Method, device and system for wireless gait recognition

Abstract

To monitor rhythmic movement such as a gait.SOLUTION: Methods, devices and systems for wireless gait recognition are described. A described system includes a transmitter, a receiver, and a processor. The transmitter is configured for transmitting a first wireless signal towards an object in a venue through a wireless multipath channel of the venue. The receiver is configured for: receiving a second wireless signal through the wireless multipath channel between the transmitter and the receiver. The second wireless signal differs from the first wireless signal due to the wireless multipath channel which is impacted by a rhythmic motion of the object. The processor is configured for: obtaining a time series of channel information (CI) of the wireless multipath channel on the basis of the second wireless signal, monitoring the rhythmic motion of the object on the basis of the time series of CI (TSCI), and triggering a response action on the basis of a result of the monitoring.SELECTED DRAWING: None

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS The entire disclosure and priority claims for each of the following cases are incorporated by reference into this application: (a) U.S. Provisional Patent Application No. 62 / 806,688, filed February 15, 2019, entitled "Method, Apparatus, and System for Wireless Gait Recognition"; (b) U.S. Provisional Patent Application No. 62 / 806,694, filed February 15, 2019, entitled "Method, Apparatus, and System for Outdoor Target Tracking"; (c) U.S. Provisional Patent Application No. 62 / 846,688, filed May 12, 2019, entitled "Method, Apparatus, and System for Processing and Providing a Lifelog Based on Wireless Signals"; (d) U.S. Provisional Patent Application No. 62 / 873,781, filed July 12, 2019, entitled "Method, Apparatus, and System for Improving the Topology of Wireless Sensing Systems"; (e) U.S. Provisional Patent Application No. 62 / 900565, filed September 15, 2019, entitled "Certified Wireless Sensing System"; (f) U.S. Provisional Patent Application No. 62 / 902,357, filed September 18, 2019, entitled "Method, Apparatus, and System for Automatic and Optimized Device-to-Cloud Connectivity for Wireless Sensing"; (g) U.S. Provisional Patent Application No. 62 / 950093, filed December 18, 2019, entitled "Method, Apparatus, and System for Target Positioning"; (h) U.S. Patent Application No. 16 / 790,610, filed February 13, 2020, entitled "Method, Apparatus, and System for Wireless Gait Recognition"; (i) U.S. patent application Ser. No. 16 / 790,627, filed February 13, 2020, entitled "Method, Apparatus, and System for Outdoor Target Tracking."

[0002] The present disclosure relates generally to gait recognition. More specifically, the present disclosure relates to identifying objects or individuals in indoor environments by recognizing their rhythmic movements, such as gait, based on wireless channel information, and to locating and tracking outdoor targets based on wireless channel information obtained from a massive multiple-input multiple-output (MIMO) system. [Background technology]

[0003] Gait, the way a person walks, is increasingly recognized not only as an essential vital sign but also as a useful biometric marker. Gait, particularly walking speed, is considered a valid and sensitive measure suitable for monitoring and assessing functional decline and overall health status, leading to its designation as the sixth vital sign. Gait reflects both functional and physiological changes and has been shown to be an indicator and predictor of many health conditions, including mobility impairment, response to rehabilitation, falls, and cognitive decline. Gait progression is associated with clinically meaningful changes in quality of life and health status. Therefore, continuous monitoring of gait at home, rather than through in-clinic clinical testing, is of great interest for personal healthcare.

[0004] Gait, on the other hand, provides a unique biometric signature of an individual and underlies a promising method for human identification. As a complex functional activity, many factors influence one's gait, making it a unique behavioral characteristic. Research has shown that gait recognition is even more reliable than facial recognition because there are dozens of distinguishing characteristics entangled in gait, making it extremely difficult to disguise someone else's walking pattern, if possible. Compared to other human recognition systems, gait recognition is particularly attractive because it can operate remotely, passively, and non-intrusively, without the active cooperation of the individual. Gait refers to the way or style of walking (e.g., of a two-legged person or an animal with two, four, or multiple legs). Gait is a simple yet subtle choreography that coordinates many muscles on a complex bone and joint structure to produce biomechanical movement. The human gait cycle includes two phases: a stance phase and a sway phase, and seven additional stages. The stance phase begins with the initial heel contact of one foot and ends when the toe of that foot leaves the ground. The swing phase immediately follows the forward swing of the leg and continues until the next heel contact.

[0005] In bipedal animals (e.g., humans or birds), the two legs can move in an alternating motion. When the left leg touches the ground, the right leg swings forward, and vice versa. In quadrupedal animals (e.g., horses, cats, dogs), in a single gait, the front two legs (and / or the back two legs) may have an alternating motion (e.g., when a horse trots). In another gait, the front two legs (and / or the back two legs) can touch the ground simultaneously or swing forward simultaneously (e.g., when a horse gallops). In multi-legged organisms (e.g., caterpillars), gait patterns can be even more complex. For example, successive pairs of legs can move in coordination (e.g., in a wave front). The two legs of each pair can have varying degrees of phase difference / delay. Gait can serve as a vital sign and a biometric cue. Gait has been shown to reflect health and functional status and can indicate and predict several health conditions, including mobility impairment, response to rehabilitation, falls, and cognitive decline. Gait progression is associated with clinically meaningful changes in quality of life and health status. Therefore, continuous monitoring of gait at home, rather than occasional in-clinic clinical testing, is of great interest for the health and well-being of individuals, especially caregivers (e.g., spouses, children, relatives, friends, and long-term care providers). Gait velocity, also known as walking speed, is the most important information measured and relevant for medical care. It has been promoted as a practical and essential clinical indicator of well-being.

[0006] There are several existing non-wireless gait measurement and recognition systems based on cameras, floor sensors, and / or wearable sensors (e.g., accelerometers) to capture gait-related information. However, existing non-wireless systems have several drawbacks. First, these systems typically require the target subject (e.g., a person) to cooperate, for example, walking in a specific direction, along a specified path, or in a specific manner. Second, existing non-wireless systems, such as instrumented walkways with floor sensors installed in or under floor mats or areas covered by installed cameras, can only monitor a specific, restricted area. The restricted area must be very small and within a short distance within line of sight (LOS). Therefore, such systems are not suitable for unlimited areas in everyday use (e.g., a person's home, a mall, a train station, or an elderly care facility). They are not convenient and / or comfortable enough for ubiquitous applications in smart homes and smart buildings. Third, when using existing systems to measure an area for gait monitoring, the devices may be too expensive and the installation may be too labor-intensive. Fourth, camera systems pose privacy issues for users. Fifth, floor sensors require significant installation effort and hardware costs. Sixth, wearables are useless if the person forgets or avoids wearing the item to be monitored.

[0007] There are several existing wireless gait monitoring systems based on the Doppler effect and radar. The drawbacks of existing wireless / RF-based systems include: (a) systems that require specialized hardware that is expensive and difficult to maintain; (b) systems that require very wide bandwidth; (c) some that require specialized phased antennas; (d) Doppler-based systems reflect only partial velocities projected in a specific direction, rather than total velocity; (e) Doppler-based systems can only operate in a narrow LOS area (typically within 4-5 meters); (f) some systems measure features that are only remotely related to gait; and (g) features can be position / location dependent, i.e., they may work in one location but not in another, thus requiring retraining for every different location.

[0008] Target localization and tracking, both indoors and outdoors, have been of interest to researchers for decades due to their crucial role in modern navigation and search and rescue systems. In general, accurate indoor localization systems can significantly improve people's lives, such as navigating passengers to airport gates or helping customers find their favorite items in large malls. Furthermore, they can also guide programmed robots to move heavy objects to their desired destinations, freeing people from tedious and time-consuming tasks and significantly improving the efficiency of modern automated production lines. Meanwhile, reliable outdoor location systems, such as the most well-known Global Positioning System (GPS), are widely used in civilian, military, and commercial applications worldwide. However, GPS requires an unobstructed line-of-sight (LOS) path to at least four GPS satellites to calculate the corresponding target's location. GPS resources available for civilian and commercial services are severely limited. Therefore, despite being centimeter-accurate, they only provide 10-meter accuracy in daily activities.

[0009] Generally, outdoor positioning methods use direction of arrival (DOA) to measure the azimuth during time of arrival (TOA) to calculate the range of the target relative to the receiver. Obviously, they require accurate time measurements, which are highly sensitive to distortion and noise in practice. Furthermore, outdoor positioning performance based on DOA and AOA is also strongly limited by its angular resolution, which is related to the aperture, dimensions, and elements of the installed antenna. Existing methods calculate source location directly from the data. To obtain more accurate results, they often require data association or center fusion processes, which are usually NP-hard problems. Finding optimal solutions to such types of questions is computationally prohibitive or requires specialized devices.

[0010] Recently, fifth-generation (5G) technology, known as massive MIMO (Multiple Input Multiple Output), was introduced, primarily focusing on communication-related issues such as spectral efficiency, resource allocation, communication complexity, inter-user interference, and channel capacity and estimation. However, related research on how to use massive MIMO to develop efficient methods for outdoor target positioning and tracking is still open. Navigation systems, with GPS being the most popular, have been widely used in modern applications. However, GPS cannot function well in non-line-of-sight (NLOS) situations because it requires unobstructed line-of-sight (LOS) to four or more GPS satellites. As a result, inertial navigation systems (INS) have been considered a complement to GPS because they are self-contained navigation technologies. INSs require estimation of the speed and direction of a moving object to dead-reckon its position. Consequently, methods for estimating the speed and direction of a target's movement have also been explored.

[0011] Accelerometers, gyroscopes, and magnetometers are the three most commonly used sensors in INS. Generally, INS employ some kind of data fusion method to jointly utilize information extracted from different sensors to estimate the target's movement speed and direction. They can be accurate when the target is relatively stable. However, they suffer from unavoidable mechanical resistance or magnetic interference, which causes accumulated errors far from the truth, especially over long periods of time.

[0012] Vision / image-based methods supported by camera devices are another popular method for detecting the moving speed and direction of subway vehicles. For example, a continuous image sequence of road surface texture is analyzed to obtain vehicle speed and direction estimation. To address the high frame rate requirement, two parallel on-board devices are employed to simultaneously capture images. Then, vehicle speed and moving direction are extracted through image matching and parameter calibration methods. Although these vision-based methods can achieve good accuracy after rounds of refinement, the requirement of sufficient image resolution and computing power poses obstacles for real-time applications.

[0013] Furthermore, fifth-generation (5G) networks are expected to be deployed, featuring ultra-wide bandwidth at gigahertz frequencies and large antenna arrays to provide gigahertz data rates. However, no effective positioning method has been disclosed for 5G networks. Thus, existing systems and methods for object positioning and tracking are not completely satisfactory, and complementary technologies that can enable high-precision outdoor positioning are desired. Summary of the Invention

[0014] The present disclosure generally relates to a system for monitoring rhythmic movements such as gait. In one embodiment, the present disclosure relates to a system for monitoring gait based on wireless signals and channel information of a wireless multipath channel affected by walking movements. The system can also recognize gait and identify and verify individuals accordingly. In another embodiment, the present disclosure discloses locating and tracking targets outdoors based on wireless channel information obtained from a massive multiple-input multiple-output (MIMO) system.

[0015] In one embodiment, a system for rhythmic motion monitoring is described. The system includes a transmitter, a receiver, and a processor. The transmitter is configured to transmit a first wireless signal toward an object at a location through a wireless multipath channel of the location. The receiver is configured to receive a second wireless signal through the wireless multipath channel between the transmitter and the receiver. The second wireless signal differs from the first wireless signal due to the wireless multipath channel being affected by the rhythmic motion of the object. The processor is configured to obtain a time series of channel information (CI) of the wireless multipath channel based on the second wireless signal, monitor the rhythmic motion of the object based on the time series of CI (TSCI), and trigger a response action based on the results of the monitoring. According to various embodiments, the processor may be physically coupled to at least one of the transmitter and the receiver.

[0016] In another embodiment, a described apparatus for rhythmic motion monitoring is located at a location where a transmitter and a receiver are disposed. The described apparatus includes a processor and at least one of the transmitter and the receiver. The transmitter is configured to transmit a first wireless signal over a wireless multipath channel of the location. The receiver is configured to receive a second wireless signal over the wireless multipath channel. The second wireless signal differs from the first wireless signal due to the wireless multipath channel being affected by rhythmic motion of an object at the location. The processor is configured to obtain a time series of channel information (CI) of the wireless multipath channel based on the second wireless signal, monitor rhythmic motion of the object based on the time series of CI (TSCI), and trigger a response action based on the results of the monitoring.

[0017] In one embodiment, the device includes a receiver but does not include a transmitter. The receiver receives the second wireless signal and extracts CI, e.g., channel state information (CSI), to perform rhythmic motion monitoring. In another embodiment, the device includes a transmitter but does not include a receiver. The CSI is extracted by the receiver and obtained by a processor for rhythmic motion monitoring. In yet another embodiment, the device includes a transmitter but does not include a receiver. The CSI is extracted in the receiver, which transmits the CSI to the transmitter. The rhythmic motion monitoring is performed in the transmitter.

[0018] In another embodiment, a method implemented by a processor, a memory communicatively coupled to the processor, and a set of instructions stored in the memory to be executed by the processor are described. The method includes obtaining a time series of channel information (CI) of a wireless multipath channel of a location. A transmitter transmits a first wireless signal through the wireless multipath channel of the location toward an object at the location. A receiver receives a second wireless signal through the wireless multipath channel and calculates a time series of CI (TSCI) of the wireless multipath channel based on the second wireless signal. The second wireless signal differs from the first wireless signal due to the wireless multipath channel being affected by rhythmic motion of the object at the location. The method further includes monitoring rhythmic motion of the object based on the TSCI and triggering a response action based on a result of the monitoring.

[0019] In one embodiment, a tracking system is described. The tracking system includes a transmitter, a receiver, and a processor. The transmitter is configured to transmit a first wireless signal through a wireless multipath channel. The receiver is configured to receive a second wireless signal through the wireless multipath channel between the transmitter and the receiver. One of the transmitter and the receiver is a device at a known location. The other of the transmitter and the receiver is a mobile device. The second wireless signal differs from the first wireless signal due to the wireless multipath channel affected by movement of the mobile device. At least one of the transmitter and the receiver includes multiple antennas, the number of which is greater than a threshold. The processor is configured to obtain multiple time series of channel information (CI) of the wireless multipath channel based on the second wireless signal, calculate spatial-temporal information (STI) of the mobile device based on at least one of the multiple time series of CI (TSCI) and past STI, and track the mobile device based on the STI. According to various embodiments, the processor may be physically coupled to at least one of the transmitter and the receiver.

[0020] In one embodiment, an additional transmitter can be placed at a reset location (e.g., at the entrance or above the door frame) that repeatedly broadcasts an additional radio signal to a specific area (e.g., the door opening). The additional radio signal acts like a beacon from a lighthouse: any receiver that picks up the beacon signal knows where it is.

[0021] The location is adaptively determined based on at least one of the mobile device's height and height probability near the target area. An additional transmitter may be located on top of a door frame that directs the beacon signal downward. The coverage area is essentially a beam or cone with a relatively large uncertainty in the receiver's location. If the receiver's (mobile device's) height is known, the mobile device's location can be narrowed down. That is, the coverage area can be narrowed based on height.

[0022] In one embodiment, the additional transmitter is stationary. In another embodiment, the additional transmitter (and / or transmitters) are not stationary and do not use directional antennas. Instead, the additional transmitter (and / or transmitters) can move around and use omni-directional antennas. The additional transmitter (and / or transmitters) have a way to obtain their instantaneous location (e.g., based on GPS, Bluetooth). When the receiver receives a beacon signal from the additional transmitter (and / or transmitters), it can obtain a "reset" location based on the instantaneous location of the additional transmitter (and / or transmitters). There may be a server (location database) that tracks the instantaneous location of the additional transmitter (and / or transmitters) and shares that location with mobile devices.

[0023] In one embodiment, the object may have complex motion (e.g., both leg motion and hand motion), and a wireless signal may capture the person's leg motion and an additional wireless signal may capture the person's hand motion.

[0024] In one embodiment, there is only one wearable receiver receiving sounding signals from two different transmitters. Due to different multipaths, the two received radio (sounding) signals may be dominated by different movements of the person. If one transmitter is located low, the radio signal may primarily capture foot / leg movements. Perhaps the second transmitter is located high, so that the second radio signal primarily captures hand movements.

[0025] In another embodiment, an apparatus for object tracking is disclosed. The described apparatus includes a processor and at least one of a transmitter and a receiver. The transmitter is configured to transmit a first wireless signal over a wireless multipath channel. The receiver is configured to receive a second wireless signal over the wireless multipath channel. One of the transmitter and the receiver is a stationary device. The other of the transmitter and the receiver is a mobile device that moves with the object. The second wireless signal is different from the first wireless signal due to the wireless multipath channel being affected by the object's motion. At least one of the transmitter and the receiver includes at least 16 antennas. The processor is configured to: acquire a plurality of time series of channel information (CI) of the wireless multipath channel based on the second wireless signal; calculate an intermediate quantity (IQ) of the current movement of the mobile device based on a set of similarity scores associated with many pairs of CIs of the plurality of time series of CIs (TSCI), each pair comprising two temporally adjacent CIs of the plurality of TSCIs; and calculate spatio-temporal information (STI) of the current movement of the mobile device based on at least one of the following: the IQ, the plurality of TSCIs, a time quantity associated with the current movement, past IQs, past STIs, tracking based on the STI, and at least one of the mobile device and the object.

[0026] In one embodiment, the device includes a receiver but does not include a transmitter. The receiver receives the second wireless signal and extracts CI, such as channel state information (CSI), for performing object tracking. In another embodiment, the device includes a transmitter but does not include a receiver. The CSI is extracted by the receiver and obtained by a processor for object tracking. In yet another embodiment, the device includes a transmitter but does not include a receiver. The CSI is extracted in a receiver that transmits the CSI to the transmitter. The object tracking is performed in the transmitter.

[0027] In another embodiment, a method is described. The method includes obtaining a plurality of time-series channel information (CIs) of a wireless multipath channel using a processor, a memory communicatively coupled to the processor, and a set of instructions stored in the memory to be executed by the processor. The plurality of time-series CIs (TSCIs) are extracted from a wireless signal transmitted between a transmitter and a receiver over the wireless multipath channel. Each of the plurality of TSCIs is associated with a pair of transmit antennas on the transmitter and a receive antenna on the receiver. One of the transmitter and the receiver is an apparatus at a known location. The other of the transmitter and the receiver is a mobile device that moves with an object. The wireless multipath channel is affected by the object's motion. At least one of the transmitter and the receiver includes 16 or more antennas. The method further includes calculating spatial-temporal information (STI) of the mobile device based on at least one of the plurality of TSCIs and past STIs, and tracking the object based on the STI.

[0028] Other concepts relate to software for implementing the present disclosure for wireless rhythmic motion monitoring and object tracking in rich clutter environments. Additional novel features will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following drawings and accompanying drawings, or may be learned by the manufacture or operation of the embodiments. The novel features of the present disclosure may be realized and attained by practice or use of various aspects of the methods, instrumentalities, and combinations described in the detailed embodiments discussed below. [Brief explanation of the drawings]

[0029] The methods, systems and / or devices will be further described with respect to exemplary embodiments, which will be described in detail with reference to the drawings, which are non-limiting exemplary embodiments, and in which like reference numerals represent like structure throughout the several views of the drawings.

[0030] [Figure 1A] FIG. 1A shows an exemplary multipath model in a rich, scattering indoor environment, where an object (eg, a human body) is simplified so that one reflection pair produces only one dominant reflection path.

[0031] [Figure 1B] FIG. 1B illustrates another exemplary multipath model in a rich, cluttered indoor environment, where objects scatter signals, resulting in multiple paths, according to an embodiment of the present disclosure.

[0032] [Figure 2A] FIG. 2A illustrates an example of a synthesized velocity signal according to an embodiment of the present disclosure.

[0033] [Figure 2B] FIG. 2B shows an example of compound rate differences according to an embodiment of the present disclosure.

[0034] [Figure 2C] FIG. 2C illustrates an example matrix of a composite velocity signal according to an embodiment of the present disclosure.

[0035] [Figure 2D] FIG. 2D illustrates an example matrix of differences in the composite velocity signal according to an embodiment of the present disclosure.

[0036] [Figure 3] FIG. 3 illustrates an exemplary performance of velocity estimation according to an embodiment of the present disclosure.

[0037] [Figure 4] FIG. 4 illustrates a comparison of the performance of exemplary velocity estimation methods according to an embodiment of the present disclosure.

[0038] [Figure 5] FIG. 5 illustrates an exemplary performance of the periodic autocorrelation function (ACF) of walking speed according to an embodiment of the present disclosure.

[0039] [Figure 6] FIG. 6 illustrates an exemplary performance of gait cycle estimation according to an embodiment of the present disclosure.

[0040] [Figure 7] FIG. 7 illustrates an exemplary performance of an extracted stable gait cycle according to an embodiment of the present disclosure.

[0041] [Figure 8] FIG. 8 illustrates exemplary performance of continuous monitoring of gait parameters over time according to an embodiment of the present disclosure.

[0042] [Figure 9] FIG. 9 shows an exemplary performance of the harmonic ratio over time for a short walking trace according to an embodiment of the present disclosure.

[0043] [Figure 10] FIG. 10 illustrates an example relationship between stride, cycle time, and velocity according to an embodiment of the present disclosure.

[0044] [Figure 11A] FIG. 11A illustrates an exemplary profile of velocity deviation according to an embodiment of the present disclosure.

[0045] [Figure 11B] FIG. 11B illustrates an exemplary histogram of velocity deviation according to an embodiment of the present disclosure.

[0046] [Figure 12] FIG. 12 illustrates exemplary recurrent plots for different users according to an embodiment of the present disclosure.

[0047] [Figure 13] FIG. 13 illustrates an exemplary scaled ACF feature according to an embodiment of the present disclosure.

[0048] [Figure 14] FIG. 14 illustrates an exemplary feature correlation matrix according to an embodiment of the present disclosure.

[0049] [Figure 15] FIG. 15 illustrates an exemplary method for gait recognition according to an embodiment of the present disclosure.

[0050] [Figure 16] FIG. 16 illustrates an exemplary performance of RR versus number of users according to an embodiment of the present disclosure.

[0051] [Figure 17] FIG. 17 illustrates an exemplary setup of a base station with massive MIMO antennas according to an embodiment of the present disclosure.

[0052] [Figure 18A] FIG. 18A illustrates an exemplary signal propagation shape between ro and rs in a 3D model according to an embodiment of the present disclosure.

[0053] [Figure 18B] FIG. 18B illustrates an exemplary signal propagation shape between ro and rs in a 2D model according to an embodiment of the present disclosure.

[0054] [Figure 19A] FIG. 19A illustrates an example ACF variance near a predetermined position (x=0, y=0) for velocity and position estimation according to an embodiment of the present disclosure.

[0055] [Figure 19B] FIG. 19B illustrates an example ACF distribution along the cross-beam direction position (x=0) for velocity and position estimation according to an embodiment of the present disclosure.

[0056] [Figure 19C] FIG. 19C illustrates an example ACF variance (x=0, y>=0) and corresponding peak definition for velocity and position estimation according to an embodiment of the present disclosure.

[0057] [Figure 20] FIG. 20 illustrates an exemplary signal propagation shape as r0 moves according to an embodiment of the present disclosure.

[0058] [Figure 21] FIG. 21 illustrates an exemplary curve fitting by local regression according to an embodiment of the present disclosure.

[0059] [Figure 22] FIG. 22 illustrates an example speed and direction estimation based on a scenario with two base stations according to an embodiment of the present disclosure.

[0060] [Figure 23] FIG. 23 illustrates an exemplary angle ambiguity created by two complementary angles whose sum is a right angle according to an embodiment of the present disclosure.

[0061] [Figure 24] FIG. 24 illustrates an example velocity estimation based on a scenario with three base stations according to an embodiment of the present disclosure.

[0062] [Figure 25] FIG. 25 shows two adjacent positions rs1 and rs2 according to an embodiment of the present disclosure.

[0063] [Figure 26A] FIG. 26A shows a geometric illustration of the opposite crossing angle ambiguity for two stations according to an embodiment of the present disclosure.

[0064] [Figure 26B] FIG. 26B shows a geometric illustration of the opposite crossing angle ambiguity for three stations according to an embodiment of the present disclosure.

[0065] [Figure 27] FIG. 27 shows a geometric illustration of two adjacent locations rS1 and rS2 according to an embodiment of the present disclosure.

[0066] [Figure 28] FIG. 28 shows another geometric illustration of opposing crossing angle ambiguity according to an embodiment of the present disclosure.

[0067] [Figure 29A] FIG. 29A illustrates rate estimation error versus number of antennas according to an embodiment of the present disclosure.

[0068] [Figure 29B] FIG. 29B illustrates location estimation error versus number of antennas according to an embodiment of the present disclosure.

[0069] [Figure 29C] FIG. 29C illustrates normalized location estimation error versus number of antennas according to an embodiment of the present disclosure.

[0070] [Figure 30] FIG. 30 illustrates an exemplary method for updating orientation according to an embodiment of the present disclosure.

[0071] [Figure 31]FIG. 31 illustrates a convex hull formed by AoA measurements according to an embodiment of the present disclosure.

[0072] [Figure 32] FIG. 32 illustrates a method for target positioning using RSS-based ranging according to an embodiment of the present disclosure.

[0073] [Figure 33] FIG. 33 illustrates a flowchart of an exemplary method for wireless object tracking according to an embodiment of the present disclosure.

[0074] [Figure 34] FIG. 34 illustrates an example network topology of four devices according to an embodiment of the present disclosure.

[0075] [Figure 35] FIG. 35 illustrates an example network topology of three devices according to an embodiment of the present disclosure.

[0076] [Figure 36] FIG. 36 illustrates an example flowchart and components of a master origin device according to an embodiment of the present disclosure.

[0077] [Figure 37] FIG. 37 illustrates an example network topology of six devices according to an embodiment of the present disclosure.

[0078] [Figure 38] FIG. 38 illustrates an example network topology of nine devices in a local area network according to an embodiment of the present disclosure.

[0079] [Figure 39] FIG. 39 illustrates exemplary motion detection performance based on passive infrared (PIR) sensing and WiFi sensing, according to an embodiment of the present disclosure.

[0080] [Figure 40]FIG. 40 illustrates an exemplary configuration for respiratory monitoring according to an embodiment of the present disclosure.

[0081] [Figure 41] FIG. 41 shows an exemplary diagram of a system for object motion detection according to an embodiment of the present disclosure.

[0082] [Figure 42] FIG. 42 shows an exemplary diagram of a system for object motion detection according to an embodiment of the present disclosure.

[0083] [Figure 43] FIG. 43 illustrates an exemplary daily display showing differentiated examples of sleep according to an embodiment of the present disclosure.

[0084] [Figure 44A] FIG. 44A illustrates an exemplary weekly view shown on a 24-hour scale according to an embodiment of the present disclosure. [Figure 44B] FIG. 44B illustrates an exemplary weekly view shown on a 24-hour scale according to an embodiment of the present disclosure.

[0085] [Figure 45A] FIG. 45A illustrates an exemplary home view showing real-time respiration rate and movement index according to an embodiment of the present disclosure. [Figure 45B] FIG. 45B illustrates an exemplary home view showing real-time respiration rate and movement index according to an embodiment of the present disclosure.

[0086] [Figure 46] FIG. 46 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 47] FIG. 47 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 48] FIG. 48 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 49]FIG. 49 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 50] FIG. 50 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 51] FIG. 51 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. [Figure 52] FIG. 52 shows a more exemplary illustration of a display of a lifelog according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0087] In the following detailed description, numerous specific details are set forth, by way of example, to provide a thorough understanding of the relevant disclosure. However, it should be apparent to those skilled in the art that the present disclosure may be practiced without such details. In other instances, well-known methods, procedures, components, and / or circuits have been described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present disclosure.

[0088] In the following detailed description, numerous specific details are set forth, by way of example, to provide a thorough understanding of the relevant disclosure. However, it should be apparent to those skilled in the art that the present disclosure may be practiced without such details. In other instances, well-known methods, procedures, components, and / or circuits have been described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present disclosure.

[0089] In one embodiment, the present disclosure discloses a method, apparatus, device, system, and / or software (method / apparatus / device / system / software) for a wireless monitoring system. Time-series channel information (CI) of a wireless multipath channel may be obtained (e.g., dynamically) using a processor, a memory communicatively coupled to the processor, and a set of instructions stored in the memory. A time series of CI (TSCI) may be extracted from wireless signals (signals) transmitted between a type 1 heterogeneous wireless device (e.g., a wireless transmitter, TX) and a type 2 heterogeneous wireless device (e.g., a wireless receiver, RX) at a location over the channel. The channel may be affected by representations (e.g., motion, movement, representation, and / or changes in position / pose / shape / representation) of objects at the location. Objects and / or their motion characteristics and / or spatial-temporal information (STI, e.g., motion information) may be monitored based on the TSCI. Tasks may be performed based on the characteristics and / or STI. A presentation associated with the task may be generated in a user interface (UI) on the user's device. The TSCI may be a wireless signal stream. The TSCI or each CI may be pre-processed. The device may be a station (STA). The symbol "A / B" means "A and / or B" in this disclosure.

[0090] Expressions can include placement, placement of movable parts, location, position, orientation, identifiable location, area, spatial coordinates, presentation, state, expression, static representation, size, length, width, height, angle, scale, shape, curve, surface, area, volume, pose, posture, sign, body representation, dynamic representation, dynamic noun, movement, movement sequence, gesture, stretch, contraction, distortion, deformation, body representation (e.g., head, face, eyes, mouth, tongue, hair, voice, neck, limbs, arms, hands, legs, feet, muscles, moving parts), surface representation (e.g., shape, texture, material, color, electromagnetic (EM) properties, visual pattern, humidity, reflectivity, translucency, flexibility), material properties (e.g., living tissue, hair, fabric, metal, wood, leather, plastic, metallic, artificial material, solid, liquid, gas, temperature), movement, activity, behavior, change in expression, and / or any combination.

[0091] Wireless signals include transmit / receive signals, EM emissions, RF signals / transmissions, signals in licensed / unlicensed / ISM bands, band-limited signals, baseband signals, wireless / mobile / cellular communications signals, wireless / mobile / cellular network signals, mesh signals, optical signals / communications, downlink / uplink signals, unicast / multicast / broadcast signals, standard (e.g., WLAN, WWAN, WPAN, WBAN, international, national, industry, de facto, IEEE, IEEE 802, 802.11 / 15 / 16, WiFi, 802.11n / ac / ax / be, 3G / 4G / LTE / 5G / 6G / 7G / 8G, 3GPP, blue tooth, BLE, Zigbee, RFID, UWB, WiMax) compliant signals, protocol signals, standard frames, beacon / pilot / search / inquiry / acknowledge / handshake / synchronization signals, management / control / data signals, standardized wireless / cellular communications protocols, reference signals, source signals, motion probe / detection / The CI may include a sensing signal and / or a series of signals. The wireless signal may include line-of-sight (LOS) and / or non-LOS components (or paths / links). Each CI may be extracted / generated / calculated / detected at a layer of the Type 2 device (e.g., the PHY / MAC layer of the OSI model) and obtained by an application (e.g., software, firmware, driver, app, wireless monitoring software / system).

[0092] A wireless multipath channel can include: a communication channel, an analog frequency channel (e.g., analog carrier frequencies around 700 / 800 / 900 MHz, 1.8 / 1.8 / 2.4 / 3 / 5 / 6 / 27 / 60 GHz), a coded channel (e.g., CDMA), and / or a channel of a wireless network / system (e.g., WLAN, WiFi, mesh, LTE, 4G / 5G, Bluetooth, Zigbee, UWB, RFID, microwave). It can include two or more channels. The channels can be contiguous (e.g., adjacent / overlapping bands) or non-contiguous (e.g., non-overlapping WiFi channels, one at 2.4 GHz and one at 5 GHz).

[0093] The TSCI can be extracted from a wireless signal at a layer of a Type-2 device (e.g., a layer of the OSI reference model, a physical layer, a data link layer, a logical link control layer, a media access control (MAC) layer, a network layer, a transport layer, a session layer, a presentation layer, an application layer, a TCP / IP layer, an Internet layer, or a link layer). The TSCI may also be extracted from a derived signal (e.g., a baseband signal, a motion detection signal, or a motion sensing signal) derived from a wireless signal (e.g., an RF signal). It may be a (wireless) measurement detected by a communication protocol (e.g., a standardized protocol) using an existing mechanism (e.g., a wireless / cellular communication standard / network, 3G / LTE / 4G / 5G / 6G / 7G / 8G, WiFi, IEEE 802.11 / 15 / 16). The derived signal may include a packet having at least one of a preamble, a header, and a payload (e.g., for data / control / management in a wireless link / network). The TSCI may be extracted from a probe signal in a packet (e.g., a training sequence, STF, LTF, L-STF, L-LTF, L-SIG, HE-STF, HE-LTF, HE-SIG-A, HE-SIG-B, or CEF). The motion detection / sensing signal may be recognized / identified based on the probe signal. The packet may be a standard-compliance protocol frame, a management frame, a control frame, a data frame, a sounding frame, an excitation frame, an illumination frame, a null data frame, a beacon frame, a pilot frame, a probe frame, a request frame, a response frame, an association frame, a reassociation frame, a disassociation frame, an authentication frame, an action frame, a report frame, a poll frame, an announcement frame, an extension frame, an inquiry frame, an acknowledgement frame, an RTS frame, a CTS frame, a QoS frame, a CF-Poll frame, a CF-Ack frame, a block acknowledgement frame, a reference frame, a training frame, and / or a synchronization frame.

[0094] The packet may contain control data and / or motion detection probes. Data (e.g., Type 1 device IDs, parameters, characteristics, settings, control signals, commands, instructions, notifications, and broadcast-related information) may be obtained from the payload. The wireless signal may be transmitted by a Type 1 device and received by a Type 2 device. A database (e.g., in a local server, hub device, cloud server, or storage network) may be used to store TSCI, characteristics, STI, signatures, patterns, behaviors, trends, parameters, analysis, output responses, identification information, user information, device information, channel information, location (e.g., map, environment model, network, proximity device / network) information, task information, class / category information, presentation (e.g., UI) information, and / or other information.

[0095] A Type 1 / Type 2 device may include at least one of electronics, circuitry, transmitter (TX) / receiver (RX) / transceiver, RF interface, "origin satellite" / "tracker bot," unicast / multicast / broadcast device, wireless power device, power / destination device, wireless node, hub device, target device, motion detection device, sensor device, remote / wireless sensor device, wireless communication device, wireless enabled device, standard compliant device, and / or receiver. A Type 1 (or Type 2) device may be heterogeneous because, if multiple instances of a Type 1 (or Type 2) device exist, they may have different circuitry, enclosure, structure, purpose, auxiliary functionality, chips / ICs, processor, memory, software, firmware, network connectivity, antenna, brand, model, appearance, form, shape, color, material, and / or specifications. A Type 1 / Type 2 device may include an access point, a router, a mesh router, an Internet of Things (IoT) device, a wireless terminal, one or more radio / RF subsystems / radio interfaces (e.g., 2.4 GHz radio, 5 GHz radio, fronthaul radio, backhaul radio), a modem, an RF front end, an RF / radio chip or integrated circuit (IC).

[0096] At least one of Type 1 devices, Type 2 devices, links between them, objects, characteristics, STI, motion monitoring, and tasks may be associated with an identification (ID) such as a UUID. Type 1, Type 2, or another device may acquire, store, retrieve, access, preprocess, condition, process, analyze, monitor, or apply TSCI. Type 1 and Type 2 devices may communicate network traffic on other channels (e.g., Ethernet, HDMI, USB, Bluetooth, BLE, WiFi, LTE, other networks, wireless multipath channels) in parallel with wireless signals. Type 2 devices may passively observe, monitor, or receive wireless signals from Type 1 devices on wireless multipath channels without establishing a connection (e.g., association / authentication) with or requesting service from the Type 1 device.

[0097] A transmitter (i.e., Type 1 device) can function (act as) a receiver (i.e., Type 2 device) temporarily, sporadically, continuously, repeatedly, interchangeably, alternately, simultaneously, in parallel, and / or simultaneously, and vice versa. A device can function as a Type 1 device (transmitter) and / or a Type 2 device (receiver) temporarily, sporadically, continuously, repeatedly, simultaneously, in parallel, and / or simultaneously. There may be multiple wireless nodes, each of which is a Type 1 (TX) and / or a Type 2 (RX) device. TSCI may be obtained for each two nodes when exchanging / communicating wireless signals. An object's characteristics and / or STI may be monitored individually based on the TSCI or jointly based on two or more (e.g., all) TSCIs. An object's movement may be monitored actively (in that its Type 1 device, Type 2 device, or both are wearable / associated with the object) and / or passively (in that both the Type 1 device and the Type 2 device are not wearable / associated with the object). It can be passive because the object may not be associated with a Type 1 device and / or a Type 2 device. The object (e.g., a user, an automated guided vehicle, or an AGV) may not need to carry / place any wearable / fixture (i.e., Type 1 and Type 2 devices are not wearable / attached equipment that the object needs to carry to perform a task). It can be active because the object may be associated with either a Type 1 or Type 2 device. The object can carry (or place) a wearable / attachment (e.g., a Type 1 device, a Type 2 device, or equipment communicatively coupled to either a Type 1 or Type 2 device).

[0098] The presentation may be visual, audio, image, video, animation, graphic presentation, text, etc. The computation of the task may be performed by a processor (or logic unit) of the Type 1 device, a processor (or logic unit) of the IC of the Type 1 device, a processor (or logic unit) of the Type 2 device, a processor of the IC of the Type 2 device, a local server, a cloud server, a data analysis subsystem, a signal analysis subsystem, and / or another processor. This work may be performed with or without a radio fingerprint or baseline (e.g., collection, processing, processing, transmission, and / or training phase / previous survey / latest survey / initial radio survey, passive indications), training, profile, trained profile, static profile, static profile, survey, initial radio survey, initial setup, installation, retraining, update, and reset).

[0099] A Type 1 device (TX device) may include at least one heterogeneous radio transmitter. A Type 2 device (RX device) may include at least one heterogeneous radio receiver. A Type 1 device and a Type 2 device may be co-located. A Type 1 device and a Type 2 device may be the same device. Any device may have a data processing unit / apparatus, a computing unit / system, a network unit / system, a processor (e.g., a logic unit), a memory communicatively coupled to the processor, and a set of instructions stored in the memory to be executed by the processor. Some processors, memories, and sets of instructions may cooperate. There may be multiple Type 1 devices interacting (e.g., communicating, exchanging signals / control / notifications / other data) with the same Type 2 device (or multiple Type 2 devices) and / or there may be multiple Type 2 devices interacting with the same Type 1 device. Multiple Type 1 / Type 2 devices may be synchronous and / or asynchronous, with the same / different window widths / sizes and / or time shifts, with the same / different synchronization start times, synchronization end times, etc. The wireless signals transmitted by multiple Type 1 devices may be sporadic, intermittent, continuous, repetitive, synchronous, simultaneous, concurrent, and / or simultaneous. Multiple Type 1 / Type 2 devices may operate independently and / or cooperatively. Type 1 and / or Type 2 devices may have / include heterogeneous hardware circuits (e.g., heterogeneous chips or ICs capable of generating / receiving wireless signals, extracting CI from received signals, or making CI available). They may be communicatively coupled to the same or different servers (e.g., cloud servers, edge servers, local servers, hub devices).

[0100] The operation of one device can be based on the operation, state, internal state, storage, processor, memory output, physical location, computational resources, or network of another device. Different devices may communicate directly and / or through another device / server / hub device / cloud server. A device can be associated with one or more users and have associated settings. Settings may be selected once, pre-programmed, and / or changed (e.g., adjusted, changed, modified) / varied over time. There may be additional steps in a method. Method steps and / or additional steps may be performed in the order shown or in a different order. Any steps may be performed in parallel, iteratively, or otherwise iteratively or otherwise. A user may be a human, adult, elderly adult, male, female, infant, child, baby, pet, animal, living being, machine, computer module / software, etc.

[0101] For one or more Type 1 devices interacting with one or more Type 2 devices, any processing (e.g., time domain, frequency domain) can be different for different devices. Processing can be based on location, orientation, direction, role, user-related characteristics, settings, configuration, available resources, available bandwidth, network connection, hardware, software, processor, co-processor, memory, battery life, available power, antenna, antenna type, antenna directional / unidirectional characteristics, power settings, and / or other parameters / characteristics of the device.

[0102] The wireless receiver (e.g., a Type 2 device) may receive a signal and / or another signal from the wireless transmitter (e.g., a Type 1 device). The wireless receiver may receive another signal from another wireless transmitter (e.g., a second Type 1 device). The wireless transmitter may transmit a signal and / or another signal to another wireless receiver (e.g., a second Type 2 device). The wireless transmitter, the wireless receiver, another wireless receiver, and / or another wireless transmitter may move with the object and / or another object. The other object may be tracked.

[0103] A Type 1 and / or Type 2 device may be capable of wirelessly coupling with at least two Type 2 and / or Type 1 devices. The Type 1 device may be triggered / controlled to switch / establish a wireless coupling (e.g., association, authentication) from the Type 2 device to a second Type 2 device at another location in the location. Similarly, the Type 2 device may be triggered / controlled to switch / establish a wireless coupling from the Type 1 device to a second Type 1 device at yet another location in the location. The switching may be controlled by a server (or hub device), a processor, the Type 1 device, the Type 2 device, and / or another device. The radios used before and after the switching may be different. A second wireless signal (second signal) may be transmitted through the channel between the Type 1 device and the second Type 2 device (or between the Type 2 device and the second Type 1 device). A second TSCI of the channel may be obtained from the second signal. The second signal may be the first signal. A characteristic, STI, and / or another quantity of the object may be monitored based on the second TSCI. The Type 1 and Type 2 devices may be the same. The characteristics, STIs, and / or other quantities with different timestamps may form a waveform. The waveform may be displayed in a presentation.

[0104] The wireless signal and / or another signal may have data embedded therein. The wireless signal may be a series of probe signals (e.g., repeated transmission of a probe signal, reuse of one or more probe signals). The probe signal may vary / change over time. The probe signal may be a standard-compliant signal, a protocol signal, a standardized wireless protocol signal, a control signal, a data signal, a wireless communication network signal, a cellular network signal, a WiFi signal, an LTE / 5G / 6G / 7G signal, a reference signal, a beacon signal, a motion detection signal, and / or a motion sensing signal. The probe signal may be formatted according to a wireless network standard (e.g., WiFi), a cellular network standard (e.g., LTE / 5G / 6G), or another standard. The probe signal may include a packet having a header and a payload. The probe signal may have data embedded therein. The payload may include data. The probe signal may replace a data signal. The probe signal may be embedded in a data signal. The wireless receiver, the wireless transmitter, the other wireless receiver, and / or the other wireless transmitter may be associated with at least one processor, a memory communicatively coupled to the respective processor, and / or a respective set of instructions stored in the memory that, when executed, cause the processor to perform any and / or all steps necessary to determine the object's STI (e.g., motion information), initial STI, initial time, direction, instantaneous position, instantaneous angle, and / or velocity. The processor, memory, and / or set of instructions may be associated with the Type 1 device, at least one Type 2 device, the object, a device associated with the object, another device associated with the location, a cloud server, a hub device, and / or another server.

[0105] A Type 1 device can broadcast a signal to at least one Type 2 device(s) over a location channel. The signal is transmitted without the Type 1 device establishing a wireless connection (e.g., association, authentication) with any Type 2 device and without the Type 2 device requesting service from the Type 1 device. A Type 1 device can transmit to a specific media access control (MAC) address common to multiple Type 2 devices. Each Type 2 device can tune its MAC address to a specific MAC address. The specific MAC address can be associated with a location. The association can be recorded in an association table in an association server (e.g., a hub device). A location can be identified by a Type 1 device, a Type 2 device, and / or another device based on the specific MAC address, a series of probe signals, and / or at least one TSCI extracted from the probe signals. For example, a Type 2 device can be moved to a new location (e.g., from another location). A Type 1 device can be reconfigured at a location such that Type 1 and Type 2 devices are unaware of each other. During setup, the Type 1 device may be instructed / guided / caused / controlled (e.g., using a dummy receiver, using hardware pin configuration / connection, using saved configuration, using local configuration, using remote configuration, using downloaded configuration, using a hub device, or using a server) to send a series of probe signals to specific MAC addresses. Upon powering up, the Type 2 device may scan for probe signals according to a table (e.g., stored in a designated source, server, hub device, cloud server) of MAC addresses that can be used to broadcast in different locations (e.g., different MAC addresses used for different locations such as a house, office, enclosure, floor, multi-story building, store, airport, mall, stadium, hall, station, subway, lot, region, area, district, province, city, country, continent, etc.).When a Type 2 device detects a probe signal sent to a specific MAC address, the Type 2 device can use a table to identify a location based on the MAC address. The location of the Type 2 device can be calculated based on the specific MAC address, the sequence of probe signals, and / or at least one TSCI obtained by the Type 2 device from the probe signals. The calculation can be performed by the Type 2 device. The specific MAC address can be changed (e.g., adjusted, modified, or revised) over time. It can be changed according to a timetable, rule, policy, mode, condition, situation, and / or change. The specific MAC address can be selected based on MAC address availability, a preselected list, collision patterns, traffic patterns, data traffic between the Type 1 device and other devices, available bandwidth, random selection, and / or a MAC address switching plan. The specific MAC address can be the MAC address of a second wireless device (e.g., a dummy receiver or a receiver acting as a dummy receiver).

[0106] The Type 1 device may transmit a probe signal on a channel selected from a set of channels. At least one CI of the selected channel may be acquired by each Type 2 device from the probe signal transmitted on the selected channel. The selected channel may be changed (e.g., adjusted, changed, modified) over time. The change may be according to a timetable, rule, policy, mode, condition, situation, and / or change. The selected channel may be selected based on channel availability, random selection, a preselected list, co-channel interference, inter-channel interference, channel traffic patterns, data traffic between the Type 1 device and another device, effective bandwidth associated with the channel, security criteria, channel switching plan, criteria, quality criteria, signal quality conditions, and / or considerations.

[0107] The specific MAC address and / or selected channel information can be communicated between a Type 1 device and a server (e.g., a hub device) over a network. The specific MAC address and / or selected channel information can also be communicated between a Type 2 device and a server (e.g., a hub device) over another network. A Type 2 device can communicate the specific MAC address and / or selected channel information to another Type 2 device (e.g., via a mesh network, Bluetooth, WiFi, NFC, ZigBee, etc.). The specific MAC address and / or selected channel can be selected by a server (e.g., a hub device). The specific MAC address and / or selected channel can be signaled in an announcement channel by a Type 1 device, a Type 2 device, and / or a server (e.g., a hub device). Any information can be preprocessed before being communicated.

[0108] A wireless connection (e.g., association, authentication) between a Type 1 device and another wireless device can be established (e.g., using a signal handshake). The Type 1 device can send a first handshake signal (e.g., a sounding frame, a probe signal, a request to send RTS) to the other device. The other device can respond by sending a second handshake signal (e.g., a command or a clear to send CTS) to the Type 1 device, triggering the Type 1 device to send a signal (e.g., a series of probe signals) in a broadcast manner to multiple Type 2 devices without establishing a connection with any Type 2 devices. The second handshake signal can be a response or acknowledgment (e.g., an ACK) to the first handshake signal. The second handshake signal can include data having location and / or Type 1 device information. The other device can be a dummy device with a purpose (e.g., primary purpose, secondary purpose) to establish a wireless connection with the Type 1 device, receive the first signal, and / or send the second signal. The other device can be physically attached to the Type 1 device.

[0109] In another example, another device can send a third handshake signal to a Type-1 device that triggers the Type-1 device to broadcast a signal (e.g., a series of probe signals) to multiple Type-2 devices without establishing a connection (e.g., association, authentication) with any of the Type-2 devices. The Type-1 device can respond to the third special signal by sending a fourth handshake signal to the other device. Another device can be used to trigger multiple Type-1 devices to broadcast. The triggering may be sequential, partially sequential, partially parallel, or fully parallel. The other device may have multiple radio circuits to trigger multiple transmitters in parallel. Parallel triggering can also be achieved by using at least one additional device to perform a trigger in parallel with another device (as the other device does). The other device cannot communicate (or suspend communication) with the Type-1 device after establishing a connection with it. The suspended communication may resume. The other device can enter an inactive mode, dormant mode, sleep mode, standby mode, low power mode, OFF mode, and / or power-down mode after establishing a connection with the Type-1 device. The other device may have a specific MAC address such that the Type 1 device sends signals to the specific MAC address. The Type 1 device and / or the other device may be controlled and / or coordinated by a first processor associated with the Type 1 device, a second processor associated with the other device, a third processor associated with the specified source, and / or a fourth processor associated with the other device. The first and second processors may coordinate with each other.

[0110] A first series of probe signals may be transmitted to at least one first Type 2 device by a first antenna of the Type 1 device through a first channel at a first location. A second series of probe signals may be transmitted to at least one second Type 2 device by a second antenna of the Type 1 device through a second channel at a second location. The first series of probe signals and the second series of probe signals may be different or not different. The at least one first Type 2 device may be different or not different from the at least one second Type 2 device. The first and / or second series of probe signals may be broadcast without an established connection (e.g., association, authentication) between the Type 1 device and any Type 2 device. The first and second antennas may be the same or different. The two locations may have different sizes, shapes, and multipath characteristics. The first and second locations may overlap. The immediate areas in the vicinity of the first and second antennas may overlap. The first and second channels may be the same or different. For example, the first may be WiFi and the second may be LTE, or both may be WiFi, but the first may be 2.4GHz WiFi and the second may be 5GHz WiFi, or both may be 2.4GHz WiFi, but with different channel numbers, SSID names, and / or WiFi settings.

[0111] Each Type 2 device can obtain at least one TSCI from each series of probe signals, where the CI is each channel between the Type 2 device and the Type 1 device. Some first Type 2 device(s) and some second Type 2 device(s) can be the same. The first and second series of probe signals can be synchronous / asynchronous. The probe signals can be transmitted with data or replaced with data signals. The first and second antennas can be the same. The first series of probe signals can be transmitted at a first rate (e.g., 30 Hz). The second series of probe signals can be transmitted at a second rate (e.g., 200 Hz). The first and second rates can be the same / different. The first and / or second rates can be changed (e.g., adjusted, varied, modified) over time. The change can be according to a schedule, rule, policy, mode, condition, situation, and / or change. Any rate can be changed (e.g., adjusted, varied, modified) over time. The first and / or second series of probe signals may be transmitted to a first MAC address and / or a second MAC address, respectively. The two MAC addresses may be the same or different. The first series of probe signals may be transmitted in a first channel. The second series of probe signals may be transmitted in a second channel. The two channels may be the same or different. The first or second MAC address and the first or second channel may change over time. Any change may be according to a timetable, rule, policy, mode, state, condition, and / or change.

[0112] A Type 1 device and another device may be controlled and / or coordinated, physically attached to a common device, or of / within a common device. They may be controlled by / connected to a common data processor or connected to a common bus interconnect / network / LAN / Bluetooth network / NFC network / BLE / wired network / wireless network / mesh network / mobile network / cloud. They may share common memory or be associated with a common user, user device, profile, account, identity (ID), identifier, home, residence, physical address, location, geographic coordinates, IP subnet, SSID, home device, office device, and / or manufacturing device. Each Type 1 device may be a signal source for a respective set of Type 2 devices (i.e., it sends a respective signal (e.g., a respective series of probe signals) to a respective set of Type 2 devices). Each Type 2 device selects a Type 1 device from among all Type 1 devices as its signal source. Each Type 2 device may select a Type 1 device asynchronously. At least one TSCI may be obtained by each Type 2 device from each series of probe signals from the Type 1 device, and the CI is a channel between the Type 2 device and the Type 1 device. Each Type 2 device selects a Type 1 device as its signal source from all Type 1 devices based on the identity (ID) or Type 1 / Type 2 device identifier, the task to be performed, past signal sources, history (e.g., of past signal sources, the Type 1 device, another Type 1 device, each Type 2 receiver, and / or another Type 2 receiver), a threshold for switching signal sources, and / or user information, account, access information, parameters, characteristics, and / or signal strength (e.g., associated with the Type 1 device and / or each Type 2 receiver). Initially, the Type 1 device may be the signal source of the initial set of each Type 2 device (i.e., the Type 1 device sends each signal (series of probe signals) to each initial set of Type 2 devices).Each initial respective Type 2 device selects a Type 1 device from among all Type 1 devices as its signal source.

[0113] A particular Type 2 device's signal source (Type 1 device) may be changed (e.g., adjusted, altered, modified) if: (1) the time interval between two adjacent probe signals (e.g., between the current probe signal and the immediately previous probe signal, or between the next probe signal and the current probe signal) received from the Type 2 device's current signal source exceeds a first threshold; (2) the signal strength associated with the Type 2 device's current signal source is less than a second threshold; (3) the processed signal strength associated with the Type 2 device's current signal source is less than a third threshold, where the signal strength is processed with a low-pass filter, a band-pass filter, a median filter, a moving average filter, a weighted average filter, a linear filter, and / or a nonlinear filter; and / or (4) the signal strength (or processed signal strength) associated with the Type 2 device's current signal source is below a fourth threshold for a significant percentage (e.g., 70%, 80%, 90%) of a recent time window. The percentage can exceed a fifth threshold. The first, second, third, fourth and / or fifth thresholds may be time-varying.

[0114] Condition (1) can occur when a Type 1 device and a Type 2 device gradually move farther away from each other, resulting in some probe signals from the Type 1 device becoming too weak to be received by the Type 2 device. Conditions (2) through (4) can occur when the two devices move far enough away from each other that the signal strength becomes very weak.

[0115] The signal source of a Type 2 device may remain unchanged if another Type 1 device has a signal strength weaker than the current signal source by a factor (e.g., 1, 1.1, 1.2, or 1.5). If the signal source is changed (adjusted, modified, modified, etc.), the new signal source may become effective in the near future (e.g., each next time). The new signal source may be the Type 1 device with the strongest signal strength and / or processed signal strength. The current signal source and the new signal source may be the same or different.

[0116] A list of available Type 1 devices may be initialized and maintained by each Type 2 device. The list may be updated by examining signal strengths and / or processed signal strengths associated with each set of Type 1 devices. A Type 2 device may select between a first series of probe signals from a first Type 1 device and a second series of probe signals from a second Type 1 device based on their respective probe signal rates, MAC addresses, channels, characteristics / properties / statuses, tasks to be performed by the Type 2 device, the first and second series of signal strengths, and / or other considerations.

[0117] The series of probe signals may be transmitted at a constant rate (e.g., 100 Hz). The series of probe signals may be scheduled at regular intervals (e.g., 0.01 seconds for 100 Hz), although each probe signal may experience small time perturbations, perhaps due to timing requirements, timing control, network control, handshaking, message passing, collision avoidance, carrier sensing, congestion, resource availability, and / or other considerations. The rate may be changed (e.g., adjusted, altered, modified). The change may be according to a schedule (e.g., hourly), a rule, a policy, a mode, a condition, and / or a change (e.g., whenever an event occurs). For example, the rate may be normally 100 Hz but changed to 1000 Hz in demanding situations and to 1 Hz in low-power / standby situations. The probe signals may be transmitted in bursts.

[0118] The probe signal rate may vary based on the task performed by the Type 1 or Type 2 device (e.g., a task may normally require 100 Hz, temporarily 1000 Hz for 20 seconds). In one example, transmitters (Type 1 devices), receivers (Type 2 devices), and associated tasks may be adaptively (and / or dynamically) associated with classes (e.g., classes that are low priority, high priority, emergency, critical, normal, privileged, non-subscribed, subscribed, paid, and / or unpaid). The (transmitter's) rate may be adjusted for some classes (e.g., high priority classes). If the needs of that class change, the rate can be changed (e.g., adjusted, modified, modified). If the receiver has critically low power, the rate may be reduced to reduce the receiver's power consumption for responding to the probe signal. In one example, the probe signal may be used to wirelessly transfer power to the receiver (Type 2 device), and the rate may be adjusted to control the amount of power transferred to the receiver.

[0119] The rate may be changed by (or based on) the following: a server (e.g., a hub device), a Type 1 device, and / or a Type 2 device. Control signals may be communicated between them. The server may monitor, track, predict, and / or anticipate the needs of the Type 2 device and / or the tasks performed by the Type 2 device, and control the Type 1 device to change the rate. The server may make scheduled changes to the rate according to a timetable. The server may detect an emergency and immediately change the rate. The server may detect a developing condition and gradually adjust the rate. Characteristics and / or STIs (e.g., movement information) may be monitored individually based on TSCIs associated with a particular Type 1 device and a particular Type 2 device, and / or jointly based on any TSCIs associated with a particular Type 1 device and any Type 2 device, and / or jointly based on any TSCIs associated with a particular Type 2 device and any Type 1 device, and / or globally based on any TSCIs associated with any Type 1 device and any Type 2 device. Any collaborative monitoring may relate to: a user, a user account, a profile, a home, a map of the location, an environmental model of the location, and / or a user history.

[0120] A first channel between a Type 1 device and a Type 2 device may be different from a second channel between another Type 1 device and another Type 2 device. The two channels may be associated with different frequency bands, bandwidths, carrier frequencies, modulations, wireless standards, coding, encryption, payload characteristics, networks, network IDs, SSIDs, network characteristics, network settings, and / or network parameters. The two channels may be associated with different types of wireless systems (e.g., two of the following: WiFi, LTE, LTE-A, LTE-U, 2.5G, 3G, 3.5G, 4G, Beyond 4G, 5G, 6G, 7G, cellular network standards, UMTS, 3GPP, GSM, EDGE, TDMA, FDMA, CDMA, WCDMA, TD-SCDMA, 802.11 systems, 802.15 systems, 802.16 systems, mesh networks, Zigbee, NFC, WiMax, Bluetooth, BLE, RFID, UWB, microwave systems, and radar-like systems). For example, one channel is WiFi and the other is LTE. The two channels may be associated with similar types of wireless systems, but different networks. For example, a first channel may be associated with a WiFi network named "Pizza and Pizza" in the 2.4 GHz band with a bandwidth of 20 MHz, while a second channel may be associated with a WiFi network with an SSID of "StarBud Hotspot" in the 5 GHz band with a bandwidth of 40 MHz. The two channels may be different channels within the same network (e.g., the "StarBud Hotspot" network).

[0121] In one embodiment, the wireless monitoring system can include training a classifier for multiple events at a location based on training TSCIs associated with the multiple events. The CIs or TSCIs associated with an event can consider / include wireless samples / characteristics / fingerprints associated with the event (and / or location, environment, object, object movement, state / emotional state / mental state / condition / stage / gesture / gait / action / movement / activity / daily activity / history / object event, etc.). For each of multiple known events occurring at a location at a respective training (e.g., survey, wireless survey, initial wireless survey) time associated with the known event, a respective training wireless signal (e.g., a respective series of training probe signals) can be transmitted to at least one first Type-2 heterogeneous wireless device through a wireless multipath channel at the location at the respective training time by an antenna of the first Type-1 heterogeneous wireless device using a processor, memory, and set of instructions of the first Type-1 device.

[0122] At least one respective time series of training CIs (training TSCIs) may be asynchronously acquired by each of the at least one first Type-2 device from the (respective) training signal. The CIs may be CIs of a channel between the first Type-2 device and the first Type-1 device at a training time associated with a known event. The at least one training TSCI may be preprocessed. The training may be a wireless survey (e.g., during installation of the Type-1 and / or Type-2 devices).

[0123] For a current event occurring within a location in a current time period, a current wireless signal (e.g., a series of current probe signals) may be transmitted to at least one second Type 2 heterogeneous wireless device through a channel of the location in a current time period related to the current event by an antenna of the second Type 1 heterogeneous wireless device using a processor, memory, and set of instructions of the second Type 1 device. At least one time series of current CIs (current TSCIs) may be asynchronously obtained by each of the at least one second Type 2 device from the current signals (e.g., the series of current probe signals). The CIs may be CIs of a channel between the second Type 2 device and the second Type 1 device in a current time period related to the current event. The at least one current TSCI may be preprocessed.

[0124] A classifier may be applied by at least one second Type 2 device to classify at least one current TSCI obtained from the series of current probe signals, to classify at least a portion of a particular current TSCI, and / or to classify a combination of at least a portion of a particular current TSCI with another portion of another TSCI. The classifier may divide the TSCIs (or features / STIs or other analysis or output responses) into clusters and associate the clusters with particular events / objects / subjects / locations / movements / activities. Labels / tags may be generated for the clusters. The clusters may be stored and searched. A classifier may be applied to associate the current TSCI (or perhaps a characteristic / STI or other analysis / output response related to the current event) with: a cluster, a known / specific event, a class / category / group / grouping / list / cluster, a set of known events / object / location / movement / activity, an unknown event, a class / category / group / grouping / list / cluster, a set of unknown events / object / location / movement / activity, and / or another event / object / location / movement / activity / class / category / group / grouping / list / cluster. Each TSCI may include at least one CI, each associated with a respective timestamp. Two TSCIs associated with two Type 2 devices differ by different start times, durations, stop times, amount of CIs, sampling frequencies, and sampling periods. The CIs may have different characteristics. The first and second Type 1 devices may be at the same location. They may be the same device. At least one second Type 2 device (or their location) may be a replacement for at least one first Type 2 device (or their location). A particular second Type 2 device and a particular first Type 2 device may be the same device. The subset of first Type 2 devices and the subset of second Type 2 devices may be the same. At least one second Type 2 device and / or the subset of at least one second Type 2 device may be a subset of at least one first Type 2 device.At least one first Type 2 device and / or a subset of at least one first Type 2 device may be a replacement for a subset of at least one second Type 2 device. At least one second Type 2 device and / or a subset of at least one second Type 2 device may be a replacement for a subset of at least one first Type 2 device. At least one second Type 2 device and / or a subset of at least one second Type 2 device may be in the same respective locations as a subset of at least one first Type 2 device. At least one first Type 2 device and / or a subset of at least one first Type 2 device may be in the same respective locations as a subset of at least one second Type 2 device.

[0125] The antenna of the Type 1 device and the antenna of the second Type 1 device may be in the same location. The antenna of the at least one second Type 2 device and / or the antenna of the subset of at least one second Type 2 device may be in the same respective location as each antenna of the subset of at least one first Type 2 device. The antenna of the at least one first Type 2 device and / or the antenna of the subset of at least one first Type 2 device may be in the same respective location as each antenna of the subset of at least one second Type 2 device.

[0126] A first section of a first duration of a first TSCI and a second section of a second duration of a second TSCI may be aligned. A map between items in the first section and items in the second section may be calculated. The first section may include a first segment (e.g., a subset) of the first TSCI having a first start / end time and / or another segment (e.g., a subset) of the processed first TSCI. The processed first TSCI may be the first TSCI processed by the first operation. The second section may include a second segment (e.g., a subset) of the second TSCI having a second start time and a second end time, and another segment (e.g., a subset) of the processed second TSCI. The processed second TSCI may be the second TSCI processed by the second operation. The first operation and / or the second operation may include subsampling, resampling, interpolation, filtering, transformation, feature extraction, preprocessing, and / or other operations.

[0127] A first item in a first section may be mapped to a second item in a second section. A first item in a first section may also be mapped to another item in a second section. Another item in a first section may also be mapped to a second item in a second section. The mapping may be one-to-one, one-to-many, many-to-one, or many-to-many. At least one function of at least one of the first item in a first section of a first TSCI, another item in a first TSCI, a timestamp of the first item, a time difference of the first item, a time difference of the first item, an adjacent timestamp of the first item, an adjacent timestamp of the first item, another timestamp related to the first item, a second item in a second section of a second TSCI, another item in a second TSCI, a timestamp of the second item, a time difference of the second item, a time difference of the second item, an adjacent timestamp of the second item, and another timestamp related to the second item may satisfy at least one constraint.

[0128] One constraint may be that the difference between the timestamp of the first item and the timestamp of the second item may be bounded upper by an adaptive (and / or dynamically adjusted) upper threshold and lower by an adaptive lower threshold.

[0129] The first section may be the entire first TSCI. The second section may be the entire second TSCI. The first time duration may be equal to the second time duration. The time duration of a section of a TSCI may be determined adaptively (and / or dynamically). A provisional section of the TSCI may be calculated. A start time and an end time of a section (e.g., provisional section, section) may be determined. A section may be determined by removing the start and end portions of the provisional section. The start portion of the provisional section may be determined as follows: Iteratively, items in the provisional section with increasing timestamps may be considered as the current item, one item at a time.

[0130] At each iteration, at least one activity measure / index may be calculated and / or considered. The at least one activity measure may be associated with at least one of the following: a current item associated with the current timestamp, a past item in the provisional section with a timestamp not greater than the current timestamp, and / or a future item in the provisional section with a timestamp not less than the current timestamp. If at least one criterion (e.g., quality criterion, signal quality condition) associated with the at least one activity measure is met, the current item may be added to the beginning of the provisional section.

[0131] The at least one criterion associated with the activity measure may include at least one of the following: (a) the activity measure is less than an adaptive (dynamically adjusted) upper threshold, (b) the activity measure is greater than an adaptive lower threshold, (c) the activity measure is less than an adaptive upper threshold consecutively for at least a predetermined amount of consecutive timestamps, (d) the activity measure is greater than an adaptive lower threshold consecutively for at least another predetermined amount of consecutive timestamps, (e) the activity measure is less than an adaptive upper threshold consecutively for at least a predetermined percentage of a predetermined amount of consecutive timestamps, (f) the activity measure is greater than an adaptive lower threshold consecutively for at least another predetermined percentage of another predetermined amount of consecutive timestamps, (g) another activity measure associated with another timestamp related to the current timestamp is less than another adaptive upper threshold and greater than another adaptive lower threshold, or (h) at least one activity measure associated with at least one respective timestamp related to the current timestamp is less than a respective upper threshold and greater than a respective lower threshold. (i) the percentage of timestamps associated with the activity measures in a set of timestamps associated with the current timestamp that are less than their respective upper thresholds and greater than their respective lower thresholds exceeds a threshold; and (j) another criterion (e.g., quality criterion, signal quality condition).

[0132] The activity measure / index associated with the item at time T1 may include at least one of the following: (1) a first function of the item at time T1 and the item at time T1-D1, where D1 is a predetermined positive quantity (e.g., a fixed time offset); (2) a second function of the item at time T1 and the item at time T1+D1; (3) a third function of the item at time T1 and the item at time T2, where T2 is a predetermined quantity (e.g., a fixed initial reference time; T2 may change (e.g., adjusted, varied, modified) over time; T2 may be updated periodically; T2 may be the start of a period and T1 may be a sliding time in the period); and (4) a fourth function of the item at time T1 and other items.

[0133] At least one of the first function, the second function, the third function, and / or the fourth function may be a function (e.g., F(X, Y, ...)) with at least two arguments X and Y. The two arguments may be scalars. The function (e.g., F) may be at least one of X, Y, (XY), (YX), abs(XY), X^a, Y^b, abs(X^aY^b), (XY)^a, (X / Y), (X+a) / (Y+b), (X^a / Y^b), and ((X / Y)^ab), where a and b may be certain predetermined quantities. For example, the function may be simply abs(XY), or (XY)^2, (XY)^4. The function may be a robust function. For example, the function is (XY)^2 when abs(XY) is less than a threshold T, and (XY)+a when abs(XY) is greater than T. Alternatively, the function may be a constant when abs(XY) is greater than T. Also, when abs(Xy) is greater than T, the function may be bounded by a slowly increasing function so that outliers cannot significantly affect the results. Another example of this function may be (abs(X / Y)-a) (where a=1). In this way, when X=Y (i.e., no change or activity), the function yields a value of 0. When X is greater than Y, (X / Y) is greater than 1 (assuming X and Y are positive), and the function is positive. When X is less than Y, (X / Y) is less than 1, and the function is negative. In another example, both arguments X and Y may be n-tuples, with X = (x_1, x_2, ..., x_n) and Y = (y_1, y_2, ..., y_n). The function may be at least one of x_i, y_i, (x_i - y_i), (y_ix_i), abs(x_i - y_i), x_i^a, y_i^b, abs(x_i^a - y_i^b), (x_i - y_i)^a, (x_i / y_i), (x_i + a) / (y_i + b), (x_i^a / y_i^b), and ((x_i / y_i)^ab), where i is a component index of the n-tuples X and Y, and 1 <= i <= n. For example, the component index of x_1 is i=1, and the component index of x_2 is i=2.The function may include a sum of another function per component of at least one of x_i, y_i, (x_i-y_i), (y_ix_i), abs(x_i-y_i), x_i^a, y_i^b, abs(x_i^a-y_i^b), (x_i-y_i)^a, (x_i / y_i), (x_i+a) / (y_i +b), (x_i^a / y_i^b), and ((x_i / y_i)^ab), where i is the component index of the n-tuples X and Y. For example, the function may be in the form sum_{i=1}^n(abs(x_i / y_i)-1) / n, or sum_{i=1}^nw_i*(abs(x_i / y_i)-1), where w_i is the weight of component i.

[0134] The map may be computed using dynamic time warping (DTW). The DTW may include constraints on at least one of the map, the items of the first TSCI, the items of the second TSCI, the first duration, the second duration, the first section, and / or the second section. Suppose the i^{th} domain item is mapped to the j^{th} range item in the map. The constraint may be on the allowable combinations of i and j (constraints on the relationship between i and j). A mismatch cost between the first section of the first duration of the first TSCI and the second section of the second duration of the second TSCI may be computed.

[0135] The first section and the second section may be aligned such that a map including a plurality of links may be established between a first item of the first TSCI and a second item of the second TSCI. Each link may associate one of the first items with a first timestamp and one of the second items with a second timestamp. A mismatch cost between the aligned first section and the aligned second section may be calculated. The mismatch cost may include a function of a cost for an item between the first item and the second item associated by a particular link of the map and a cost for a link associated with the particular link of the map.

[0136] The aligned first section and the aligned second section may be represented as a first vector and a second vector, respectively, of the same vector length. The mismatch cost may include at least one of a dot product, a dot product-like measure, a correlation-based measure, a correlation indicator, a covariance-based measure, a discrimination score, a distance, a Euclidean distance, an absolute distance, an Lk distance (e.g., L1, L2,...), a weighted distance, a distance-like measure, and / or another similarity value between the first vector and the second vector. The mismatch cost may be normalized by the respective vector lengths.

[0137] A parameter derived from the mismatch cost between a first section of a first duration of a first TSCI and a second section of a second duration of a second TSCI may be modeled with a statistical distribution. At least one of a scale parameter, a position parameter, and / or another parameter of the statistical distribution may be estimated. The first section of the first duration of the first TSCI may be a sliding section of the first TSCI. The second section of the second duration of the second TSCI may be a sliding section of the second TSCI. A first sliding window may be applied to the first TSCI, and a corresponding second sliding window may be applied to the second TSCI. The first sliding window of the first TSCI and the corresponding second sliding window of the second TSCI may be aligned.

[0138] A mismatch cost between the aligned first sliding window of the first TSCI and the corresponding aligned second sliding window of the second TSCI can be calculated, and the current event may be associated with at least one of a known event, an unknown event, and / or another event based on the mismatch cost.

[0139] The classifier may be applied to at least one of each first section of a first duration of the first TSCI and / or each second section of a second duration of the second TSCI to obtain at least one provisional classification result, each provisional classification result being associated with a respective first section and a respective second section.

[0140] The current event can be associated with at least one of a known event, an unknown event, a class / category / group / grouping / list / set of unknown events, and / or another event based on the mismatch cost. The current event can be associated with at least one of a known event, an unknown event, and / or another event based on the most numerous provisional classification results in the multiple sections of the first TSCI and the multiple sections of the corresponding second TSCI. For example, the current event can be associated with a specific known event if the mismatch cost points to the specific known event N consecutive times (e.g., N=10). In another example, the current event can be associated with a specific known event if the percentage of mismatch costs within the immediately preceding N consecutive N times that point to the specific known event exceeds a predetermined threshold (e.g., >80%). In another example, the current event can be associated with the known event that achieves the smallest mismatch cost the most times in time. The current event can be associated with the known event that achieves the smallest overall mismatch cost, which is a weighted average of at least one mismatch cost associated with at least one first section. The current event may be associated with a particular known event that achieves a minimum of another overall cost. The current event may be associated with an “unknown event” if none of the known events achieves a mismatch cost lower than a first threshold T1 in a sufficient percentage of at least one first section. The current event may also be associated with an “unknown event” if none of the known events achieves an overall mismatch cost lower than a second threshold T2. The current event may be associated with at least one of a known event, an unknown event, and / or another event based on the mismatch cost and additional mismatch cost associated with at least one additional section of the first TSCI and at least one additional section of the second TSCI. The known event may include at least one of a door-closed event, a door-opened event, a window-closed event, a window-opened event, a multi-state event, an on-state event, an off-state event, an intermediate state event, a continuous state event, a discrete state event, a person-present event, a person-absent event, a life-present event, and / or a life-absent event.

[0141] A projection for each CI may be trained using a dimensionality reduction method based on the training TSCI. The dimensionality reduction method may include at least one of principal component analysis (PCA), PCA with different kernels, independent component analysis (ICA), Fisher's linear discriminant, vector quantization, supervised learning, unsupervised learning, self-organizing maps, autoencoders, neural networks, deep neural networks, and / or another method. The projection may be applied to at least one of the training TSCI associated with at least one event and / or the current TSCI for a classifier. A classifier for at least one event may be trained based on the projection associated with the at least one event and the associated training TSCI. The at least one current TSCI may be classified / categorized based on the projection and the current TSCI. The projection may be retrained using at least one of the dimensionality reduction method and another dimensionality reduction method based on at least one of the training TSCI, the at least one current TSCI before retraining the projection, and / or additional training TSCI. Other dimensionality reduction methods may include at least one of principal component analysis (PCA), PCA with different kernels, independent component analysis (ICA), Fisher's linear discriminant, vector quantization, supervised learning, unsupervised learning, self-organizing maps, autoencoders, neural networks, deep neural networks, and / or other methods. The classifier for the at least one event may be retrained based on at least one of the retrained projections, the training TSCI associated with the at least one event, and / or the at least one current TSCI. The at least one current TSCI may be classified based on the retrained projections, the retrained classifier, and / or the current TSCI.

[0142] Each CI may include a vector of complex values. Each complex value may be preprocessed to provide a magnitude of the complex value. Each CI may be preprocessed to provide a vector of non-negative real numbers containing the magnitude of the corresponding complex value. Each training TSCI may be weighted in training the projection. The projection may include multiple projection components. The projection may include at least one most significant projection component. The projection may include at least one projected component that may be useful to the classifier.

[0143] Channel information (CI) includes signal strength, signal amplitude, signal phase, spectral power measures, modem parameters (e.g., used in connection with modulation / demodulation in digital communication systems such as WiFi, 4G / LTE, etc.), dynamic beamforming information, transfer function components, radio state (e.g., used in digital communication systems to decode digital data, baseband processing state, RF processing state, etc.), measurable variables, sensing data, layer coarse-grained / fine-grained information (e.g., physical layer, data link layer, MAC layer, etc.), digital settings, gain settings, RF filter settings, RF front-end switch settings, DC offset settings, DC correction settings, IQ correction settings, effects of the environment (e.g., location) on the radio signal during propagation, input signal (radio signal transmitted by Type 1 device) and output signal (Type 2) The CI may be associated with or include a transformation into a radio signal (radio signal received by the device), steady state behavior of the environment, a condition profile, radio channel measurements, a received signal strength indicator (RSSI), channel state information (CSI), a channel impulse response (CFR), a channel frequency response (CFR), characteristics of frequency components (e.g., subcarriers) in the bandwidth, channel characteristics, channel response, a timestamp, auxiliary information, data, metadata, user data, account data, access data, security data, session data, status data, supervision data, home data, identification (ID), identifiers, device data, network data, proximity data, environmental data, real-time data, sensor data, stored data, encrypted data, compressed data, protected data, and / or other channel information. Each CI may be associated with a timestamp and / or a time of arrival. The CSI can equalize / restore / minimize / reduce multipath channel effects (transmission channel) and demodulate signals similar to those transmitted by the transmitter through the multipath channel. The CI may be associated with information related to a frequency band, a frequency signature, a frequency phase, a frequency amplitude, a frequency trend, a frequency characteristic, a frequency-like characteristic, a time domain element, a frequency domain element, a time-frequency domain element, an orthogonal decomposition characteristic, and / or a non-orthogonal decomposition characteristic of a signal passing through a channel. The TSCI may be a stream of wireless signals (e.g., CIs).

[0144] The CI may be pre-processed, processed, post-processed, stored (e.g., in a local memory, portable / mobile memory, removable memory, storage network, cloud memory, in a volatile manner, in a non-volatile manner), retrieved, transmitted, and / or received. One or more modem parameters and / or radio condition parameters may be kept constant. The modem parameters may be applied to a radio subsystem. The modem parameters may represent radio conditions. A motion detection signal (e.g., a baseband signal and / or packets decoded / demodulated from the baseband signal, etc.) may be obtained by processing (e.g., downconverting) a first radio signal (e.g., an RF / WiFi / LTE / 5G signal) by the radio subsystem using the radio conditions represented by the stored modem parameters. The modem parameters / radio conditions may be updated (e.g., using previous modem parameters or previous radio conditions). Both the previous and updated modem parameters / radio conditions may be applied to a radio subsystem of a digital communication system. Both the previous and updated modem parameters / radio conditions may be compared / analyzed / processed / monitored in a task.

[0145] The channel information may also be modem parameters (e.g., stored or newly calculated) used to process the wireless signal. The wireless signal may include multiple probe signals. The same modem parameters can be used to process multiple probe signals. The same modem parameters can also be used to process multiple wireless signals. The modem parameters may include parameters indicating settings or overall configurations for operation of the radio subsystem or baseband subsystem (or both) of the wireless sensor device. The modem parameters may include one or more of gain settings, RF filter settings, RF front-end switch settings, DC offset settings, or IQ compensation settings for the radio subsystem, or digital DC correction settings, digital gain settings, and / or digital filtering settings (e.g., for the baseband subsystem). CI may also relate to information related to time, time signature, timestamp, time amplitude, time phase, time trend, and / or time characteristics of a signal. CI may be associated with information related to the time-frequency division, signature, amplitude, phase, trend, and / or characteristics of a signal. CI may relate to signal decomposition. A CI may relate to information related to direction, angle of arrival (AoA), angle of a directional antenna, and / or phase of a signal passing through a channel. A CI may relate to the attenuation pattern of a signal passing through a channel. Each CI may be associated with a Type 1 device and a Type 2 device. Each CI may be associated with an antenna of a Type 1 device and an antenna of a Type 2 device.

[0146] The CI can be obtained from communication hardware (e.g., a Type 2 device or a Type 1 device) capable of providing the CI. The communication hardware can be a WiFi-enabled chip / IC (integrated circuit), a chip compliant with 802.11 or 802.16 or other wireless / wireless standards, a next-generation WiFi-enabled chip, an LTE-enabled chip, a 5G-enabled chip, a 6G / 7G / 8G-enabled chip, a Bluetooth-enabled chip, an NFC (near field communication)-enabled chip, a BLE (Bluetooth low energy)-enabled chip, a UWB chip, or other communication chips (e.g., Zigbee, WiMax, mesh networks). The communication hardware calculates the CI and stores the CI in a buffer memory so that the CI is available for extraction. The CI can include data related to channel state information (CSI) and / or at least one matrix. The at least one matrix can be used for channel equalization, beamforming, etc. The channel can be associated with a location. Attenuation can be due to signal propagation at the location, signal propagation through / at / near the air (e.g., the air at the location), reflection, refraction, diffraction, refractive media / reflective surfaces such as walls, doors, furniture, obstacles, and / or barriers, etc. Attenuation can be due to reflections off surfaces and obstacles (e.g., reflective surfaces, obstacles) such as floors, ceilings, furniture, fixtures, objects, people, pets, etc. Each CI can be associated with a timestamp. Each CI can include N components (e.g., N frequency-domain components in CFR, N time-domain components in CIR, or N decomposed components). Each component can be associated with a component index. Each component can be a real, imaginary, or complex quantity, magnitude, phase, flag, and / or set. Each CI can include a vector or matrix of complex numbers, a set of mixed quantities, and / or a multidimensional collection of at least one complex number.

[0147] Components of the TSCI associated with a particular component index may form respective component time series associated with the respective index. The TSCI may be divided into N component time series. Each individual component time series is associated with a respective component index. Object motion characteristics / STI may be monitored based on the component time series. In one example, one or more ranges of CI components (e.g., one range from component 11 to component 23, a second range from component 44 to component 50, and a third range having only one component) may be selected based on some criteria / cost function / signal quality metric (e.g., based on signal-to-noise ratio and / or interference level) for further processing.

[0148] A component-specific characteristic of the TSCI component-feature time series may be calculated. The component-specific characteristic may be a scalar (e.g., energy) or a function with a domain and range (e.g., autocorrelation function, transform, inverse transform). The object motion characteristic / STI may be monitored based on the component-specific characteristics. A total characteristic (e.g., aggregate characteristic) of the TSCI may be calculated based on the component-specific characteristics of each component time series of the TSCI. The total characteristic may be a weighted average of the component-specific characteristics. The object motion characteristic / STI may be monitored based on the total characteristic. A total quantity may be a weighted average of the individual quantities.

[0149] Type 1 devices and Type 2 devices may support WiFi, WiMax, 3G / 3G Beyond, 4G / 4G Beyond, LTE, LTE-A, 5G, 6G, 7G, Bluetooth, NFC, BLE, Zigbee, UWB, UMTS, 3GPP, GSM, EDGE, TDMA, FDMA, CDMA, WCDMA, TD-SCDMA, mesh networks, proprietary wireless systems, IEEE 802.11 standards, 802.15 standards, 802.16 standards, 3GPP standards, and / or other wireless systems.

[0150] A common wireless system and / or a common wireless channel may be shared by a Type 1 transceiver and / or at least one Type 2 transceiver. The at least one Type 2 transceiver may transmit respective signals simultaneously (or: asynchronously, synchronously, sporadically, continuously, repeatedly, in parallel, simultaneously, and / or at one time) using the common wireless system and / or the common wireless channel. The Type 1 transceiver may transmit signals to the at least one Type 2 transceiver using the common wireless system and / or the common wireless channel.

[0151] Each Type 1 device and Type 2 device may have at least one transmit and receive antenna. Each CI may be associated with one of the transmit antennas of the Type 1 device and one of the receive antennas of the Type 2 device. Each pair of transmit and receive antennas may be associated with a link, path, communication path, signal hardware path, etc. For example, if a Type 1 device has M (e.g., 3) transmit antennas and a Type 2 device has N (e.g., 2) receive antennas, there may be M x N (e.g., 3 x 2 = 6) links or paths. Each link or path may be associated with a TSCI.

[0152] At least one TSCI may correspond to various antenna pairs between a Type 1 device and a Type 2 device. The Type 1 device may have at least one antenna. The Type 2 device may also have at least one antenna. Each TSCI may be associated with an antenna of the Type 1 device and an antenna of the Type 2 device. Averaging or weighted averaging across antenna links may be performed. The averaging or weighted averaging may be across at least one TSCI. The averaging may optionally be performed over a subset of the at least one TSCI corresponding to a subset of the antenna pairs.

[0153] The timestamps of some CIs of a TSCI may be irregular and may be corrected so that the corrected timestamps of the time-corrected CIs are evenly spaced in time. In the case of multiple Type 1 devices and / or multiple Type 2 devices, the corrected timestamps may be related to the same clock or different clocks. An original timestamp associated with each of the CIs may be determined. The original timestamps may not be evenly spaced in time. The original timestamps of all CIs of a particular portion of a particular TSCI in the current sliding time window may be corrected so that the corrected timestamps of the time-corrected CIs are evenly spaced in time.

[0154] Characteristics and / or STI (e.g., motion information) may include location, location coordinates, change in location, location (e.g., initial location, new location), location on map, height, horizontal position, vertical position, distance, displacement, speed, acceleration, rotational speed, rotational acceleration, direction, motion angle, orientation, direction of motion, rotation, path, deformation, translation, contraction, extension, gait, gait cycle, head movement, repetitive movement, periodic movement, pseudo-periodic movement, impulse movement, sudden movement, falling movement, transient movement, behavior, transient behavior, movement cycle, movement frequency, time trend, temporal profile, temporal characteristics, occurrence, change, temporal change, change in CI, change in frequency, change in timing, change in gait cycle, timing, start time, start time, end time, duration, movement history, movement type, movement classification, frequency, frequency spectrum, frequency characteristics, presence, absence, proximity, proximity retract, retract, object identification / identifier, object composition, head movement velocity, head movement direction, mouth-related rate, eye-related rate, respiration rate, heart rate, tidal volume, respiration depth, inhalation time, exhalation time, inhalation to exhalation time ratio, airflow rate, heart rate interval, heart rate variability, hand movement rate, hand movement direction, leg movement, body movement, walking speed, hand movement velocity, position characteristics, object movement-related characteristics (e.g., change in position / location), tool movement, machine movement, compound movement, and / or combination of multiple movements, event, signal statistics, signal dynamics, anomaly, movement statistics, movement parameters, motion detection indication, motion magnitude, motion phase, similarity score, distance score, Euclidean distance, weighted distance, L_1 norm, L_2 norm, L_k norm for k>2, statistical distance, correlation, correlation indicator,Autocorrelation, covariance, autocovariance, cross-covariance, inner product, Cartesian product, motion signal transformation, motion features, motion presence, motion absence, motion localization, motion discrimination, motion recognition, object presence, object absence, object entrance, object exit, object change, movement cycle, number of movements, gait cycle, movement rhythm, movement deformation, gesture, handwriting, head movement, mouth movement, cardiac movement, visceral movement, motion trend, size, length, area, volume, volume, shape, morphology, tag, start / start position, end position, open The characteristics and / or STIs may include start / start quantity, end quantity, event, fall event, security event, accident event, home event, office event, factory event, warehouse event, manufacturing event, assembly line event, maintenance event, car-related event, navigation event, tracking event, door event, door open event, door close event, window event, window open event, window close event, repeatable event, one-time event, consumption quantity, unconsumed quantity, state, physical state, health state, comfort state, emotional state, mental state, other event, analysis, output response, and / or other information. Characteristics and / or STIs may be calculated / monitored based on features calculated from the CI or TSCI (e.g., feature calculation / extraction). Static segments or profiles (and / or dynamic segments / profiles) may be identified / calculated / analyzed / monitored / extracted / acquired / marked / presented / indicated / highlighted / stored / communicated based on feature analysis. Analysis may include motion detection / motion assessment / presence detection. Computational workloads may be shared among Type 1 devices, Type 2 devices, and other processors.

[0155] The Type 1 device and / or the Type 2 device may be a local device, which may be a smartphone, a smart device, a TV, a sound bar, a set-top box, an access point, a router, a repeater, a wireless signal repeater / extender, a remote control, a speaker, a fan, a refrigerator, a microwave oven, a coffee machine, a hot water pot, an appliance, a table, a chair, a light, a lamp, a door lock, a camera, a microphone, a motion sensor, a security device, a fire hydrant, a garage door switch, a power adapter, a computer, a dongle, a computer peripheral, an electronic pad, a sofa, a tile, an accessory, a home device, a vehicle device, an office device, a building equipment, a manufacturing equipment, a watch, a glass, a clock, a television, an oven, an air conditioner, an accessory, a utility, an appliance, a smart machine, a smart vehicle, an Internet of Things (IoT), a smart house, a smart office, a smart building, a smart parking lot, a smart system, and other devices.

[0156] Each Type 1 device may be associated with a respective identifier (e.g., ID). Each Type 2 device may also be associated with a respective identification (ID). The ID may include numbers, a combination of text and numbers, a name, a password, an account, an account ID, a web link, a web address, an index to some information, and / or another ID. The ID may be assigned. The ID may be assigned by hardware (e.g., hardwired, via a dongle, and / or other hardware), software, and / or firmware. The ID may be stored (e.g., in a database, in memory, in a server (e.g., a hub device), in the cloud, locally stored, remotely stored, permanently stored, or temporarily stored) and may be searched. The ID may be associated with at least one record, account, user, household, address, phone number, social security number, customer number, another ID, another identifier, timestamp, and / or collection of data. The ID and / or a portion of the ID of the Type 1 device may be made available to the Type 2 device. The ID may be used by Type 1 devices and / or Type 2 devices for registration, initialization, communication, identification, verification, detection, recognition, authentication, access control, cloud access, networking, social networking, logging, recording, cataloging, classification, tagging, association, pairing, transactions, electronic transactions, and / or intellectual property control.

[0157] The object may be a person, user, subject, passenger, child, elderly, infant, sleeping infant, infant in a vehicle, patient, worker, high value worker, expert, medical specialist, waiter, customer in a mall, traveller at an airport / train station / bus terminal / shipping terminal, staff / laborer / customer service person in a factory / mall / supermarket / office / workplace, service person in a sewer / air ventilation system / lift well, lift in a lift well, elevator, inmate, person to be tracked / monitored, animal, plant, living thing, pet, dog, cat, smartphone, phone accessory, computer, tablet, portable computer, dongle, computer accessory equipment, network equipment, WiFi equipment, IoT device, smart watch, smart glasses, smart device, speaker, key, smart key, wallet, wallet, handbag, backpack, goods, cargo, luggage, equipment, motor, machine, air conditioner, fan, air conditioning equipment, lighting fixture, movable light, television, camera, audio and / or video equipment, It can be stationery, surveillance equipment, parts, signs, tools, carts, tickets, parking passes, passes, plane tickets, credit cards, plastic cards, access cards, food packaging, utensils, tables, chairs, cleaning equipment / tools, vehicles, automobiles, cars in parking facilities, goods in a warehouse / store / supermarket / distribution center, boats, bicycles, airplanes, drones, remote controlled cars / planes / boats, robots, manufacturing equipment, assembly lines, materials / unfinished parts / robots / trolleys / transports on a factory floor, tracked objects in an airport / shopping mart / supermarket, non-objects, absence of objects, presence of objects, objects with shape, objects changing shape, shapeless objects, mass of a fluid, mass of a liquid, mass of a gas / smoke, fire, flames, electromagnetic (EM) sources, EM media, and / or other objects.The object itself may be communicatively coupled to several networks, such as WiFi, MiFi, 3G / 4G / LTE / 5G / 6G / 7G, Bluetooth, NFC, BLE, WiMax, Zigbee, UMTS, 3GPP, GSM, EDGE, TDMA, FDMA, CDMA, WCDMA, TD-SCDMA, mesh networks, ad-hoc networks, and / or other networks. The object itself may be AC-powered and bulky, and may be moved during installation, cleaning, maintenance, renovation, etc. The object may also be placed on a mobile platform, such as a lift, pad, mobile platform, elevator, conveyor belt, robot, drone, forklift, car, boat, or vehicle. The object may have multiple parts, each with a different motion (e.g., change of location / position). For example, the object may be a person walking in front. While walking, his left and right hands may move in different directions with different instantaneous speeds, accelerations, and motions.

[0158] The wireless transmitter (e.g., a Type 1 device), the wireless receiver (e.g., a Type 2 device), another wireless transmitter, and / or another wireless receiver may travel with the object and / or another object (e.g., in a previous trip, a current trip, and / or a future trip). They may be communicatively coupled to one or more nearby devices. They may transmit TSCIs and / or information related to the TSCIs to nearby devices and / or to each other. They may be associated with nearby devices. The wireless transmitter and / or wireless receiver may be part of a small (e.g., coin-sized, cigarette-pack-sized, or even smaller) lightweight portable device. The portable device may be wirelessly coupled to the nearby device.

[0159] The nearby device may be a smartphone, an iPhone, an Android phone, a smart device, a smart appliance, a smart vehicle, a smart gadget, a smart TV, a smart refrigerator, a smart speaker, a smart watch, smart glasses, a smart pad, an iPad, a computer, a wearable computer, a notebook computer, a gateway. The nearby device may be connected to a cloud server, a local server (e.g., a hub device), and / or other servers via the Internet, a wired Internet connection, and / or a wireless Internet connection. The nearby device may be portable. The portable device, nearby devices, local server (e.g., hub device), and / or cloud server can share tasks (e.g., acquiring TSCI, determining object characteristics / STI related to object movement (e.g., position / change in position), calculating time series of power (e.g., signal strength) information, determining / calculating specific functions, searching for local extrema, classification, identifying specific values ​​of offset time, denoising, processing, simplification, cleaning, wireless smart sensing tasks, extracting CIs from signals, switching, segmenting, estimating trajectory / path / track, processing maps, processing trajectory / path / track based on environmental models / constraints / limits, correction, correction adjustment, tuning, map-based (or model-based) correction, error detection, checking for boundary hits, thresholding) and calculation and / or storage of information (e.g., TSCI). Nearby devices may not move with the object. Nearby devices can be portable / non-portable / mobile / non-mobile. Nearby devices can use battery power, solar, AC power, and / or other power sources. The nearby device may have replaceable / non-replaceable and / or rechargeable / non-rechargeable batteries. The nearby device may resemble the object. The nearby device may have the same (and / or similar) hardware and / or software as the object.The nearby devices may be smart devices, network-enabled devices, devices with connections to WiFi / 3G / 4G / 5G / 6G / Zigbee / Bluetooth / NFC / UMTS / 3GPP / GSM / EDGE / TDMA / FDMA / CDMA / WCDMA / TD-SCDMA / ad hoc networks / other networks, smart speakers, smart watches, smart clocks, smart appliances, smart machines, smart appliances, smart tools, smart vehicles, Internet of Things (IoT) devices, Internet-enabled devices, computers, portable computers, tablets, and other devices. Nearby devices and / or at least one processor associated with a wireless receiver, a wireless transmitter, another wireless receiver, another wireless transmitter, and / or a cloud server (in the cloud) may determine an initial STI for the object. Two or more of them may jointly determine initial spatio-temporal information. Two or more of them may share intermediate information in determining the initial STI (e.g., initial location).

[0160] In one example, a wireless transmitter (e.g., a Type 1 device or a tracker bot) moves with an object. The wireless transmitter can send a signal to a wireless receiver (e.g., a Type 2 device or an Origin Register) to determine the object's initial STI (e.g., initial location). The wireless transmitter can also send a signal and / or another signal to another wireless receiver (e.g., another Type 2 device or another Origin Register) to monitor the object's movement (spatio-temporal information). The wireless receiver can also receive a signal and / or another signal from the wireless transmitter and / or another wireless transmitter to monitor the object's movement. The location of the wireless receiver and / or another wireless receiver can be known. In another example, a wireless receiver (e.g., a Type 2 device or a tracker bot) can move with the object. The wireless receiver can receive a signal transmitted from the wireless transmitter (e.g., a Type 1 device or an Origin Register) to determine the object's initial spatio-temporal information (e.g., initial location). The wireless receiver may also receive a signal and / or another signal from another wireless transmitter (e.g., another Type 1 device or another origin register) for monitoring the current movement (e.g., space-time information) of the object. The wireless transmitter may also transmit a signal and / or another signal to the wireless receiver and / or another wireless receiver (e.g., another Type 2 device or another tracker bot) for monitoring the movement of the object. The location of the wireless transmitter and / or another wireless transmitter may be known.

[0161] Locations include sensing areas, rooms, houses, offices, property, workspaces, corridors, lifts, lift wells, escalators, elevators, sewers, ventilation systems, stairs, assembly areas, ducts, air ducts, pipes, enclosed spaces, enclosed structures, semi-enclosed structures, enclosed areas with at least one wall, plants, machines, engines, structures, structures with wood, structures with glass, structures with metal, structures with walls, structures with doors, structures with gaps, structures with reflective surfaces, structures with liquids, buildings, rooftops, stores, factories, assembly lines, homes, etc. Hotel rooms, museums, classrooms, schools, universities, government buildings, warehouses, garages, malls, airports, train stations, bus terminals, hubs, transportation hubs, cargo terminals, government buildings, public facilities, schools, universities, entertainment venues, recreational facilities, hospitals, pediatric / neonatal wards, nursing homes, elderly care facilities, community centers, stadiums, playgrounds, fields, basketball courts, tennis courts, soccer stadiums, baseball fields, gymnasiums, garages, shopping marts, knolls, supermarkets, manufacturing facilities, parking facilities Facilities, construction sites, mining facilities, transportation facilities, highways, roads, valleys, forests, trees, terrain, landscapes, caves, patios, land, roads, amusement parks, urban areas, rural areas, suburban areas, metropolitan areas, gardens, squares, plazas, music halls, downtown facilities, open facilities, semi-open facilities, closed areas, train platforms, train stations, distribution centers, warehouses, shops, distribution centers, storage facilities, underground spaces, spatial (e.g. above ground, space) facilities, floating facilities, caves, tunnel facilities, indoor facilities, outdoor facilities, outdoor facilities with some walls / doors / reflective barriers Areas such as facilities, open facilities, semi-open facilities, automobiles, trucks, buses, vans, containers, ships / boats, submarines, trains, trams, airplanes, vehicles, mobile platforms, caves, tunnels, pipes, channels, metropolitan areas, downtown areas with relatively tall buildings, valleys, wells, ducts, pathways, gas lines, oil pipes, water pipes, interconnecting pathways / arrays / roads / tubes / cavities / caves / pipe-like structures / voids / fluid spaces, human bodies, animal bodies, body cavities, organs, bones, teeth, soft tissue, hard tissue, rigid tissue, non-hard tissue, blood / body fluid ducts, wind ducts, air ducts, burrows, etc. The location may be an indoor space, an outdoor space, and the location may include both inside and outside spaces.For example, a location can include both the inside and outside of a building. For example, a location can be a building with one or more floors, and part of the building can be underground. The shape of the building can be, for example, round, square, rectangular, triangular, or irregular. These are merely examples. The present disclosure can be used to detect events in other types of locations or spaces.

[0162] The wireless transmitter (e.g., a Type 1 device) and / or wireless receiver (e.g., a Type 2 device) may be embedded in a portable device (e.g., a module or a device having a module) that may travel with the object (e.g., in a previous travel and / or a current travel). The portable device may be communicatively coupled to the object using a wired connection (e.g., via USB, micro USB, Firewire, HDMI, serial port, parallel port, and other connectors) and / or a connection (e.g., Bluetooth, Bluetooth Low Energy (BLE), WiFi, LTE, NFC, ZigBee). The portable device may be a lightweight device. The portable device may be powered by batteries, rechargeable batteries, and / or AC power. The portable device may be very small (e.g., on the sub-millimeter and / or sub-centimeter scale) and / or small (e.g., coin-sized, card-sized, pocket-sized, or larger). The portable device may be large, bulky, and / or heavy installed machinery. Portable devices include WiFi hotspots, access points, Mobile WiFi (MiFi), dongles with USB / micro USB / Firewire / other connectors, smartphones, portable computers, computers, tablets, smart devices, Internet of Things (IoT) devices, WiFi enabled devices, LTE enabled devices, smart watches, smart glass, smart mirrors, smart antennas, smart batteries, smart lights, smart pens, smart rings, smart doors, smart windows, smart clocks, smart batteries, smart wallets, smart belts, smart handbags, smart cloth / garments, smart ornaments, smart packaging, smart paper / books / magazines / posters / printed materials / signage / displays / illuminated systems / lighting systems, smart keys / tools, smart bracelets / chains / necklaces / clothes / accessories, smart pads / cushions, smart tiles / blocks / bricks / building materials / other materials,Smart trash can / waste container, smart food carriage / storage, smart ball / racket, smart chair / sofa / bed, smart shoes / footwear / carpet / mat / shoe rack, smart gloves / handwear / ring / handwear, smart hat / cap / cosmetics / sticker / tattoo, smart mirror, smart toy, smart pill, smart cookware, smart bottle / food container, smart tool, smart device, IoT device, WiFi enabled device, network enabled device, 3G / 4G / 5G / 6G enabled device, UMTS device, 3GPP device, GSM device, EDGE device, TDMA device, FDMA device, CDMA device, WCDMA device, TD-SCDMA device, embedded device, embeddable device, air conditioner, refrigerator, heater, furnace, furniture, oven, cooking device , TV / Set Top Box (STB) / DVD Player / Audio Player / Video Player / Remote Control, Hi-Fi, Audio Device, Speaker, Lamp / Light, Wall, Door, Window, Roof, Tile / Roofing Shingle / Structure / Attic Structure / Device / Feature / Installation / Fixture, Lawn Mower / Garden Equipment / Tools / Machine Tools / Garage Tools, Garbage Can / Container, 20ft / 40ft Container, Storage Container, Factory / Production / Manufacturing Equipment, Repair Tool, Fluid Container, Machine, Installed Machine, Vehicle, Cart, Wagon, Warehouse Vehicle, Automobile, Bicycle, Motorcycle, Boat, Watercraft, Airplane, Basket / Box / Bag / Bucket / Container, Smart Plate / Cup / Bowl / Pot / Mat / Utensil / Kitchenware / Kitchen Accessories / Cabinet / Table / Chair / Tile / Light / Water Pipe / Faucet / Gas Range / Oven / Dishwasher / , etc. Portable devices may have batteries that may be replaceable, non-replaceable, rechargeable, and / or non-rechargeable. Portable devices may be charged wirelessly. The portable device may be a smart payment card. The portable device may be a payment card used in parking lots, highways, entertainment parks, or other locations / facilities requiring payment. The portable device may have an identity (ID) / identifier, as described above.

[0163] Events can be monitored based on the TSCI. The events can be object-related events such as an object (e.g., a person and / or a patient) falling, rolling, hesitating, resting, impact (e.g., a person hitting a punching bag, door, window, bed, chair, table, desk, cabinet, box, another person, animal, bird, flying, table, chair, ball, bowling ball, tennis ball, football, soccer ball, baseball, basketball, volleyball), two-body action (e.g., leaving a balloon, catching a fish, molding clay, writing a paper, a person typing into a computer), moving a car in a garage, a person carrying a smartphone and walking around an airport / mall / government office / etc, an autonomous mobile object / machine moving around (e.g., a vacuum cleaner, utility vehicle, car, drone, self-driving car), etc. The tasks or wireless smart sensing tasks are object detection, presence detection, proximity detection, object recognition, activity recognition, object verification, object counting, daily activity monitoring, health monitor, vital signs monitoring, health status monitoring, baby monitoring, elderly monitoring, sleep monitoring, sleep stage monitoring, gait monitoring, movement monitoring, tool detection, tool recognition, tool verification, patient detection, patient monitoring, patient verification, machine detection, machine verification, human detection, human recognition, human verification, baby detection, baby recognition, baby verification, human breathing detection, human breathing recognition, human breathing estimation, human breathing verification, human heart rate detection, human heart rate recognition, human heart rate estimation, human heart rate verification, fall detection, fall recognition, fall estimation, fall verification, emotion detection, emotion recognition, emotion estimation, emotion verification, motion detection, motion degree estimation, motion recognition, motion estimation, motion verification, periodic motion detection, periodic motion estimation, periodic motion verification, repetitive motion detection, periodic motion recognition, repetitive motion estimation, repetitive motion verification, static motion recognition, static motion detection, static motion estimation, static motion verification, cyclostationary motion detection, cyclostationary motion recognition, cyclostationary motion estimation, cyclostationary motion verification, transient motion detection, transient motion recognition, transient motion estimation, transient motion verification, trend detection, trend recognition, trend estimation, trend verification, breathing detection, breathing recognition, breathing estimation, human biometric detection, human biometric recognition, human biometric estimation, human biometric verification,Environmental informatics detection, environmental informatics recognition, environmental informatics estimation, environmental informatics verification, gait detection, gait recognition, gait estimation, gait verification, gesture detection, gesture recognition, gesture estimation, gesture verification, machine learning, supervised learning, unsupervised learning, semi-supervised learning, clustering, feature extraction, feature training, principal component analysis, eigenvalue decomposition, frequency decomposition, time decomposition, Time-Frequency Decomposition, Function Decomposition, Other Decompositions, Training, Discriminative Training, Supervised Training, Unsupervised Training, Semi-Supervised Training, Neural Networks, Sudden Motion Detection, Fall Detection, Hazard Detection, Life-Threat Detection, Regular Motion Detection, Stationary Motion Detection, Cyclostationary Motion Detection, Intrusion Detection, Suspicious Motion Detection, Security, Safety Monitoring, Navigation, Guidance, Map-Based Processing, Map-Based Correction, Model-Based Processing / Correction, Irregularity Detection, Localization, Room Sensing, Tracking, Multiple Object Tracking, Indoor Tracking, Indoor Positioning, Indoor Navigation, Energy Management, Power Transmission, Wireless Power Transmission, Object Counting, Car Tracking in Parking Garages, Device / System Activation (e.g., Security Systems, Access Systems, Alarms, Sirens, Speakers, Televisions, Entertainment Systems, Cameras, Heating / Air Conditioning (HVAC) Systems, Ventilation Systems, Lighting Systems, Gaming Systems, Coffee Machines, Cooking Appliances, Cleaning Equipment, Housekeeping Equipment), Geometric Estimation The tasks may include communication, augmented reality, wireless communication, data communication, signal broadcasting, networking, coordination, management, encryption, protection, cloud computing, other processing, and / or other tasks. The tasks may be performed by a Type 1 device, a Type 2 device, another Type 1 device, another Type 2 device, a nearby device, a local server (such as a hub device), an edge server, a cloud server, and / or another device. The tasks may be based on TSCI between any pair of Type 1 and Type 2 devices. A Type 2 device may also be a Type 1 device, and vice versa. A Type 2 device may fulfill the role (e.g., functionality) of a Type 1 device temporarily, continuously, sporadically, simultaneously, and / or concurrently, and / or vice versa. A first portion of the tasks may include pre-processing, processing, signal conditioning, signal processing, post-processing,may include at least one of sporadic / continuous / concurrent / simultaneous / dynamic / adaptive / on-demand / as-needed processing, calibration, noise removal, feature extraction, coding, encryption, transformation, mapping, motion detection, motion estimation, motion change detection, motion pattern detection, motion pattern estimation, motion pattern recognition, vital sign detection, vital sign estimation, vital sign recognition, periodic motion detection, periodic motion estimation, repetitive motion detection / estimation, breathing rate detection, breathing rate estimation, breathing pattern detection, breathing pattern estimation, breathing pattern recognition, heart rate detection, heart rate estimation, cardiac pattern detection, cardiac pattern estimation, cardiac pattern recognition, gesture detection, gesture estimation, gesture recognition, velocity detection, velocity estimation, object location, object tracking, navigation, acceleration estimation, acceleration detection, fall detection, change detection, intruder (and / or tampering) detection, baby detection, baby monitoring, patient monitoring, object recognition, wireless power transmission, and / or wireless charging;

[0164] The second part of the task may be a smart home task, a smart office task, a smart building task, a smart factory task (e.g., manufacturing with a machine or assembly line), a smart Internet of Things (IoT) task, a smart system task, a smart home operation, a smart office operation, a smart building operation, a smart manufacturing operation (e.g., movement of supplies / parts / raw materials to a machine / assembly line), an IoT operation, a smart system operation, turning on lights, turning off lights, controlling light in at least one of a room, area, and / or location, playing a sound clip, playing a sound clip in at least one of a room, area, and / or location, playing at least one sound clip of welcome, greeting, farewell, a first message, and / or a second message related to the first part of the task, turning on an appliance, turning off an appliance, The control may include at least one of controlling appliances in at least one of the rooms, areas, and / or locations; turning on an electrical system; turning off an electrical system; controlling an electrical system in at least one of the rooms, areas, and / or locations; turning on a security system; turning off a security system; controlling a security system in at least one of the rooms, areas, and / or locations; turning on a mechanical system; turning off a mechanical system; controlling a mechanical system in at least one of the rooms, areas, and / or locations; and / or controlling at least one of an air conditioning system, a heating system, a ventilation system, a lighting system, a heater, a stove, an entertainment system, a door, a fence, a window, a garage, a computer system, a networked device, a networked system, a home appliance, an office appliance, a lighting device, a robot (e.g., a robotic arm), a smart vehicle, a smart machine, an assembly line, a smart device, an Internet of Things (IoT) device, a smart home device, and / or a smart office device.

[0165] The tasks are to detect when the user comes home, detect when the user leaves, detect when the user moves from one room to another, detect when a window / door / garage door / blinds / curtains / panels / solar panels / sunshades are controlled / locked / unlocked / opened / closed / partially opened, detect pets, detect / monitor when the user is doing something (e.g. sleeping on the sofa, sleeping in the bedroom, running on the treadmill, cooking, sitting on the sofa, watching TV, eating in the kitchen, eating in the dining room, going up and down stairs, going out / coming back, in the bathroom), monitor / detect the location of the user / pet and do something automatically when detected (e.g. sending a message, notifying / reporting), do something to the user when a user is detected, turn on / off / dim the lights, turn on / off the music / radio / home entertainment system, turn on / off the TV / may include turning on / off / adjusting / controlling hi-fi / set-top box (STB) / home entertainment system / smart speaker / smart device, turning on / off / adjusting air conditioning system, turning on / off / adjusting ventilation system, turning on / off / adjusting heating system, adjusting / controlling curtains / light shades, turning on / off / starting computer, turning on / off / preheating / controlling coffee machine / hot water kettle, turning on / off / preheating / controlling cooker / oven / microwave / other cooking appliance, checking / adjusting temperature, checking weather forecast, checking phone message box, checking email, checking system, controlling / adjusting system, checking / controlling / arming / disarming security system / baby monitor, checking / controlling refrigerator, reporting (e.g. through speaker such as Google Home, Amazon Echo, via web page / email / messaging system / notification system).

[0166] For example, if a user arrives at home in their vehicle, the task may automatically detect that the user or their vehicle is approaching, open the garage door upon detection, turn on the driveway / garage lights as the user approaches the garage, turn on the air conditioner / heater / fan, etc. As the user enters the house, the task may automatically turn on the entrance lights, turn off the driveway / garage lights, play a greeting message to warm up the user, turn on music, turn on the radio and tune it to the user's favorite radio news channel, open the curtains / blinds, monitor the user's mood, adjust the lighting and sound environment according to the user's mood or a current / impending event on the user's daily calendar (e.g., romantic lighting and music because the user is having dinner with his / her girlfriend in an hour), microwave the food the user prepared that morning, perform a diagnostic check of all systems in the house, check the weather forecast for tomorrow's tasks, check news of interest to the user, and update the user's calendar, to-do list, etc. Check reminders, check phone answering systems, messaging systems, email, communicate verbal reports using dialogue systems / speech synthesis, remind the user of their mother's birthday (e.g., using audible tools such as speakers, hi-fi, speech synthesis, sound, voice, music, song, sound field, background sound field, dialogue system, etc.; using visual tools such as TV / entertainment system / computer / notebook / smartpad / display / light / color / brightness / pattern, using tactile tools / virtual reality tools / gestures / tools, using smart devices / appliances / materials / furniture / fixtures, using web tools / servers / hub devices / cloud servers / fog servers / edge servers / home networks / mesh networks, using messaging tools / notification tools / communication tools / scheduling tools / email, using user interfaces / GUIs, using scents / smells / aromas / tastes, using neural tools / neural system tools, or a combination), create reports, provide reports (e.g., using reminding tools as described above).A task may be to proactively start an air conditioner / heater / ventilation system or proactively adjust a smart thermostat temperature setting. When a user moves from the front door to the living room, the tasks may be to turn on the living room lights, open the living room curtains, open the windows, turn off the front door light behind the user, turn on the TV and set-top box, turn on the set-top box, set the TV to the user's preferred channel, adjust the appliances according to the user's preferences and conditions / states (e.g., adjust the lighting, select / play music to create a romantic atmosphere), etc.

[0167] Another example could be: When a user wakes up in the morning, the task could be to detect the user moving around in the bedroom, open the blinds / curtains, open the windows, turn off the alarm clock, adjust the room temperature profile from a night temperature profile to a day temperature profile, turn on the bedroom lights, turn on the toilet light as the user approaches the bathroom, check the radio or streaming channels, play the morning news, turn on the coffee machine, preheat water, turn off the security system, etc. When the user walks from the bedroom to the kitchen, the task could be to turn on the kitchen and hallway lights, turn off the bedroom and toilet lights, move music / messages / reminders from the bedroom to the kitchen, turn on the kitchen TV, change the TV to the morning news channel, lower the kitchen blinds, open the kitchen window to let in fresh air, unlock the back door so the user can check the back yard, adjust the kitchen temperature setting, etc. Another example could be: When the user leaves home for work, the tasks may be to detect the user's departure, say goodbye and / or have a nice day, open / close the garage door, turn on / off the garage and driveway lights, turn off / dim to save energy (only if the user fails), close / lock all windows / doors (only if the user fails), turn off appliances (especially the stove, oven, microwave), turn on / arm the home security system to protect the home against intruders, adjust the air conditioning / heating / ventilation system to an "away from home" profile to save energy, send alerts / reports / updates to the user's smartphone, etc.

[0168] The motions are classified as no motion, rest motion, motionless motion, motion, change of location / position, deterministic motion, transient motion, falling motion, repetitive motion, periodic motion, pseudo-periodic motion, periodic / repetitive motion related to breathing, periodic / repetitive motion related to heartbeat, periodic / repetitive motion related to living organisms, periodic / repetitive motion related to machines, periodic / repetitive motion related to man-made objects, periodic / repetitive motion related to nature, complex motion related to transient and periodic elements, repetitive motion, non-deterministic motion, stochastic motion, chaotic motion, random motion, complex motion with non-deterministic and deterministic elements, stationary random motion, pseudo-stationary random motion, cyclostationary random motion, non-stationary random motion, non-stationary random motion with a periodic autocorrelation function (ACF), random motion with a periodic ACF over time. motion, pseudo-stationary random motion over time, random motion where the instantaneous ACF has a pseudo-periodic / repetitive component over time, machine motion, mechanical motion, vehicle motion, drone motion, air-related motion, wind-related motion, weather-related motion, water-related motion, fluid-related motion, ground-related motion, change in magnetic properties, subsurface motion, earthquake motion, plant motion, animal motion, animal motion, human motion, normal motion, abnormal motion, dangerous motion, warning motion, suspicious motion, rain, fire, flood, tsunami, explosion, collision, imminent collision, human motion, head motion, face motion, eye motion, oral motion, tongue motion, neck motion, finger motion, hand motion, arm motion, shoulder motion, body motion, chest motion, abdominal motion, hip motion, leg motion, foot motion, body joint motion, knee motion, elbow motion, upper body motion, lower body motion, skin motion, subcutaneous motion, subcutaneous tissue motion. The motion may include at least one of blood vessel movement, venous movement, organ movement, heart movement, lung movement, stomach movement, intestinal movement, bowel movement, eating movement, breathing movement, facial expression, eye expression, mouth expression, vocal movement, singing movement, eating movement, gesture, hand gesture, arm gesture, keystroke, typing stroke, user interface gesture, man-machine interaction, gait, dance movement, coordinated movement, and / or coordinated body movement.

[0169] The heterogeneous ICs of the Type 1 device and / or any Type 2 receiver may include a low noise amplifier (LNA), a power amplifier, a transmit-receive switch, a media access controller, a baseband radio, a 2.4 GHz radio, a 3.65 GHz radio, a 4.9 GHz radio, a 5 GHz radio, a 5.9 GHz radio, a sub-6 GHz radio, a 60 GHz radio, a sub-60 GHz radio, and / or another radio. The heterogeneous IC may include a processor, a memory communicatively coupled to the processor, and a set of instructions stored in the memory for execution by the processor. The IC and / or any processor may include at least one of a general purpose processor, a special purpose processor, a microprocessor, a multiprocessor, a multi-core processor, a parallel processor, a CISC processor, a RISC processor, a microcontroller, a central processing unit (CPU), a graphical processor unit (GPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), an embedded processor (e.g., ARM), a logic circuit, another programmable logic device, discrete logic, and / or a combination. Heterogeneous ICs are used in broadband networks, wireless networks, mobile networks, mesh networks, cellular networks, wireless local area networks (WLANs), wide area networks (WANs), metropolitan area networks (MANs), WLAN standards, WiFi, LTE, LTE-A, LTE-U, 802.11 standards, 802.11a, 802.11b, 802.11g, 802.11n, 802.11ac, 802.11af, 802.11ah, 802.11ax, 802.11ay, mesh networking standards 802.16, 3G, 3.5G, 4G, Beyond 4G, 4.5G, 6G, 7G, 8G, 9G, UMTS, 3GPP, GSM, EDGE, TDMA, FDMA, CDMA, WCDMA, TD-SCDMA, Bluetooth, Bluetooth Low-Energy It may support Bluetooth Low Energy (BLE), NFC, Zigbee, WiMax, and other wireless network protocols.

[0170] The processor may include a general-purpose processor, a special-purpose processor, a microprocessor, a microcontroller, an embedded processor, a digital signal processor, a central processing unit, a graphical processing unit (GPU), a multiprocessor, a multi-core processor, and / or a processor with graphics capabilities, and / or a combination thereof. The memory may be volatile, non-volatile, random-access memory (RAM), read-only memory (ROM), programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), a hard disk, flash memory, CD-ROM, DVD-ROM, magnetic storage, optical storage, organic storage, a storage system, a storage network, network storage, cloud storage, edge storage, local storage, external storage, internal storage, or any other form of non-transitory storage medium known in the art. The set of instructions (machine-executable code) corresponding to the method steps may be directly embodied in hardware, software, firmware, or a combination thereof. The set of instructions may be embedded, pre-loaded, loaded at boot time, loaded on-the-fly, loaded on-demand, pre-installed, installed, and / or downloaded.

[0171] The presentation may be visual (e.g., using a combination of sights, graphics, text, symbols, color, shade, video, animation, sound, voice, audio, etc.), graphical (e.g., using GUI, animation, video), textual (e.g., web page with text, messages, animated text), symbolic (e.g., emojis, signs, hand gestures), or mechanical (e.g., vibration, actuator movement, haptics, etc.) presentation.

[0172] The computational workload associated with this method is shared among the processor, the type 1 heterogeneous wireless device, the type 2 heterogeneous wireless device, the local server (eg, the hub device), the cloud server, and other processors.

[0173] Operations, pre-processing, processing, and / or post-processing may be applied to the data (e.g., TSCI, autocorrelation, TSCI features). Operations may include pre-processing, processing, and / or post-processing. Pre-processing, processing, and / or post-processing may be operations. Operations may include pre-processing, processing, post-processing, scaling, calculating confidence coefficients, calculating line-of-sight (LOS) quantities, non-LOS calculations, and the like. Computing non-linear (NLOS) quantities, computing quantities including line-of-sight and non-linear (NLOS), computing quantities for a single link (e.g., path, communication path, link between a transmitting antenna and a receiving antenna), computing quantities including multiple links, computing functions of operands, filtering, linear filtering, nonlinear filtering, folding, grouping, energy computation, low-pass filtering, band-pass filtering, high-pass filtering, median filtering, rank filtering, quartile filtering, percentile filtering, finite impulse response (FIR) filtering, infinite impulse response (IIR) filtering, moving average (MA) filtering, autoregressive (AR) filtering, autoregressive moving average (ARMA) filtering, selective filtering, adaptive filtering, interpolation, decimation, subsampling, upsampling, resampling, time correction, time-based correction, phase correction, magnitude correction, phase cleaning, amplitude cleaning, matched filtering, enhancement, restoration, noise removal, smoothing, signal conditioning, enhancement, restoration, linear transform, nonlinear transform, inverse transform, frequency transformation, inverse frequency transformation, Fourier transform (FT), discrete-time FT (DTFT), Discrete FT (DFT), Fast FT (FFT), Wavelet Transform, Laplace Transform, Hilbert Transform, Hadamard Transform, Trigonometric Transform, Sin Transform, Cosine Transform, DCT, Power of 2 Transform, Sparse Transform, Graph Based Transform, Graph Signal Processing, Fast Transform, Transform Combined with Zero Padding, Cyclic Padding, Padding, Zero Padding, Feature Extraction, Decomposition, Projection, Orthogonal Projection, Non-Orthogonal Projection, Overprojection (oecomlee)ojecion), eigendecomposition, singular value decomposition (SVD), principle component analysis (ICA), independent component analysis (ICA), grouping, sorting, thresholding, soft thresholding, hard thresholding, clipping, soft clipping, first derivative, second derivative, higher derivative, convolution, multiplication, division, addition, subtraction, integration, maximization, minimization, least squared error, recursive least squares, constrained least squares, batch least squares, least absolute deviation, least mean squared deviation, least absolute deviation, local maximization, local minimization, cost function optimization, neural networks, recognition, labeling, training, class filtering, machine learning, supervised learning, unsupervised learning, semi-supervised learning, comparison with other TSCI, similarity score calculation, quantization, vector quantization, matching pursuit, compression, encryption, encoding, storage, transmission, normalization, time normalization, frequency domain normalization, classification, clustering, labeling, tagging, learning, detection, estimation, learning network, mapping, remapping, expansion, storage, search, transmission, reception, representation, combining, merging, splitting, tracking, monitoring, matched filtering, Kalman filtering, particle filtering, interpolation, extrapolation, histogram estimation The operations may include importance sampling, Monte Carlo sampling, compressed sensing, representation, merging, combining, dividing, scrambling, error protection, forward error correction, do nothing, time-varying processing, adjusted averaging, weighted averaging, arithmetic mean, geometric mean, harmonic mean, averaging over selected frequencies, averaging over antenna links, logical operations, permutation, combination, sorting, AND, OR, XOR, union, intersection, vector addition, vector subtraction, vector multiplication, vector division, inverse, norm, distance, and / or another operation. The operations may be pre-processing, processing, and / or post-processing. Operations may be applied jointly on multiple time series or functions.

[0174] Functions (e.g., functions of operands) can be scalar functions, vector functions, discrete functions, continuous functions, polynomial functions, properties, characteristics, magnitude, phase, exponential functions, logarithmic functions, trigonometric functions, transcendental functions, boolean functions, linear functions, algebraic functions, nonlinear functions, piecewise functions, real functions, complex functions, vector-valued functions, inverse functions, derivative functions, integral functions, circular functions, functions of other functions, one-to-one functions, one-to-many functions, many-to-one functions, many-to-many functions, zero crossings, absolute functions, index functions, mean, mode, median, range, statistics, histogram, variance, standard deviation, measure of change, expansion, dispersion, deviation, divergence, range, interquartile range, total deviation, absolute deviation, arithmetic mean, geometric mean, harmonic mean, trimmed mean, percentile, square, cube, square root, power, sine, cosine, tangent, cotangent ent, elliptic functions, parabolic functions, hyperbolic functions, game functions, zeta functions, absolute value, threshold, limit functions, floor functions, rounding functions, sign functions, quantization, piecewise constant functions, composite functions, functions of functions, time functions processed by operations (e.g. filtering), probabilistic functions, stochastic functions, deterministic functions, periodic functions, iterative functions, transformations, frequency transforms, inverse frequency transforms, discrete time transforms, Laplace transforms, Hilbert transforms, sine transforms, cosine transforms, trigonometric transforms, wavelet transforms, integer transforms, power of two transforms, sparse transforms, projections, decompositions, principal component analysis (PCA), neural networks, feature extraction, moving functions, functions for moving windows of adjacent items in a time series, filtering functions, convolutions, mean functions, histograms, variance / standard deviation functions, statistical functions, short time transforms, discrete transforms, Discrete Fourier Transform, Discrete Cosine Transform, Discrete Sine Transform, Hadamard Transform, Eigenvalue Decomposition, Eigenvalue, Singular Value Decomposition (SVD), Singular Value, Orthogonal Decomposition, Matching Pursuit, Sparse Transform, Arbitrary Decomposition, Graph Based Processing, Graph Based Transform, Graph Signal Processing, Classification, Class / Group / Category Identification, Labeling, Learning, Machine Learning, Detection, Estimation, Feature Extraction, Learning Networks, Feature Extraction, Noise Reduction, Signal Enhancement, Coding, Encryption, Mapping, Remapping, Vector Quantization, Low-Pass Filtering, High-Pass Filtering, Band-Pass Filtering, Matched Filtering, Kalman Filtering, Pre-Processing, Post-Processing, Particle Filtering, FIR Filtering, IIR Filtering, Autoregressive (AR) Filtering, Adaptive Filtering, First Derivative, Higher-Order Derivative, Integration, Zero Crossing, Smoothing, Median Filtering, Mode Filtering, Sampling, Random Sampling, Resampling Function, Downsampling, Downconverting, Upsampling, Upconverting, Interpolation, Extrapolation, Importance Sampling, Monte Carlo Sampling, Compressed Sensing, Statistics, These may include short-term statistics, long-term statistics, autocorrelation functions, cross-correlation functions, moment-generating functions, time averages, weighted averages, special functions, Bessel functions, error functions, complementary error functions, beta functions, gamma functions, integral functions, Gaussian functions, Poisson functions, etc. Machine learning, training, discriminative training, deep learning, neural networks, continuous-time processing, distributed computing, distributed storage, acceleration using GPUs / DSPs / coprocessors / multi-cores / multi-processing may be applied to the steps (or each step) of the present disclosure.

[0175] The frequency transform may include a Fourier transform, a Laplace transform, a Hadamard transform, a Hilbert transform, a sine transform, a cosine transform, a trigonometric transform, a wavelet transform, an integer transform, a power of two transform, zero-padding and combinations of transforms, a power Fourier transform with zero-padding, and / or another transform. Fast and / or approximate versions of the transforms may be performed. The transforms may be performed using floating-point and / or fixed-point arithmetic.

[0176] The inverse frequency transform may include an inverse Fourier transform, an inverse Laplace transform, an inverse Hadamard transform, an inverse Hilbert transform, an inverse sine transform, an inverse cosine transform, an inverse trigonometric transform, an inverse wavelet transform, an inverse integer transform, an inverse power of two transform, a combination of zero padding and transforms, an inverse Fourier transform with zero padding, and / or another transform. Fast and / or approximate versions of the transform may be performed. The transform may be performed using floating-point and / or fixed-point arithmetic.

[0177] Quantities / features from TSCI can be calculated. Quantities include: movement, location, map coordinates, height, speed, acceleration, movement angle, rotation, dimension, volume, time trend, one-time pattern, recurring pattern, evolving pattern, time pattern, mutually exclusive pattern, associated / correlated pattern, cause-effect, short-term / long-term correlation, tendency, slope, preference, statistics, typical behavior, atypical behavior, time trend, time profile, periodic movement, repetitive movement, repetition, tendency, change, sudden change, gradual change, frequency, transient, breathing, gait, behavior, event, suspicious event, dangerous event, warning event, warning, belief, proximity, collision, power, signal, signal power, signal strength, signal quantity, received signal strength indicator (RSSI), signal amplitude, signal phase, signal frequency component, signal frequency band component, channel state information (CSI), map, time, frequency, time-frequency, The statistics may include at least one of decomposition, orthogonal decomposition, non-orthogonal decomposition, tracking, respiration, palpitations, statistical parameters, cardiopulmonary statistics / analysis (e.g., output response), daily activity statistics / analysis, chronic disease statistics / analysis, medical statistics / analysis, early (or instantaneous or simultaneous or delayed) indicator / suggestion / sign / marker / verifier / detection / symptom / condition / state, biometrics, baby, patient, machine, device, temperature, vehicle, parking lot, location, lift, elevator, space, road, fluid flow, home, room, office, house, building, warehouse, storage, system, ventilation, fan, pipe, duct, people, human, car, boat, truck, plane, drone, downtown, crowd, impulse event, cyclostationary, environment, vibration, material, surface, 3D, 2D, local, global, presence, and / or other measurable quantity / variable.

[0178] The sliding time window may have a time-varying window width. It may be smaller initially to allow for rapid acquisition and may increase over time to a steady-state size. The steady-state size may be related to the monitored frequency, repetitive motion, transient motion, and / or STI. Even in the steady state, the window size may be adaptively (and / or dynamically) changed (e.g., adjusted, varied, modified) based on battery life, power consumption, available computing power, changes in the volume of interest, the nature of the monitored motion, etc.

[0179] The time shift between two sliding time windows at adjacent time instances can be constant / variable / locally adaptive / dynamically adjusted over time. If a shorter time shift is used, any monitoring updates can be more frequent, which can be used for rapidly changing conditions, object movements, and / or objects. A longer time shift can be used for slower conditions, object movements, and / or objects. The window width / size and / or time shift can be changed (e.g., adjusted, changed, modified) according to user requests / selections. The time shift can be changed automatically (e.g., as controlled by a processor / computer / server / hub device / cloud server) and / or adaptively (and / or dynamically).

[0180] At least one characteristic (e.g., a characteristic value or characteristic point) of a function (e.g., an autocorrelation function, an autocovariance function, a cross-correlation function, a cross-covariance function, a power spectral density, a time function, a frequency domain function, a frequency transform) may be determined (e.g., by the object tracking server, a processor, a Type 1 heterogeneous device, a Type 2 heterogeneous device, and / or another device). At least one characteristic of the function may include a maximum, a minimum, an extremum, a local maximum, a local minimum, a local extremum, a local extremum with a positive time offset, a first local extremum with a positive time offset, an nth local extremum with a positive time offset, a first local extremum with a negative time offset, a bounded maximum, a bounded minimum, a bounded extremum, a significant maximum, a significant minimum, a significant extremum, a gradient, a derivative, a higher order derivative, a maximum gradient, a minimum gradient, a local maximum gradient, a local maximum gradient with a positive time offset, a local minimum gradient, a bounded maximum gradient, a bounded minimum gradient, a maximum higher order derivative, a minimum higher order derivative, a bounded higher order derivative, a zero crossing, a zero crossing with a positive time offset, an nth zero crossing with a positive time offset, a zero crossing with a negative time offset, an nth zero crossing with a negative time offset, a bounded zero crossing, a zero crossing of a gradient, a zero crossing of a gradient of a higher order derivative, and / or other characteristics. At least one argument of the function associated with at least one characteristic of the function may be identified. A quantity (eg, spatial-temporal information of an object) may be determined based on at least one argument of the function.

[0181] Characteristics (e.g., characteristics of an object's movement at a location) can include instantaneous characteristics, short-term characteristics, recurring characteristics, recursive characteristics, history, incremental characteristics, change characteristics, deviation characteristics, phase, amplitude, degree, time characteristics, frequency characteristics, time-frequency characteristics, decomposition characteristics, orthogonal decomposition characteristics, non-orthogonal decomposition characteristics, deterministic characteristics, probability characteristics, stochastic characteristics, autocorrelation function (ACF), mean, variance, standard deviation, measure of change, spread, variance, deviation, divergence, range, interquartile range, total variation, absolute deviation, total deviation, statistics, duration, timing, trend, periodic characteristics, recurring characteristics, long-term characteristics, historical characteristics, average characteristics, recent characteristics, past characteristics, future characteristics, predicted characteristics, position, distance, height, speed, direction, velocity, acceleration, change in acceleration, angle, angular speed, change in angular velocity, of an object The change in the angular acceleration includes at least one of: angular acceleration, change in angular acceleration, orientation of the object, angle of rotation, deformation of the object, shape of the object, change in shape of the object, change in size of the object, change in structure of the object, and / or change in properties of the object.

[0182] At least one local maximum and at least one local minimum of the function may be identified. At least one local signal-to-noise ratio-like (SNR-like) parameter may be calculated for each pair of adjacent local maximums and minima. The SNR-like parameter may be a function (e.g., linear, logarithmic, exponential, monotonic) of the fraction of the amount (e.g., power, magnitude) of the local maximum over the same amount of the local minimum. It may also be a function of the difference between the amount of the local maximum and the same amount of the local minimum. Significant local peaks may be identified or selected. Each significant local peak may be a local maximum with an SNR-like parameter greater than a threshold T1 and / or a local maximum with an amplitude greater than a threshold T2. At least one local minimum and at least one local minimum in the frequency domain may be identified / calculated using a persistence-based approach.

[0183] A set of selected significant local peaks may be selected from the set of identified significant local peaks based on a selection criterion (e.g., quality criteria, signal quality state). An object characteristic / STI may be calculated based on the set of selected significant local peaks and frequency values ​​associated with the set of selected significant local peaks. In one example, the selection criterion may always correspond to selecting the strongest peak in the range. The strongest peak may be selected, but non-selected peaks may still be significant (or even strong).

[0184] Unselected significant peaks may be saved and / or monitored as "reserved" peaks for use in future selections in future sliding time windows. As an example, there may be a particular peak (at a particular frequency) that appears consistently over time. Initially, even though it is significant, it may not be selected (because other peaks may become stronger). However, at a later time, the peak may become stronger and more dominant and may be selected. If it is "selected," it may be back-traced in time and deemed "selected" at an earlier time that was significant but not selected. In such a case, the back-traced peak may replace a previously selected peak at an earlier time. The replaced peak may be a relatively weak peak or a peak that appears isolated in time (i.e., appears for only a short time in time).

[0185] In other examples, the selection criteria may not correspond to selecting the strongest peak in the range. Instead, it may consider not only the "strength" of the peak, but also the "trace" of peaks that may have occurred in the past, especially peaks that have been identified over time. For example, if a finite state machine (FSM) is used, it may select peaks based on the state of the FSM. The decision threshold may be adaptively (and / or dynamically) calculated based on the state of the FSM.

[0186] The similarity score and / or component similarity score may be calculated (e.g., by a server (e.g., a hub device), a processor, a Type 1 device, a Type 2 device, a local server, a cloud server, and / or another device) based on a pair of temporally adjacent CIs of a TSCI. The pairs may be obtained from the same sliding window or two different sliding windows. The similarity score may also be based on a pair of temporally adjacent or less-adjacent CIs from two different TSCIs. The similarity score and / or component similarity score may be / include time reversal resonating strength (TRRS), correlation, cross-correlation, autocorrelation, correlation indicator, covariance, cross-covariance, autocovariance, dot product of two vectors, distance score, norm, metric, quality metric, signal quality condition, statistical property, discrimination score, neural network, deep learning network, machine learning, training, discrimination, weighted average, preprocessing, denoising, signal conditioning, filtering, time correction, time alignment, phase offset compensation, transform, component-wise operation, feature extraction, finite state machine, and / or another score. Property and / or STI may be determined / calculated based on the similarity score.

[0187] Any threshold may be predetermined, adaptively (and / or dynamically) determined, and / or determined by a finite state machine. Adaptive determination may be based on time, space, location, antenna, path, link, condition, battery life, remaining battery capacity, available power, available computational resources, available network bandwidth, etc.

[0188] A threshold applied to a test statistic to distinguish between two events (or two conditions, or two situations, or two states), A and B, may be determined. Data (e.g., CI, channel state information (CSI), power parameters) may be collected under A and / or under B in a training scenario. A trial statistic may be calculated based on the data. A distribution of the trial statistic under A may be compared to a distribution of the trial statistic under B (a reference distribution), and a threshold may be selected according to some criteria. The criteria may include maximum likelihood estimation (ML), maximum a posteriori probability (MAP), discriminative training, minimum type-1 error for a given type-2 error, minimum type-2 error for a given type-1 error, and / or other criteria (e.g., quality criteria, signal quality conditions). The threshold may be adjusted to achieve different sensitivities to A, B, and / or other events / conditions / situations / states. The threshold adjustment may be automatic, semi-automatic, and / or manual. Threshold adjustments may be applied once, occasionally, frequently, periodically, repeatedly, occasionally, sporadically, and / or on-demand. Threshold adjustments may be adaptive (and / or dynamically adjusted). Threshold adjustments may depend on objects, object movement / location / orientation / motion, object characteristics / STI / size / characteristics / traits / habits / behavior, location, at / at / of location, features / fixtures / furniture / barriers / materials / machines / creatures / objects / boundaries / surfaces / media, map, map constraints (or environmental model), events / states / scenes / conditions, time, timing, duration, current state, past history, user, and / or personal preference, etc.

[0189] The stopping criterion (or skip or bypass or blocking or pausing or passing or rejecting criterion) of an iterative algorithm may be that the change in the current parameter (e.g., offset value) in the update in the iteration is less than a threshold. The threshold may be 0.5, 1, 1.5, 2, or another number. The threshold may be adaptive (and / or dynamically adjusted). It may change as the iteration progresses. With respect to the offset value, the adaptive threshold may be determined based on the task, the initial specific value, the current time offset value, the regression window, the regression analysis, the regression function, the regression error, the convexity of the regression function, and / or the number of iterations.

[0190] The local extrema may be determined as a corresponding extremum of the regression function in the regression window. The local extrema may be determined based on a set of time offset values ​​and a set of associated regression function values ​​within the regression window. Each of the set of associated regression function values ​​associated with the set of time offset values ​​may be within a range from the corresponding extremum of the regression function in the regression window.

[0191] Searching for local extrema, robust search, minimization, optimization, statistical optimization, dual optimization, constraint optimization, convex optimization, global optimization, local optimization, energy minimization, linear regression, quadratic regression, higher order regression, linear programming, nonlinear programming, stochastic programming, combinatorial optimization, constraint programming, constraint satisfaction, computation of variations, optimal control, dynamic programming, mathematical programming, multiobjective optimization, multimodal optimization, disjunctive programming, space mapping, infinite dimensional optimization, heuristics, metaheuristics, convex programming, semidefinite programming, cone programming, second order cone programming, integer programming, quadratic programming, fractional programming, numerical analysis, simplex algorithm, iterative methods, gradient descent, subgradient methods, coordinate gradient methods, conjugate gradient methods, Newton's algorithm, sequential quadratic programming, interior point methods, elliptic methods, reduced gradient methods, quasi-Newton methods, simultaneous perturbation stochastic approximation, interpolation, pattern search methods, line search, non-differential optimization, genetic algorithms, evolutionary algorithms, dynamic relaxation The search for local extrema may involve an objective function, a loss function, a cost function, a utility function, a fitness function, an energy function, and / or an energy function.

[0192] The regression may be performed using a regression function to fit the sampled data (e.g., CIs, CI features, components of CIs) or another function (e.g., an autocorrelation function) in a regression window. The length of the regression window and / or the position of the regression window may be varied in at least one iteration. The regression function may be a linear function, a quadratic function, a cubic function, a polynomial function, and / or another function. The regression analysis may minimize at least one of the following: error, aggregate error, component error, error in a projected domain, error in a selected axis, error in a selected orthogonal axis, absolute error, squared error, absolute deviation, squared deviation, higher-order error (e.g., third-order, fourth-order), robust error (e.g., squared error versus absolute error for smaller magnitude errors and for larger errors, or a first type of error for smaller magnitude errors and a second type of error for larger magnitude errors), another error, a weighted sum (or weighted average) of absolute / squared errors (e.g., in the case of a wireless transmitter with multiple antennas and a wireless receiver with multiple antennas, each pair of transmitter antenna and receiver antenna forms a link), mean absolute error, mean squared error, mean absolute deviation, and / or mean squared deviation. Errors associated with different links may have different weights. One possibility is that some links and / or some components with greater noise or lower signal quality metrics may have smaller or larger weights. (weighted sum of squared errors, weighted sum of higher-order errors, weighted sum of robust errors, weighted sum of further errors, absolute cost, squared cost, higher-order cost, robust cost, further cost, weighted sum of absolute costs, weighted sum of squared costs, weighted sum of higher-order costs, weighted sum of robust costs, and / or weighted sum of further costs). The determined regression error can be an absolute error, a squared error, a higher-order error, a robust error, a further error, a weighted sum of absolute errors, a weighted sum of squared errors, a weighted sum of higher-order errors, a weighted sum of robust errors, and / or a weighted sum of further errors.

[0193] The time offset associated with the maximum regression error (or minimum regression error) of the regression function for a particular function within the regression window may be the updated current time offset for the iteration.

[0194] The local extrema can be searched for based on a quantity including the difference between two different errors (e.g., the difference between an absolute error and a squared error), each of which can include an absolute error, a squared error, a higher-order error, a robust error, another error, a weighted sum of absolute errors, a weighted sum of squared errors, a weighted sum of higher-order errors, a weighted sum of robust errors, and / or a weighted sum of another error.

[0195] The quantity may be compared to reference data or a reference distribution, such as an F-distribution, a central F-distribution, another statistical distribution, a threshold, a threshold associated with a probability / histogram, a threshold associated with a probability / histogram of finding a false peak, a threshold associated with an F-distribution, a threshold associated with a central F-distribution, and / or a threshold associated with another statistical distribution.

[0196] The regression window may be determined based on at least one of: a movement of the object (e.g., a change in location / position), a quantity related to the object, at least one characteristic and / or STI of the object related to the movement of the object, an estimated location of a local extremum, noise characteristics, estimated noise characteristics, a signal quality metric, an F-distribution, a central F-distribution, another statistical distribution, a threshold, a preset threshold, a threshold related to a probability / histogram, a threshold related to a desired probability, a threshold related to the probability of finding a false peak, a threshold related to an F-distribution, a threshold related to a central F-distribution, a threshold related to another statistical distribution, a condition that the quantity at the window center is a maximum within the regression window, a condition that only one of the local extrema of a particular function for a particular value exists for the first time within the regression window, another regression window, and / or other conditions.

[0197] The width of the regression window can be determined based on the particular local extrema to be searched for, including a first local maximum, a second local maximum, a higher-order local maximum, a first local maximum with a positive time offset, a second local maximum with a positive time offset, a higher-order local maximum with a positive time offset, a first local maximum with a negative time offset, a second local maximum with a negative time offset, a second local maximum with a negative time offset, a higher-order local maximum with a negative time offset, a first local minimum, a second local minimum, a higher-order local minimum, a first local minimum with a positive time offset, The local minimum may include a second local minimum with a positive time offset, a higher-order local minimum with a positive time offset, a first local minimum with a negative time offset, a second local minimum with a negative time offset, a higher-order local minimum with a negative time offset, a first local extremum, a second local extremum, a higher-order local extremum, a first local extremum with a positive time offset, a second local extremum with a positive time offset, a first local extremum with a negative time offset, a second local extremum with a negative time offset, and a higher-order extremum with a negative and / or negative time offset.

[0198] The current parameters (e.g., time offset values) may be initialized based on a target value, a target profile, a trend, a past trend, a current trend, a target velocity, a velocity profile, a target velocity profile, a past velocity trend, an object's motion or movement (e.g., a change in location / position), at least one characteristic and / or STI of the object associated with the object's motion, a position quantity of the object, an initial velocity of the object associated with the object's motion, a predefined value, an initial width and duration of the regression window, a value based on the signal's carrier frequency, a value based on the signal's subcarrier frequency, a signal's bandwidth, an antenna's aggregate value associated with the channel, noise characteristics, a signal h metric, and / or an adaptive (and / or dynamically adjusted) value. The current time offset may be at the center, left, right, and / or another fixed relative position of the regression window.

[0199] In the presentation, information may be displayed along with a map (or environmental model) of the location. Information may include location, zone, area, region, coverage area, corrected location, approximate location, location wrt a map of the location, location wrt a segmented location, direction, route, route wrt a map and / or segmentation, trace (e.g., location within a time window such as the last 5 seconds, or the last 10 seconds, where the time window duration may be adaptively (and / or dynamically) adjusted, and the time window duration may be adaptively (and / or dynamically) adjusted for speed, acceleration), route history, approximate regions / zones along the route, history / summary of past locations, history of past locations of interest, frequently visited areas, customer traffic, herd distribution, herd behavior, herd control information, speed, acceleration, movement statistics, respiration rate, heart rate, presence / absence of movement, Presence or absence of a person, pet, or object, presence or absence of vital signs, gesture control (controlling a device using gestures), position-based gesture control, information on location-based operations, identity (ID) or identifier of the object of interest (e.g., pet, person, self-guided machine / device, vehicle, drone, car, boat, bicycle, unmanned car, machine with fan, air conditioner, TV, machine with moving parts), user identification (e.g., person), user information, position / speed / acceleration / direction / movement / gesture / gesture control / movement trace, user ID or identifier, user activity, user state, user sleep / rest characteristics, user The location information may include: an emotional state of a user, vital signs of the user, environmental information of the location, weather information of the location, earthquake, explosion, storm, rain, fire, temperature, collision, impact, vibration, events, door opening events, door slamming events, window opening events, window slamming events, fall events, combustion events, freezing events, water-related events, wind-related events, air movement events, accident events, quasi-periodic events (e.g., running on a treadmill, hopping, skipping rope, somersaults, etc.), repetitive events, swarming events, vehicle events, user gestures (e.g., hand gestures, arm gestures, foot gestures, leg gestures, body gestures, head gestures, face gestures, mouth gestures, eye gestures, etc.). The location may be two-dimensional (e.g., using 2D coordinates), three-dimensional (e.g., using 3D coordinates).Location may be relative (e.g., with respect to a map or environmental model) or relational (e.g., halfway between point A and point B, around the corner, upstairs, on a table, on the ceiling, on the floor, on the couch, close to point A, a distance R from point A, within a radius of R from point A, etc.). Location may be expressed in Cartesian coordinates, polar coordinates, and / or another representation.

[0200] Information (e.g., location) may be marked with at least one symbol. The symbol may change over time. The symbol may flash and / or pulsate with or without changing color / intensity. The size may change over time. The orientation of the symbol may change over time. The symbol may be a number reflecting an instantaneous quantity (e.g., user's vital signs / respiratory rate / heart rate / gesture / status / condition / action / movement, temperature, network traffic, network connectivity, device / machine status, remaining device power, device state, etc.). The rate of change, size, orientation, color, intensity, and / or symbol may reflect the respective movement. Information may be presented visually and / or verbally explained (e.g., using pre-recorded audio or speech synthesis). Information may be written in text. Information may also be presented in a mechanical manner (e.g., animated gadgets, moving parts).

[0201] The user interface (UI) device may be a smartphone (e.g., iPhone, Android phone), a tablet (e.g., iPad), a laptop (e.g., notebook computer), a personal computer (PC), a device with a graphic user interface (GUI), a smart speaker, a device with voice / sound / speaker capabilities, a virtual reality (VR) device, an augmented reality (AR) device, a smart car, an in-car display, a voice assistant, an in-car voice assistant, etc. The map (or environmental model) may be two-dimensional, three-dimensional, and / or higher dimensional (e.g., a time-varying 2D / 3D map / environment model). Walls, windows, doors, entrances, exits, and restricted areas may be marked on the map or model. The map may include a facility floor plan. The map or model may have one or more layers (overlays). The map / model may be a maintenance map / model including water pipes, gas pipes, wiring, cable runs, air ducts, crawl spaces, ceiling layouts, and / or underground layouts. A location can be segmented / subdivided / regionalized / grouped into multiple zones / regions / geographical areas / sectors / sections / territories / districts / administrative districts / sites / neighborhoods / areas / stretches / open spaces, such as bedrooms, living rooms, storage rooms, walkways, kitchens, dining rooms, foyers, garages, first floors, second floors, restrooms, offices, conference rooms, reception areas, various office areas, various warehouse areas, various facility areas, etc. The segments / regions / regions can be presented on a map / model. Different regions may be color-coded. Different regions may be presented with characteristics (e.g., color, brightness, color intensity, texture, animation, blinking, blink rate, etc.). The logical segmentation of a location can be performed using at least one heterogeneous Type-2 device, or server (e.g., hub device), or cloud server, etc.

[0202] Here is an example of the disclosed system, device, and method. Stephan and his family want to install the disclosed wireless motion detection system to detect movement in their 2,000-square-foot, two-story townhouse in Seattle, Washington. Because his house is two stories, Stephan decides to use one Type 2 device (named A) and two Type 1 devices (named B and C) on the first floor. The first floor is centered around three rooms: the kitchen, dining room, and living room, with the dining room in the middle and arranged in a straight line. The kitchen and living room are on opposite sides of the house. He places a Type 2 device (A) in the dining room, one Type 1 device (B) in the kitchen, and another Type 1 device (C) in the living room. With this device installation, he specifically uses the motion detection system to partition the first floor into three zones: the dining room, the living room, and the kitchen. When motion is detected by the AB pair and the AC pair, the system analyzes the motion information and associates the motion with one of three zones.

[0203] When Stefan and his family go away for the weekend (e.g., going camping for a long weekend), Stefan turns on the motion detection system using a mobile phone app (e.g., Android phone app or iPhone app). When the system detects motion, an alert signal is sent to Stefan (e.g., SMS text message, email, push message to the mobile phone app, etc.). If Stefan pays a monthly fee (e.g., $10 / month), a service company (e.g., a security company) receives the alert signal through a wired network (e.g., broadband) or wireless network (e.g., home WiFi, LTE, 3G, 2.5G, etc.) and performs security procedures for Stefan (e.g., calling him to check the problem, sending someone to check the house, contacting the police on Stefan's behalf, etc.). Stefan loves his elderly mother and is concerned about her well-being when he is home alone. When his mother is home alone while the rest of the family is out (e.g., going to work, shopping, or on vacation), Stephan uses his mobile app to turn on the motion detection system to ensure his mother is okay. He then uses the mobile app to monitor his mother's movements around the house. When Stephan uses the mobile app to see his mother moving around the house between three areas, according to her daily routine, Stephan knows that his mother is okay. Stephan is grateful that the motion detection system can help him monitor his mother's well-being while he is away from home.

[0204] On a typical day, his mother wakes up around 7:00 AM. She plans to make breakfast in the kitchen in about 20 minutes. She then eats breakfast in the dining room for about 30 minutes. Then, she does her daily exercise in the living room before sitting on the sofa and watching her favorite TV show. The motion detection system allows Stephan to see the timing of movements in each of three areas of the house. When the movements fit into her daily routine, Stephan knows roughly that his mother should be doing well. However, if the movement pattern seems abnormal (e.g., no movement until 10:00 AM, staying in the kitchen too long, remaining motionless for long periods of time, etc.), Stephan suspects something is wrong and calls his mother to check on her. Stephan may even ask someone (e.g., family member, neighbor, paid staff member, friend, social worker, service provider) to check on his mother.

[0205] Occasionally, Stephan feels the need to reposition the Type 2 device. He simply unplugs the device from its original AC power plug and plugs it into another AC power plug. He is pleased that the wireless motion detection system is plug-and-play and relocation does not affect the system's operation. Once powered on, it works immediately. On another occasion, Stephan is so confident that our wireless motion detection system can indeed detect motion with very high accuracy and very low alarms that he can actually use the mobile app to monitor motion on the first floor. He decides to install a similar configuration (i.e., one Type 2 device and two Type 1 devices) on the second floor to monitor the bedrooms on the second floor. Again, he finds system setup extremely easy, as he simply plugs the Type 2 and Type 1 devices into the second floor's AC power plugs. No special installation is required. He can then use the same mobile app to monitor motion on both the first and second floors. Each Type 2 device on the first and second floors can interact with all Type 1 devices on both the first and second floors. Stephen is happy to see that as he doubles his investment in Type 1 and Type 2 devices, he has more than double the capacity of the combined system.

[0206] According to various embodiments, each CI (CI) may include at least one of channel state information (CSI), frequency domain CSI, a frequency representation of CSI, frequency domain CSI associated with at least one subband, time domain CSI, intra-domain CSI, a channel response, a channel response estimate, a channel impulse response (CIR), a channel frequency response (CFR), channel characteristics, a channel filter response, CSI of a wireless multipath channel, information of a wireless multipath channel, a timestamp, auxiliary information, data, metadata, user data, account data, access data, security data, session data, status data, management data, family data, identity (ID), identifier, device data, network data, proximity data, environmental data, real-time data, sensor data, stored data, encrypted data, compressed data, protected data, and / or another CI. In one embodiment, the disclosed system includes hardware components (e.g., a wireless transmitter / receiver with an antenna, analog circuitry, a power supply, a processor, a memory) and corresponding software components. According to various embodiments of the present disclosure, the disclosed system includes a Bot (referred to as a Type 1 device) and an Origin (referred to as a Type 2 device) for vital signs detection and monitoring, each device comprising a transceiver, a processor, and a memory.

[0207] The disclosed system can be applied in many ways. In one example, a Type 1 device (transmitter) may be a small WiFi-enabled device placed on a table. It may also be a WiFi-enabled television (TV), set-top box (STB), smart speaker (e.g., Amazon Echo), smart refrigerator, smart microwave, mesh network router, mesh network satellite, smartphone, computer, tablet, smart plug, etc. In one example, a Type 2 (receiver) may be a WiFi-enabled device placed on a table. It may also be a WiFi-enabled television (TV), set-top box (STB), smart speaker (e.g., Amazon Echo), smart refrigerator, smart microwave, mesh network router, mesh network satellite, smartphone, computer, tablet, smart plug, etc. Type 1 and Type 2 devices may be placed in / near a conference room to count people. Type 1 and Type 2 devices may be a health monitoring system for the elderly to monitor daily activities and any signs of symptoms (e.g., dementia, Alzheimer's disease). Type 1 and Type 2 devices may be used in an infant monitor to monitor the vital signs (breathing) of living infants. Type 1 and Type 2 devices can be placed in bedrooms to monitor sleep quality and any sleep apnea. Type 1 and Type 2 devices can be placed in automobiles to monitor passenger and driver health, detect driver sleep, and detect any babies left in the car. Type 1 and Type 2 devices can be used in logistics to prevent human trafficking by monitoring people hidden in trucks and containers. Type 1 and Type 2 devices can be deployed by emergency services in disaster areas to search for victims trapped in rubble. Type 1 and Type 2 devices can be placed in an area to detect the breathing of any intruders. Non-wearable wireless respiratory monitoring has many applications.

[0208] The hardware modules may be configured to include Type 1 transceivers and / or Type 2 transceivers and may be sold / used under variable brands to design, build, and sell final commercial products. The products using the disclosed systems and / or methods may be home / office security products, sleep monitoring products, WiFi products, mesh products, TVs, STBs, entertainment systems, HiFi, speakers, home appliances, lamps, stoves, ovens, microwaves, tables, chairs, beds, shelves, tools, appliances, torches, vacuum cleaners, smoke detectors, sofas, pianos, fans, doors, windows, door / window handles, locks, smoke detection equipment, car accessories, computing devices, office supplies, air conditioners, heaters, pipes, connectors, surveillance cameras, access points, computer equipment, mobile devices, LTE devices, 3G / 4G / 5G / 6G devices, UMTS devices, 3GPP devices, GSM devices, EDGE devices, TDMA devices, FDMA devices, CDMA devices, WCDMA devices, TD-SCDMA devices, gaming devices, eyeglasses, glass panels, VR goggles, necklaces, watches, waistbands, belts, wallets, pens, hats, implanted devices, tags, parking tickets, smartphones, etc.

[0209] The summarization may include: analysis, output response, selected time window, sub-sampling, transformation, and / or projection. The presentation may include presenting at least one of a month / week / day view, simplified / detailed view, cross-sectional view, small / large form factor view, color-coded view, comparison view, summary view, video, web view, audio announcement, and another presentation related to the cyclical / repetitive nature of the recurring motion.

[0210] A Type 1 / Type 2 device is any device that includes an antenna, a device having an antenna, a device having a housing (e.g., for a radio, antenna, data / signal processing unit, radio IC, circuitry), a device that interfaces / attaches / connects / links to another device / system / computer / phone / network / data aggregator, a device having a user interface (UI) / graphical UI / display, a device having a wireless transceiver, a device having a wireless transmitter, a device having a wireless receiver, an Internet of Things (IoT) device, a device having a wireless network, a device having both wired and wireless network capabilities, a device having a wireless integrated circuit (IC), a Wi-Fi device, a device having a Wi-Fi chip (e.g., compliant with 802.11a / b / g / n / ac / ax standards), a Wi-Fi access point (AP), a Wi-Fi client, a WiFi router, a Wi-Fi repeater, a WiFi hub, a WiFi mesh network router / hub / AP, a wireless mesh network router, an ad-hoc network device, a wireless mesh network device, a mobile device (e.g., 2G / 2.5G / 3G / 3G).5G / 4G / LTE / 5G / 6G / 7G, UMTS, 3GPP, GSM, EDGE, TDMA, FDMA, CDMA, WCDMA, TD-SCDMA), cellular device, base station, mobile network base station, mobile network hub, mobile network compatible device, LTE device, device with LTE module, mobile module (e.g., circuit board with mobile-enabled chip (IC) such as Wi-Fi chip, LTE chip, BLE chip), device with mobile module, smartphone, companion device for smartphone (e.g., dongle, attachment, plug-in), dedicated device, plug-in device, AC-powered device, battery-powered device, device with processor / memory / instruction set, smart device / gadget Object / Item: A Type 1 device and / or Type 2 device may be a watch, stationery, pen, user interface, paper, mat, camera, television (TV), set-top box, microphone, speaker, refrigerator, oven, machine, phone, wallet, furniture, door, window, ceiling, floor, wall, table, chair, bed, nightstand, air conditioner, heater, pipe, duct, cable, carpet, decoration, gadget, USB device, plug, dongle, lamp / light, tile, ornament, bottle, vehicle, automobile, AGV, drone, robot, laptop, tablet, computer, hard disk, network card, equipment, racket, ball, shoe, wearable device, clothing, eyeglasses, hat, necklace, food, pill, small device that moves within a living being's body (e.g., blood vessels, lymph, digestive system), and / or another device. The Type 1 device and / or Type 2 device may be communicatively coupled to the Internet, another device that accesses the Internet (e.g., a smartphone), a cloud server (e.g., a hub device), an edge server, a local server, and / or storage. Type 1 and / or Type 2 devices may operate under local control, may be controlled by another device via a wired or wireless connection, may operate automatically, or may be controlled by a central system located remotely (e.g., away from the home).

[0211] In one embodiment, a Type-B device may be a transceiver that may perform as both an origin (Type-2 device, Rx device) and a bot (Type-1 device, Tx device), i.e., a Type-B device may be both a Type-1 (Tx) device and a Type-2 (Rx) device (e.g., simultaneously or alternately), such as a mesh device, mesh router, etc. In one embodiment, a Type-A device may be a transceiver that may function only as a bot (Tx device), i.e., it may be only a Type-1 device or only a Tx, such as a simple IoT device. It may have the functionality of an origin (Type-2 device, Rx device), but in some embodiment, it functions only as a bot. All Type-A and Type-B devices form a tree structure. The root may be a Type-B device that has access to a network (e.g., the Internet). For example, it may be connected to broadband service via a wired connection (e.g., Ethernet, cable modem, ADSL / HDSL modem) or a wireless connection (e.g., LTE, 3G / 4G / 5G, WiFi, Bluetooth, microwave link, satellite link, etc.). In one embodiment, all Type A devices are leaf nodes. Each Type B device may be a root node, a non-leaf node, or a leaf node.

[0212] The present disclosure discloses a system for monitoring rhythmic movements such as gait based on wireless signals and channel information of a wireless multipath channel affected by gait movements, and the system may recognize gait and identify and authenticate individuals accordingly.

[0213] Gait recognition is particularly attractive for a variety of ubiquitous applications requiring human identification because it can be achieved remotely and without the active cooperation of a user. For example, a smart building might automatically open its doors when an authorized user enters. A smart home might change the temperature and ambient light to match the recognized user. A smart TV within the smart home might respond with the recognized user's favorite programs. Smart home devices such as Google Home and Amazon Alexa can directly interact with the recognized user in a more intimate way. For all of this to work, the user simply walks through the space as usual. However, continuous, passive gait recognition requires a system that is easy to deploy and convenient.

[0214] Conventional gait measurement / recognition systems typically rely on cameras, floor sensors, and / or wearable devices to capture gait information. The target subject must walk within a limited area (usually only an instrumented corridor) or wear body sensors (e.g., accelerometers). Therefore, they are primarily limited to research and clinical use, and lack the convenience and comfort required for ubiquitous applications in smart homes and smart buildings. Recently, a new type of gait recognition using wireless signals has emerged. Wireless signals (e.g., Wi-Fi) are ubiquitous and can penetrate obstacles such as walls and furniture, supporting the possibility of through-wall gait detection. However, existing methods require the subject to walk a predefined path in a predefined direction. Therefore, they are only suitable for limited areas with strong line-of-sight (LOS) conditions (e.g., narrow paths such as a 5-meter hallway). Furthermore, most existing techniques do not physically measure gait. They merely extract RF-based features believed to represent walking patterns. Because such RF features are usually closely related to the surrounding environment, existing techniques are location- and environment-dependent. Most importantly, none of these systems can function in non-line-of-sight (NLOS) situations, let alone detect through walls.

[0215] The wireless / RF-based system disclosed in this disclosure can achieve many advantages, including: (a) the disclosed system can operate in both LOS and non-line-of-sight (NLOS) conditions; (b) the disclosed system does not require the cooperation of the monitored person (i.e., the person is free to walk however they want); (c) the disclosed system is inexpensive because it uses common off-the-shelf (COTS) components; (d) each device used in the disclosed system can cover a large area; (e) the disclosed system is easy to install; (f) the disclosed system does not require cameras, which invade privacy; (g) the disclosed system does not require the person to wear a wearable device; and (h) the disclosed device is location-independent.

[0216] One goal of the disclosed system is to obtain computable features suitable for recognition / authentication from channel state information (CSI) without imposing constraints on the monitored subject, such as restrictions on wearing a wearable device and gait style. In general, the disclosed system can monitor a subject's rhythmic movement based on time-series CI. For example, breathing is periodic and rhythmic. Gait is somewhat periodic, but different gait cycles may have different periods. The "step" phase corresponding to the left foot touching the ground may be different from the "step" phase corresponding to the right foot. Asymmetry between left-foot and right-foot steps (e.g., in the case of a single-foot injury) is an important feature for assessing many health-related issues. Therefore, gait movement is actually generated rhythmically with alternating movements of the left and right feet, and is therefore referred to as rhythmic rather than periodic. For illustrative purposes, gait is used as an example of rhythmic movement, but the systems and methods disclosed in this disclosure are applicable to any rhythmic movement.

[0217] The Type 1 and / or Type 2 devices (together with the wireless signal, TSCI, processor, memory, and instruction set) can also be used to monitor other types of motion, such as transient motion (e.g., falls, gestures, writing, body motion, people motion, vehicle motion, and in-vehicle activity).

[0218] The Type 1 device and / or Type 2 device may be a standalone device or an embedded device, which may be connected to another system (for power supply, signal transmission, network access, etc.) using connectors / ports such as USB (e.g., Type A / B / C / D / E, micro USB, mini USB, etc.), Thunderbolt, Firewire, Lightning (e.g., in Apple devices such as iPhone, iPad, AirPods, etc.), OBD (on-board diagnostic port), cigarette lighter port (e.g., 12V), PCI (e.g., PCI, PCI Express, etc.), VGA, DVI, HDMI, parallel port, serial port, ADAT, BNC, D-SUB, F, MIDI, UHF, MCX, N, PS / 2, RCA, SATA, S-Video, SATA, mSATA, m.2, SMA, SMB, SMC, S / PDIF, RJ-11, RJ-45, SCSI, TNC, TS, TRS, UHF, mini UHF, XLR, coaxial, optical, Ethernet, DisplayPort, etc.

[0219] Any of the devices may be powered by a battery (e.g., AA batteries, AAA batteries, coin batteries, button batteries, small batteries, battery banks, mobile batteries, car batteries, hybrid batteries, vehicle batteries, container batteries, non-rechargeable batteries, rechargeable batteries, NiCd batteries, NiMH batteries, lithium ion batteries, zinc carbon batteries, zinc chloride batteries, lead acid batteries, alkaline batteries, batteries with wireless chargers, smart batteries, solar cells, marine batteries, aircraft batteries, other batteries, temporary energy storage devices, capacitors, flywheels).

[0220] Any of the devices may be powered by DC or direct current (e.g., from batteries, generators, power converters, solar panels, rectifiers, DC-DC converters as mentioned above at various voltages such as 1.2V, 1.5V, 3V, 5V, 6V, 9V, 12V, 24V, 40V, 42V, 48V, 110V, 220V, 380V, etc.) and therefore may have a DC connector or a connector with at least one pin for a DC power source.

[0221] Any of the devices may be powered by AC or alternating current (e.g., from a domestic wall socket, transformer, inverter, shore power, etc. at various voltages such as 100V, 110V, 120V, 100-127V, 200V, 220V, 230V, 240V, 220-240V, 100-240V, 250V, 380V, 50Hz, 60Hz, etc.) and therefore may have an AC connector or a connector with at least one pin for AC power. Type 1 devices and / or Type 2 devices may be positioned (e.g., installed, positioned, moved) on-site or off-site.

[0222] For example, in a vehicle (e.g., automobile, truck, lorry, bus, special vehicle, tractor, excavator, telehandler, bulldozer, crane, forklift, electric trolley, AGV, emergency vehicle, cargo, wagon, trailer, container, ship, ferry, boat, submarine, aircraft, airship, lift, monorail, train, tram, rail vehicle, rail car, etc.), the Type 1 device and / or the Type 2 device may be an embedded device built into the vehicle, or may be an add-on device (e.g., an aftermarket device) plugged into a port within the vehicle (e.g., an OBD port / socket, a USB port / socket, an accessory port / socket, a 12V auxiliary power outlet and / or a 12V cigarette lighter port / socket).

[0223] For example, one device (e.g., a Type 2 device) may be plugged into a 12V cigarette lighter / accessory port or an OBD port or a USB port (e.g., of a car / truck / vehicle), and the other device (e.g., a Type 1 device) may be plugged into a 12V cigarette lighter / accessory port or an OBD port or a USB port (e.g., of a car / truck / vehicle). The OBD port and / or USB port can provide power, signal transmission, and / or networking (of the car / truck / vehicle). The two devices can jointly monitor passengers, including children / babies, in the vehicle. They may be used to count passengers, recognize the driver, and detect the presence of passengers in specific seats / positions in the vehicle.

[0224] In another example, one device may be plugged into a 12V cigarette lighter / accessory port or OBD port or USB port of a car / truck / vehicle, and the other device may be plugged into a 12V cigarette lighter / accessory port or OBD port or USB port of another car / truck / vehicle.

[0225] In another example, there may be many devices of the same Type A (e.g., Type 1 or Type 2) in many different types of vehicles / portable devices / smart gadgets (e.g., automated guided vehicles / AGVs, shopping / luggage / mobile carts, parking tickets, golf carts, bicycles, smartphones, tablets, cameras, recording devices, smart watches, roller skates, shoes, jackets, goggles, hats, eyewear, wearables, Segways, scooters, luggage tags, sweepers, vacuum cleaners, pet tags / collars / wearables / implants), each plugged into the vehicle's 12V accessory port / OBD port / USB port or built into the vehicle. There may also be one or more other Type B devices (e.g., Type 1 B if A is Type 2, or Type 2 B if A is Type 1) installed in locations such as gas stations, lampposts, street corners, tunnels, parking structures, scattered locations to cover a wide area such as factories / stadiums / train stations / shopping malls, etc. The Type A devices may be located, tracked, or monitored based on the TSCI.

[0226] In one embodiment, a scattering multipath model is considered to derive a statistical method for passive velocity estimation, in addition to a scattering model that can capture the velocity when the target is up to 10 meters away from the link and / or on the other side of a wall. The proposed model provides new directions and opportunities for indoor wireless sensing. Furthermore, a gait recognition system called "GaitWay" is built based solely on walking speed. By extracting various physically plausible features, GaitWay can recognize a subject's gait independently of location, posture, environment, and user clothing.

[0227] Given a large number of multipaths, the disclosed system can statistically examine channel characteristics by considering all multipaths together based on the disclosed method. The target's moving speed can be calculated from the ACF of the CSI. Built on the statistical properties of a large number of multipaths, the disclosed method is independent of the environment, location, and user posture. The disclosed method automatically detects stable periods during a user's normal activities. The disclosed system can measure such gait periodicity using the ACF of speed. The disclosed system can estimate gait cycle time and segment the speed sequence by stride. To monitor and evaluate a subject's gait, the disclosed system can examine three simple characteristics: average walking speed, gait cycle time, and stride length. Furthermore, the disclosed system can also employ stability and symmetry measures, such as harmonic ratios, to evaluate gait progression. The disclosed system utilizes support vector machines (SVM) to recognize various gaits based on features extracted from speed estimates, including speed, acceleration, symmetry, rhythmicity and cadence, smoothness, recursive quantitative analysis, and ACF features.

[0228] A bipedal animal (e.g., a human) has two step segments for a gait cycle. For example, a first step segment may correspond to a forward movement of the left foot, and a second step segment may correspond to a forward movement of the right foot. A quadrupedal animal (e.g., a dog) has four step segments for a gait cycle. Each step segment may correspond to a movement of each foot. The step segments may have different lengths.

[0229] The disclosed system can explore a multipath model rich in scattering to apply the disclosed method. Instead of simplifying an object (e.g., a human body) as a single reflector generating only one dominant reflection path in the Doppler Frequency Shift (DFS) model as shown in FIG. 1A, FIG. 1B illustrates an exemplary multipath model in a scattering-rich indoor environment in which an object scatters signals to generate multiple paths, in accordance with some embodiments of the present disclosure. As shown in FIG. 1B, the human body 110 is viewed as multiple scatterers that reflect signals in various directions, and these signals are superimposed at the Tx 120 with signals scattered by other objects from all directions. Given multiple multipaths, rather than geometrically analyzing a specific reflection path or assuming a dominant path while ignoring other paths, the disclosed system can statistically explore the channel characteristics by considering all multipaths together. This allows the target's movement speed to be calculated from the ACF of the CSI. Because it is built on the statistical properties of multiple multipaths, the disclosed method is independent of the environment, location, and user posture. In contrast to previous reflection models that do not work in multipath-rich environments, the proposed model performs well in the presence of more multipath and supports detection through walls.

[0230] The disclosed system can also track velocity through walls. Consider a wireless transmission pair, each equipped with an omnidirectional antenna. The channel frequency response (CFR) of the multipath channel at time t, also known as channel state information (CSI), may be generally modeled as: TIFF0007779954000001.tif12144 where a1(t) and τl(t) represent the complex amplitude and propagation delay of the l-th multipath component (MPC), respectively, and Ω represents the set of MPCs.

[0231] Due to timing, frequency synchronization offsets and additive thermal noise, TIFF0007779954000002.tif20115 where α(t) and β(t) are the random initial distortion and linear phase distortion at time t, respectively.

[0232] Wireless signals are scattered by many scatterers such as walls, ceilings, floors, furniture, human bodies, etc. Based on the principle of EM wave superposition, the CSI, H(t,f), can be decomposed as follows: TIFF0007779954000003.tif14170 where Ωs(t) denotes the set of static scatterers, Ωd(t) denotes the set of dynamic scatterers, and H(t,f) represents the contribution of the i-th scatterer. ε(t,f) is a noise term, which can be approximated as additive white Gaussian noise (AWGN) with variance σ2(f) and is statistically independent of H(t,f). The decomposition intuitively shows that each scatterer can be treated as a "virtual Tx" that spreads the received EM waves in all directions, and these EM waves are summed at the receiving antenna after bouncing off interior objects indoors. As a result, H(t,f) actually measures the sum of the electric fields of all received EM waves. In practice, it is reasonable to assume that, within a sufficiently short time period, both the sets Ωs(t) and Ωd(t) change slowly with time, allowing them to be approximated as time-invariant sets.

[0233] We can consider a 2D scattering model where all scatterers lie in the same horizontal plane. Due to channel reciprocity, EM waves traveling in both directions experience the same physical perturbations (i.e., reflection, refraction, diffraction, etc.). Therefore, if the receiver is transmitting EM waves, the CIS "measured" at the ith scatterer or "virtual Tx" will be the same as Hi(t,f). If the velocity of the ith scatterer is vi, then the continuum-limited expression of Hi(t,f) can be written as TIFF0007779954000004.tif17144 where Fi(θ,f) denotes the complex channel gain of the MPC from the direction θ for the ith scatterer; TIFF0007779954000005.tif1077

[0234] The statistical theory of EM fields developed for reverberant cavities provides a good approximation for indoor environments. Based on this, Fi(θ,f) for ∀i can be expressed as a random variable with the following properties: for ∀θ, Fi(θ,f) is a circularly symmetric Gaussian random variable with the same variance σF2(f), and for ∀θ1 / =θ2, Fi(θ1,f) and Fi(θ2,f) are statistically independent for ∀θ1 and ∀θ2, for ∀i / =j∈Ωd.

[0235] Using the above properties, we can examine how the ACF of a CSI relates to the velocity vi. The mean of Hi(t,f) is equal to zero, i.e., E[Hi(t,f)]=0, where E[·] denotes the expectation operator. Then, we can write the covariance of two CSIs with a time lag τ as follows: JPEG0007779954000006.jpg23117If the ACF of Hi(t,f) with a time lag τ is ρHi(τ,f), it can be derived as follows: TIFF0007779954000007.tif16148

[0236] Similarly, if the ACF of H(t,f), which is CSI with a time lag τ, is ρH(τ,f), it can be obtained as follows: TIFF0007779954000008.tif21119In the formula, δ(·) is the Dirac delta function. As can be seen from the formula, ρH(τ,f) is a linear combination of the ACFs of Hi(t,f), and the weight of each term is equal to the energy scattered by the corresponding scatterer.

[0237] Consider the case where only one person is moving within the monitored area. We can approximate that all scatterers have the same velocity, i.e., vi = v for ∀i∈Ωd.

[0238] In that case, ρH(τ,f) can be simplified as follows: TIFF0007779954000009.tif20167where α(f) is defined as the gain of each subcarrier f. Equation (8) bridges the gap between the human body's moving speed and the second-order statistics of CSI, i.e., ACF.

[0239] In practice, an estimate of the ACF, the sample ACF, is used instead, and n(τ,f) can be used to denote the estimated noise of the ACF, i.e., TIFF0007779954000010.tif14139

[0240] Since the term J0(kvτ) in equation (9) is a function of the moving speed v, it will be referred to as the speed signal hereinafter. However, in reality, the signal-to-noise ratio (SNR) of the velocity signal on each subcarrier modulated by human motion may be very low, especially if the monitored person is far from the link or on the other side of a wall. As a second-order statistic, the ACF avoids phase issues and is synchronized across all subcarriers. This allows the ACFs measured on different subcarriers to be directly combined. In the following, a novel scheme is proposed based on maximal ratio combining (MRC), which combines velocity signals measured on multiple subcarriers in an optimal manner to maximize the SNR of the velocity signal. MRC is a diversity combining method in telecommunications that optimizes the SNR by combining signals received from multiple antennas. In this specification, MRC is applied by treating subcarriers as receive diversity.

[0241] When α(f) is small, i.e., when the proportion of white noise in H(t,f) is large, each tap of the ACF follows a zero-mean normal distribution with mean variance 1 / N, i.e., TIFF0007779954000012.tif1495, where N is the number of samples used in ACF estimation. Therefore, the variance of n(τ,f) in equation (9) is the same for different subcarriers. Considering that the noise terms for different subcarriers are statistically independent, it can be shown that the MRC scheme achieves the maximum SNR of the velocity signal J0(kvτ). That is, it can be expressed as follows: TIFF0007779954000013.tif10159In the formula, S(τ) is called the composite velocity signal, and w * (f) denotes the optimal combining weight for subcarrier f, and w * (f) is linearly proportional to the gain α(f).

[0242] However, the gain α(f) at each subcarrier cannot be obtained directly from the CSI. Fortunately, since J0(kvτ) is continuous at time lag 0, i.e., limτ→0 J0(kvτ)=1, we can obtain α(f)=limτ→0 ρH(τ,f) according to equation (8). Therefore, if the channel sampling rate Fs is sufficiently high, TIFF0007779954000014.tif20156

[0243] The intuition behind MRC maximizing SNR is that if you combine all the subcarriers properly, their noise terms are independent, so the "good" subcarriers boost the signal while the "bad" subcarriers help to attenuate the noise.

[0244] FIG. 2A shows an example of a composite velocity signal, and FIG. 2C shows a matrix of composite velocity signals. In the figure, each column of the matrix corresponds to a composite velocity signal. As shown in FIG. 2A, the shape of the composite velocity signal resembles a Bessel function J0(x) where x = kvτ, and the velocity v can be extracted by matching their key features, such as the location of the first peak or trough. In one embodiment of GaitWay, the first peak can be used, i.e., the velocity is calculated as follows: TIFF0007779954000015.tif1483 where x0 is a constant value corresponding to the first peak of the Bessel function J0(x), TIFF0007779954000016.tif1112 is the time lag corresponding to the first peak in the composite velocity signal, indicated by point 210 in FIG. 2C.

[0245] In one embodiment, two further steps are performed to improve the speed estimation. First, to facilitate finding the peak, the difference of the composite speed signal is used, as shown in FIGS. 2B and 2D. In this case, x0 corresponds to the first peak of the derivative of J0(x). Second, the disclosed system can use the phase difference between the two receive antennas to eliminate errors in the raw phase. Thanks to ACF, the disclosed method is insensitive to initial phase offsets. Figure 3 shows speed estimates over a 10-second period while a user is walking continuously. Figure 3 further demonstrates that MRC significantly improves speed estimation.

[0246] The disclosed system implements the DFS-based method and can compare it with the proposed method using actual measurements. Specifically, the disclosed system can set up two links (one Tx and two Rx in line), one in LOS and the other in NLOS. The user is asked to walk toward the link, and the two receivers can simultaneously measure CSI. Figure 4 shows the estimated speeds by the GaitWay-based method and the DFS-based method. As can be seen from the figure, GaitWay accurately captures the speed using either the LOS link or the NLOS link, preserving the accurate instep speed variation. The speed estimates for both links are highly consistent, with only small absolute differences. However, the DFS-based method cannot capture the accurate speed in both the LOS and NLOS cases.

[0247] Rather than only allowing a user to walk at a nearly constant speed along a predefined linear path and assuming that all data is collected during stable walking, the disclosed system aims to acquire gait information for free and natural walking. A user may perform various activities, such as walking, sitting, standing, and typing. Furthermore, a user may walk at different speeds, especially when starting to walk from a standing position, changing direction, or about to stop. During these periods, walking speed does not necessarily reflect the most distinctive and stable gait features. Therefore, the first step in gait analysis / recognition is to identify stable walking periods, during which the subject normally walks at their usual pace.

[0248] Algorithms can be devised to automatically detect periods of stability during a user's normal activity. When a user is walking smoothly, the observed speed reaches a certain range and follows a repeating pattern with a periodic walking rhythm.

[0249] The disclosed system can measure the periodicity of such walking using the speed ACF. As shown in FIG. 5, when a user is walking steadily, a clear peak is observed from the speed ACF. In contrast, the ACF becomes flatter for fluctuating walking. The disclosed system can apply a sliding window (e.g., 3 seconds) to the speed estimates and calculate the ACF for each window. The disclosed system can then use peak detection on the speed ACF to look for the first peak. A period is considered to be stable walking only if a continuous series of reliable peaks is observed.

[0250] To be more robust, the disclosed system can further check the mean central tendency of walking speed. A walking period is used for gait analysis only if the mean speed is greater than a certain value (e.g., 0.7 m / s, which is less than the normal human walking speed of 1.0 m / s to 2.0 m / s). Figure 7 shows an example of identified stable walking periods. The speed is measured when a user walks back and forth through a 10-meter corridor twice. In this case, each stable period becomes a gait instance of the speed sequence V = [v(ti), i = 1, 2, ..., M].

[0251] The disclosed system can also estimate a gait cycle, defined as the period between two consecutive heel-to-ground strikes during walking. In addition to estimating gait cycle time, the disclosed system segments the velocity series into strides. During normal human walking, the subject's velocity increases and then decreases, resulting in a velocity peak occurring with each stride. Therefore, the disclosed system can identify steps by performing simple peak detection on the velocity series, as shown in FIG. 6. To remove noise and outliers, specific constraints (including peak prominence and height) are applied to the peak detection. Once all steps have been identified, the disclosed system can extract gait periods by deleting the period before the first peak and the period after the last peak. The remaining trajectory becomes a valid gait instance for further analysis in one embodiment of GaitWay.

[0252] The disclosed system can also analyze individual gaits. From walking speed and identified gait cycles, various gait characteristics can be analyzed. To monitor and evaluate a subject's gait, the disclosed system can examine three characteristics: average walking speed, gait cycle time, and stride length. Furthermore, the disclosed system can employ a measure of stability and symmetry, i.e., harmonic ratio, to evaluate gait progression. A stride is a complete gait cycle, which includes two step segments for bipedal animals, three step segments for tripedal animals, and four step segments for quadrupedal animals.

[0253] In one embodiment, the average walking speed is simply obtained as the average of instantaneous estimates of the user's walking speed. As shown in Figure 8, the disclosed measurements show that different users have different habitual speeds and that a user's walking speed fluctuates over time.

[0254] In one embodiment, the gait cycle time is calculated as the average duration of every two consecutive steps. The middle panel of Figure 8 shows the average cycle times of two users' walking instances measured at different locations and times. Across the 20 trajectories, a variance of 0.7 ms and 0.6 ms is observed for the two users, respectively.

[0255] In one embodiment, stride length estimation can also be performed by the disclosed system. Thanks to accurate velocity estimation, stride length can be intuitively derived by integrating the velocity estimate over the duration of each step. The bottom panel of Figure 8 shows the estimated stride lengths of two users.

[0256] Harmonic ratio (HR) may be employed as a quantitative measure of gait smoothness. There has been growing interest in using HR techniques to study the effects of various pathologies and monitor rehabilitation. HR examines the inter-step symmetry within a stride by quantifying the harmonic components of acceleration for a given stride. First, a discrete Fourier transform (DFT) is performed on the acceleration within each stride. HR may be defined as the ratio of the sum of the amplitudes of even harmonics to the sum of the amplitudes of odd harmonics. The disclosed system can calculate HR using the first 20 harmonics. This is valid for a normal gait, where the majority of power occurs below 10 Hz. Figure 9 shows the HR of a gait trajectory over seven cycles (14 steps), demonstrating the progression of inter-step symmetry during gait.

[0257] The disclosed system examines the pairwise relationship between stride length, stride time, and walking speed, which is of interest for research related to pedestrian dead reckoning. To this end, the disclosed system integrates measurements from six users collected under different conditions, and the relationship can be visualized in Figure 10. As shown, stride length clearly varies for different users' walking speeds, which are represented by different grayscales. Therefore, assuming a fixed stride length would result in significant errors, and therefore, as is done in many conventional techniques for pedestrian dead reckoning, stride length is multiplied by the number of steps to derive walking distance.

[0258] Furthermore, in the disclosed measurements, data is collected and extracted automatically as the subject walks around freely, without requiring them to walk along a predefined path or in a predefined direction, or at a deliberate speed. In fact, the subjects involved do not even need to be aware of the data collection. This is a key characteristic supporting a continuous gait monitoring / recognition system. This is in stark contrast to conventional techniques that impose some or all of the above constraints on the subject to acquire environmentally reproducible features for human recognition or to derive a DFS.

[0259] Instead of directly extracting data-driven features, including low-plausibility and environment-dependent features, from CSI measurements, the disclosed system performs human identification by extracting true gait features, which are physically plausible and environment-independent, from velocity estimates. In addition to the gait monitoring parameters described above, the disclosed system can devise many features that characterize various aspects of human gait patterns.

[0260] Regarding gait velocity, the disclosed system does not directly use the average walking speed as a feature because a subject's speed varies greatly over time. The disclosed system can utilize features that are independent of the average walking speed. As shown in FIG. 11A, the disclosed system can detrend the absolute velocity by first subtracting the average central velocity. Then, the disclosed system can calculate various percentile values ​​of the velocity deviation (e.g., in one GaitWay embodiment, by obtaining the 95th, 75th, and 50th percentiles as shown in FIG. 11B). Specifically, the disclosed system can obtain these percentile values ​​for each of the positive deviation, negative deviation, and the absolute value of the total deviation.

[0261] Acceleration may be calculated as the derivative of velocity. The disclosed system can obtain the maximum, minimum, and variance of acceleration. Because walking acceleration also exhibits a sinusoidal pattern, the disclosed system can also identify the peaks and valleys in the acceleration sequence and calculate the variance of each.

[0262] The level of symmetry is also a gait feature. The disclosed system calculates the step time and stride length of the left and right legs, respectively, and can obtain their mean and standard deviation as features. The difference between the right and left legs for each feature is derived as a measure of gait symmetry.

[0263] Rhythmicity and cadence can be measured based on gait. Referring again to FIG. 5, the disclosed system can calculate the ACF of walking speed. When a user walks rhythmically, the speed ACF exhibits multiple prominent peaks and slowly decays. Thus, the ACF embodies the rhythmicity or dynamic stability of gait. Therefore, the disclosed system can develop several features based on the speed ACF. In one embodiment, the disclosed system, GaitWay, can apply a sliding window to calculate a series of ACFs for each gait instance, resulting in a speed ACF matrix. From there, the disclosed system can first identify the prominent peaks and corresponding delays of each ACF, and then extract single-valued features: the mean and variance of the value of the first ACF peak, the number of identified prominent peaks, the variance of the peak cycle (i.e., the difference in peak delay of each ACF corresponding to the difference in consecutive stride cycle times), and the proportion of ACFs in the matrix where no prominent first peak is observed.

[0264] In one embodiment, Harmonic Ratio (HR) is used as a measure of gait smoothness. The disclosed system can obtain one HR value per gait cycle during walking. To obtain a single-valued feature of the gait trajectory, the disclosed system can obtain the median and variance of the HR values.

[0265] To quantify gait variability, the disclosed system can employ recurrence quantification analysis (RQA), a method of nonlinear data analysis that quantifies the number and duration of recurrences of a dynamic system represented by its phase space trajectory. It describes recurrence properties by analyzing a recurrence plot (RP), which visualizes and reveals all the times when the phase space trajectory of a dynamic system appears in approximately the same region of phase space. Such an RP is represented by an N*N matrix R: It is represented mathematically as JPEG0007779954000017.jpg362. In one embodiment of GaitWay, a state-space trajectory X is constructed from a velocity sequence {v i , i = 1, 2, ..., L} with an embedding dimension of 5 and a delay of 10 samples. Figure 12 shows two example RPs for two different users. The upper RP exhibits more diagonals, indicating a more stable, periodic gait. RQA allows for the derivation of multiple metrics based on the RP. In one embodiment of GaitWay, four metrics are available to reflect various characteristics of the RP: (1) recursion rate, which is the percentage of recursion points in the RP; (2) decision rule, which is the percentage of recursion points that form diagonals; (3) Shannon entropy, which is the probability distribution of the diagonal lengths; and (4) the average diagonal length. The metrics show that RQA reaches a stable value when calculated over four gait cycles. This suggests a minimum length for detecting stable gait periods. In practice, the disclosed system can be relaxed to a minimum of three cycles (i.e., six steps) to collect more usable data, and walking periods of less than three cycles may not be considered for gait recognition.

[0266] The disclosed system can also examine the ACF of the CSI (shown in FIGS. 2A-2D) for feature extraction. The ACF is closely related to walking speed but independent of location and environment, and therefore serves as a signature for gait classification. Specifically, rather than using all ACFs within a walking period, the ACFs corresponding to speed peaks, as identified in FIG. 6, can be considered. Note that peak speeds may vary within a walking trajectory; that is, the location of the first peak of the ACF varies over time. Therefore, the disclosed system can scale all ACFs to correspond to the average peak speed. For greater clarity, the disclosed system can obtain the difference of each ACF and then average the scaled ACF differences. FIG. 13 shows an example of scaled ACF differences. The disclosed system can use the first 50 taps as the feature vector in the disclosed system.

[0267] The disclosed system can compile all of the above features, resulting in a 90-dimensional feature vector for each gait instance. The disclosed system can perform the following two steps for latent dimensionality deduction. First, the disclosed system can examine whether these features are correlated with each other. The disclosed system can calculate pairwise correlations for all features (except the 50-dimensional ACF feature taken as a whole) and plot a correlation matrix as shown in Figure 14. Most of the extracted features are independent of each other. As shown by the four clustered regions 1410 along the diagonal of the correlation matrix, some features are more highly correlated than others. For example, features #2-#8 are all related to cycle time and therefore highly correlated. Similarly, features #16-#23 are different measures of speed deviation. Features #30-#34 are related to acceleration, and features #38-#40 are all extracted by RQA. The disclosed system can remove one of each pair of highly correlated features for classification. Second, the disclosed system can employ a result-based method for feature selection. Specifically, the disclosed system performs 10 cross-validation runs with and without a particular feature and can retain that feature only if it improves the output classification accuracy.

[0268] Given the extracted features, the disclosed system can identify users (from other humans) by their gait patterns. Following gait recognition, two identification cases can be considered: single-user authentication, which verifies whether the user is a target subject or an unknown stranger, and multi-user recognition, which identifies which target subjects the user is among a set of candidates. The disclosed system can utilize a classification technique, support vector machines (SVM), for this purpose. Because the primary goal here is to demonstrate the effectiveness of speed estimates and physically plausible features for gait recognition, the disclosed system can use SVM instead of common deep learning techniques. In one embodiment, the application of deep learning can continue in the disclosed system.

[0269] For single-user authentication, the disclosed system can train a gait model for a subject by creating a binary classifier that considers the subject's gait instances as a positive class and some benchmark users as a negative class. Benchmark data can be obtained from available standard public databases. In one GaitWay embodiment, they are randomly selected from experimental participants. To authenticate a subject, the disclosed system can calculate the probability that an instance fits the target class. If the probability is higher than a threshold, the test gait instance is considered to belong to the subject of interest; if the probability is below the threshold, the instance is rejected. In practice, the threshold can be defined as different levels of sensitivity by users to suit different authentication applications.

[0270] To recognize multiple users, the disclosed system can train a one-vs-all binary classifier for each user, with gait instances from the target user as the positive class and instances from all other candidates as the negative class. Given a test gait instance, the disclosed system can input it into all classifiers and obtain the matching probability that the instance belongs to each class. The gait instance may be assigned to the user with the classifier that observes the highest matching probability. In one embodiment, the disclosed system can use an SVM tool with a radial basis function (RBF) kernel. Optimal values ​​for parameters γ and c are selected by grid search using 10-fold cross-validation. Features may be scaled to [0,1] for classification.

[0271] In one embodiment, the rhythmic movement of an object, e.g., a human gait, can be monitored based on a time series of intermediate quantities (IQ) calculated by the disclosed system based on channel information of a wireless multipath channel. The wireless multipath channel is affected by the rhythmic movement of the object. In one embodiment, after a time series of velocity (IQ) is obtained, the time series of velocity is analyzed for a time window of stable walking (e.g., rhythmic movement). If there is a stable gait, the average (mean) velocity should not be too small and there should be a strong maximum in the autocorrelation function of the IQ. In another embodiment, peak detection is performed to calculate the maximum velocity value in the time window of the stable rhythmic movement of the object.

[0272] In one embodiment, a "step" segment of a time window is defined as the time from one local maximum to the next local maximum. The step segment may approximately correspond to the left foot touching the ground, and the next step segment may approximately correspond to the right foot touching the ground. The term "approximately corresponds" refers to the heel not necessarily touching or lifting off the ground at maximum velocity (IQ). Two consecutive step segments can be combined to define a gait cycle or movement cycle. In another embodiment, a local maximum can be replaced with a local minimum in this claim. In another embodiment, the time window for stable rhythmic movement may be adjusted, for example, by pruning / removing some IQs from both ends of the time window, because rhythmic movement may not be completely stable or steady-state at both ends of the time window. When a subject starts walking from a stationary state, they may still be in "acceleration" mode during the first few moments within the time window and may already be in "deceleration" mode during the last few moments of the time window. One way to identify these non-steady-state IQs may be an increase or decrease in the time difference between adjacent local maxima. In a steady state, the time difference should be somewhat stable. There are several features related to gait speed. Gait speed (e.g., walking speed) itself is important for health-related monitoring. However, gait speed may not be very useful for gait recognition (i.e., gait-based human recognition). Instead, analyses derived from gait speed, such as speed deviation, peak variance, valley variance, and harmonic ratio, may be more important. Gait or movement features may be related to step segments (e.g., half a movement cycle for bipedal animals, one-quarter of a movement cycle for quadrupedal animals, or 1 / N of a movement cycle for N-legged animals). As an N-legged animal moves forward, N legs sequentially have maximum forward (or positive) speed. In the movement cycle of an N-legged animal, IQ (speed) may have N local maxima. A step segment may include the period from one local maximum to the next local maximum or a phase-shifted version of that period.

[0273] For a bipedal human, a motion cycle has two step segments. An odd-numbered step segment may correspond to a left foot movement (or a left step in the motion cycle), and an even-numbered step segment may correspond to a right foot movement (or a right step in the motion cycle). As an example, each left step may correspond to the period from when the left foot touches the ground to when the right foot touches the ground. Each right step may correspond to the period from when the right foot touches the ground to when the left foot touches the ground. Alternatively, each left step may correspond to the period from when the left foot leaves the ground to when the right foot leaves the ground. Each right step may correspond to the period from when the right foot leaves the ground to when the left foot leaves the ground.

[0274] Some motion features are associated with strides or motion cycles. For quadrupedal animals, odd-numbered motion cycles may correspond to two legs, and even-numbered cycles may correspond to the other two legs. In galloping, for example, odd-numbered cycles may correspond to the two front legs, and even-numbered cycles may correspond to the two rear legs. In trotting, odd-numbered cycles may correspond to the left front leg and the right rear leg, and even-numbered cycles may correspond to the left rear leg and the right front leg. Some motion features are associated with N step segments (e.g., a tripedal device may have three step segments). A motion cycle may form a wavefront with three phases. Phase 1 (cycles 1, 4, 7, ...) may correspond to foot 1 (or foot 2 and foot 3). Phase 2 (cycles 2, 5, 8, ...) may correspond to foot 2 (or foot 1 and foot 3). Phase 3 (cycles 3, 6, 9, ...) may correspond to foot 3 (or foot 1 and foot 2).

[0275] 15 illustrates an example method 1500 for gait recognition according to some embodiments of the present disclosure. As shown in FIG. 15, first, in operation 1502, channel state information (CSI) of a wireless multipath channel is collected. The wireless multipath channel is affected by the gait of a moving human or animal. In operation 1504, the CSI is preprocessed. Based on the preprocessing, in operation 1506, a speed signal is maximized to achieve a maximum SNR, for example, based on MRC. In operation 1508, a walking speed is estimated based on the maximum SNR.

[0276] At act 1510, periods of stable walking are detected. At act 1512, gait cycles are estimated based on the periods of stable walking. Next, at act 1514, gait-related features are extracted from the gait cycles. At act 1516, a gait is recognized based on the extracted gait features. For example, a user ID of a user moving with the gait is determined by comparing the gait features with features stored in a database.

[0277] The disclosed system, GaitWay, can be implemented on commodity WiFi devices, allowing experiments to be conducted in a typical building with an area of ​​approximately 5,000 square feet. In one embodiment, different configurations can be explored by placing the WiFi Tx and Rx in different locations during multiple data collection sessions. In one embodiment, the disclosed system can have six different configurations, with the Tx and Rx mounted on stands approximately 1 meter high. In all configurations, the Tx and Rx are spaced 8 to 11 meters apart and shielded by one or more walls. Both the Tx and Rx are commercially available laptops equipped with commercially available WiFi network interface cards and unmodified omnidirectional chip antennas. The disclosed system can use a 5.8 GHz channel (channel 161 by default) with a 40 MHz bandwidth. Multiple WiFi devices coexist on the same channel.

[0278] In one embodiment, the disclosed system can collect gait instances from 11 subjects, five female and six male. During data collection, users walked freely and continuously in a natural gait. Users were also free to walk in any area. Some users read the news, played mobile games, or talked on the phone while walking. The experiment was conducted on four different days over a six-month period. The disclosed system collected data for two sessions at different times on each day. The disclosed system acquired data for a total of eight sessions, each with different settings. Users wore different clothing (summer clothing to autumn clothing) during the different data collection sessions. In each session, the disclosed system measured approximately 10 to 20 minutes of walking for each subject. All subjects involved in the data collection were approved by an IRB. The data was anonymized to protect privacy. The disclosed system collected approximately 1,030 minutes of walking data from 11 participants, from which it extracted approximately 970 minutes of walking (i.e., approximately 60 minutes when the subjects were outside the link's effective range and speed data was not captured). From these data, the disclosed system extracted 5,283 valid stable walking instances, which accounted for approximately 680 minutes, or approximately 67% of the total walking time. Furthermore, the effective proportion of data collection was limited because users frequently stopped, started walking, or changed direction, making these free-flowing walks unreliable for gait measurements. In practice, training data collection is more efficient if users are cooperative with gait measurement. Compared to many conventional techniques that require users to repeatedly walk a fixed route, the disclosed system, GaitWay, significantly simplifies the task and significantly improves the scale of gait collection.

[0279] The performance of three cases can be studied separately: single-user authentication, two-user distinction, special case of multi-user recognition, and general multi-user recognition. Following the above-mentioned gait recognition, the disclosed system can use the receiver's operating curve (ROC) for the false acceptance rate (FAR) and false rejection rate (FRR), the equal error rate (EER) on the ROC where FAR and FRR are equal to evaluate authentication, and the recognition rate (RR) for recognition evaluation. All the following results of the disclosed system, GaitWay, are obtained based on 10-fold verification.

[0280] To evaluate GaitWay's performance for single-user authentication, the disclosed system can test each subject in a dataset by using all other users' gait instances as a negative class. The disclosed system can shuffle the training and test data 10 times and present the combined results. In one embodiment, GaitWay achieves an EER of 12.58% when using 70% of the training gait instances. In one embodiment, performance degrades slightly when more sessions are included over time.

[0281] Because it is very common for two people to share an office or two residents to live in one apartment, it is interesting to study the special case of two users before evaluating multi-user recognition. This is particularly useful if the disclosed system can distinguish between two people. The disclosed system can perform binary classification for all subject pairs in the dataset. Precision and recall can be studied by treating one user in each group as the positive class and the other as the negative class. In one embodiment, GaitWay achieves excellent performance with an average precision of 94.84% and a recall of 95.21% for 55 user pairs when using 70% of the training data. Accuracy exceeding 90% can be achieved using only 20% of the training data. Using automated data collection / gait extraction, such a large amount of data can be easily collected in approximately 20 minutes of walking. This makes GaitWay easy to use for user enrollment.

[0282] Recognizing multiple users is more difficult than verifying or distinguishing between two people. In one embodiment, RR varies with gait variability when different sessions are included, and performance decreases as the number of sessions increases. This is because a user's gait speed can vary significantly over time. For example, in an experiment, one user's walking speed was approximately 0.8 m / s in one session and approximately 1.4 m / s in another session. Although GaitWay avoids using absolute speed, performance can still be affected by large changes in walking speed because the disclosed system is built solely on walking speed. Unless otherwise specified, the disclosed system, GaitWay, can use data from all eight sessions in the following evaluations.

[0283] Since the data is collected and extracted automatically, the duration and number of steps of each gait instance varies. Therefore, the disclosed system can analyze whether the length of the gait sample affects the recognition accuracy. The disclosed system can analyze the length distribution of all test gait instances. A subject's gait may change during a very long walk, but an instance that is too short will not comprehensively embody the gait features.

[0284] The disclosed system can analyze the impact of different sessions by examining the distribution of error sources across data from different sessions. In one embodiment, most data sessions show similar performance, but session #3 shows the lowest performance at only 42.19%. This is because the Tx-Rx link is heavily obstructed by two walls and a reinforced concrete main pillar measuring approximately 1m x 1m, making it difficult for the Rx to capture the scattered signal. Examining the velocity estimates from session #3 reveals that it is noisier than the other sessions.

[0285] To study the effect of the number of subjects, the results in FIG. 16 can be aggregated across all 2036 possible combinations of 11 subjects. Generally, the RR gradually decreases with an increase in the number of users involved. The disclosed system, GaitWay, maintains an excellent RR of over 80% when there are five subjects, demonstrating its promising potential for smart homes, which typically have only a few residents. In one embodiment, subjects with low RR have relatively larger FAR and FRR than other subjects, indicating that their gait patterns are less discriminable.

[0286] The disclosed system, GaitWay, can be run in real time on a personal computer. For one minute of data, the total processing time is approximately 27 seconds (20 seconds for velocity estimation, 6 seconds for stable period identification, and less than 1 second for feature extraction). The SVM takes 164 seconds and 2 seconds for training and testing, respectively, when using 1,000 instances. This cost is negligible because training can be done offline.

[0287] Gait / rhythmic movement recognition based on the disclosed systems and methods is particularly attractive for various ubiquitous applications requiring human identification because it can be achieved remotely and without the active cooperation of a user. For example, a smart building automatically opens its doors when an authorized user enters. A smart home changes the temperature and ambient light to match the recognized user. A smart TV in a smart home responds with the recognized user's favorite programs. Smart home devices such as Google Home and Amazon Alexa can directly interact with the recognized user in a more intimate way. According to various embodiments of the present disclosure, a user simply walks through a space as they normally would for all of this to work.

[0288] In one embodiment, the disclosed system can monitor rhythmic motion of an object based on channel information of a wireless multipath channel and trigger a responsive action based on the results of the monitoring. According to various embodiments, the object can be a human, and the monitored rhythmic motion can be a human gait, hand movement, or breathing. The monitoring can include identifying the object. Upon identifying a person, the disclosed system can trigger a responsive action, such as generating a tailored action and / or controlling a device based on settings associated with the person. The settings can be based on time / date, month / season, or daily / weekly / monthly / yearly routines. For example, the disclosed system may send signals to turn on / off or adjust lamps / lights / windows / window blinds / fans / TV / audio / vacuum cleaner settings, generate conversational greetings / recommendations / suggestions / reports / brief reports / date announcements, play music / radio on a smart speaker or user interface, adjust the channel / source of a radio / TV / streaming device / media player, preheat / start a coffee machine / oven / cooking device / car, prime a car / vehicle / garage, unlock a door, turn an alarm on / off, start / stop a system, set an air conditioner / heater to a particular temperature, power on a computer, heat / cook food, etc. based on preferred settings, favorite settings or time settings associated with the person.

[0289] In one example, the rhythmic movement may be a complex rhythmic movement (e.g., a human dance movement) formed from many simple coordinated movements (e.g., a human's head movement, neck movement, left / right hand movement, right / left wrist movement, left finger movement, left arm movement, right arm movement, left shoulder movement, right shoulder movement, hip movement, hip movement, left leg movement, left foot movement, right leg movement, etc.). A stride may correspond to a period or cycle of the complex rhythmic movement. A step segment may correspond to a simple movement within a stride (a "step" within a cycle of the complex movement). A stride may be formed by combining a series of N consecutive step segments. A stride may be broken down into N step segments. A step segment may correspond to a simple movement or no movement. In one embodiment, monitoring may include detecting some event / action, such as a person falling, a person acting quickly, a person dancing, playing, sitting, standing up, hopping, jumping, a good / bad mood / a particular emotional state, etc. Responses may include alerting a caregiver / manager / supervisor / police / doctor / emergency response service or team, recording / monitoring activity for health monitoring purposes, counting steps, determining whether daily activity / exercise goals are being met, or prompting the user to perform a preferred action (e.g., playing appropriate music while exercising) or environmental setting. In one embodiment, the object may be a dog and the rhythmic movement may be a gait. Response actions may include preparing dog food / drinks, water, locking / unlocking doors, adjusting temperature / windows / fans, etc.

[0290] In one embodiment, a system for monitoring rhythmic movement is described. The system includes a transmitter, a receiver, and a processor. The transmitter is configured to transmit a first wireless signal toward an object at a location through a wireless multipath channel of the location. The receiver is configured to receive a second wireless signal through the wireless multipath channel between the transmitter and the receiver. The second wireless signal differs from the first wireless signal because the wireless multipath channel is affected by the rhythmic movement of the object. The processor is configured to obtain time-series channel information (CI) of the wireless multipath channel based on the second wireless signal, monitor the rhythmic movement of the object based on the time-series CI (TSCI), and trigger a response action based on the monitoring results. According to various embodiments, the processor may be physically coupled to at least one of the transmitter and the receiver.

[0291] In one embodiment, the processor is further configured to calculate a time series of intermediate quantities (IQ) based on the TSCI and monitor rhythmic motion of the object based on the time series of IQ. In one embodiment, the system further includes an additional transmitter configured to transmit a third wireless signal through an additional wireless multipath channel and an additional receiver configured to receive a fourth wireless signal through the additional wireless multipath channel, where the fourth wireless signal differs from the third wireless signal because the additional wireless multipath channel is affected by rhythmic motion of the object at the location, and the processor is further configured to obtain an additional TSCI for the additional wireless multipath channel based on the fourth wireless signal, calculate an additional time series of IQ based on the additional TSCI, and monitor rhythmic motion based on the time series of IQ and the additional time series of IQ. In one embodiment, the additional transmitter is a transmitter and the third wireless signal is the first wireless signal. In another embodiment, the additional receiver is a receiver.

[0292] In one embodiment, the processor is further configured to calculate at least one local feature of IQ within a time window of stable rhythmic movement, the at least one local feature comprising at least one of a local maximum, a local minimum, a zero crossing, a local maximum of a derivative of IQ, a local minimum of a derivative, and a zero crossing of a derivative of IQ; and further configured to segment the time window into at least one step segment based on a time stamp associated with the at least one local feature of IQ, each step segment spanning a time associated with a local feature to another time associated with a next local feature, each movement cycle comprising N consecutive step segments, where N is a positive integer. In another embodiment, each of the at least one local feature is a local maximum.

[0293] The present disclosure further discloses a system for tracking targets or objects based on time reversal and massive MIMO in outdoor environments. Outdoor environments are known to have poor or insufficient multipath, making it difficult to apply time reversal. The present disclosure discloses an alternative method for implementing time reversal (TR) in outdoor environments: using massive MIMO antenna arrays to contribute additional multipath. In outdoor environments, multipath may not be abundant for each antenna, but with massive MIMO, the number of antennas is large, so combining multipaths can be abundant. Similar to the spatial "focusing ball" achieved by time reversal in indoor environments, a spatial "focusing beam" can be achieved by combining time reversal and massive MIMO. Based on this spatial focusing beam, the present disclosure discloses a high-precision target localization method suitable for outdoor environments. Extensive simulation results show that the disclosed system can achieve centimeter-level accuracy for outdoor localization and tracking.

[0294] One major advantage of the disclosed system over GPS-based tracking and navigation systems is NLOS operation. Because GPS requires direct line of sight with multiple GPS satellites in space, it does not operate in partially covered / obstructed areas such as football stadiums (e.g., open-topped Super Bowl stadiums), parking lots, downtown areas with dense tall buildings / skyscrapers (e.g., the Manhattan area of ​​New York City), forests with moderately dense trees / vegetation, mazes, canyons / valleys, etc. The disclosed system can operate equally well in both LOS and NLOS conditions. Thus, unlike GPS, the disclosed system can operate perfectly well in these areas where GPS has problems.

[0295] In the primary setting, wireless transmitters (Type 1 devices) are fixed devices (e.g., installed routers or access points, APs) with a large number of antennas (e.g., 21 or more), and wireless receivers (Type 2 devices) are mobile devices (e.g., smartphones, automobiles, AGVs) with at least one antenna. Mobile devices (Type 2 devices) are devices whose location is tracked.

[0296] In another embodiment, in an alternative configuration, the wireless transmitter (Type 1 device) is a fixed device (e.g., an installed router or access point, AP) having at least one antenna, and the wireless receiver (Type 2 device) is a mobile device (e.g., a smartphone, automobile, AGV) having a large number of antennas (e.g., 21 or more). The mobile device (Type 2 device) is a tracked device. In another embodiment, in an alternative configuration, the wireless receiver (Type 2 device) is a fixed device (e.g., an installed router or access point, AP) having a large number of antennas (e.g., 21 or more), and the wireless transmitter (Type 1 device) is a mobile device (e.g., a smartphone, automobile, AGV) having at least one antenna. The mobile device (Type 1 device) is a tracked device. In another embodiment, in an alternative configuration, the wireless receiver (Type 2 device) is a fixed device (e.g., an installed router or access point, AP) having at least one antenna, and the wireless transmitter (Type 1 device) is a mobile device (e.g., a smartphone, automobile, AGV) having a large number of antennas (e.g., 21 or more). A mobile device (Type 1 device) is a device that is tracked.

[0297] In another embodiment, instead of "distance", the "velocity" or "acceleration" of the current movement of the moving device is calculated. In yet another embodiment, instead of "distance", a function of "distance", "velocity" and / or "acceleration" is calculated. In one embodiment, "distance" may be based on autocorrelation focusing strength (ACFS).

[0298] The "first spatiotemporal information" or first STI may include at least one of distance, height, depth, displacement, location, velocity, acceleration, angle, angular velocity, angular acceleration, derivative, higher-order derivative, integral, direction, time, period, timing, time trend, incremental change in spatiotemporal information (STI), total change in STI, time-dependent change in STI, set of STI, and / or another spatiotemporal quantity. Similarly, the second STI may include at least one of distance, height, depth, displacement, location, velocity, acceleration, angle, angular velocity, angular acceleration, derivative, higher-order derivative, integral, direction, time, period, timing, time trend, incremental change in spatiotemporal information (STI), total change in STI, time-dependent change in STI, set of STI, and / or another spatiotemporal quantity. The second STI may be calculated based on the first STI. For example, the first STI may be distance, directional distance, or a projection of distance in a particular direction. In one embodiment, the first STI may be a distance in an orthogonal direction of an "axis." The axis connects the center of the MIMO antenna of the fixed device (e.g., base station) with the center of the antenna of the mobile device (or object). In another example, the mobile device may have MIMO antennas (e.g., a large number of antennas, such as 50, 100, 200, 1,000, or 10,000), and the fixed device may have a small number of antennas (e.g., 1, 2, 3, or 4).

[0299] The first STI may be calculated based on the "focused beam principle." That is, a CSI-based feature, such as time-reversal resonance intensity or autocorrelation focused intensity, may be a sinc function of distance in an orthogonal direction. The CSI-based feature may form a beam around an axis, and the beam intensity is a sinc function of radial distance. The CSI-based feature (e.g., ACFS) may be calculated first, and then the first STI (e.g., distance) may be calculated based on the CSI-based feature and the sinc function (e.g., using a table lookup or by calculating / estimating an inverse sinc function).

[0300] The ACFS may be calculated between a current CSI at time t and a past CSI at time tk (k is an integer). k may be varied from 1 to 2, 2 to 3, etc., to find k such that the ACFS between the CSI at time t and the CSI at time tk is a feature point of the ACFS (e.g., a minimum, a maximum, a first minimum, a first maximum, etc.).

[0301] The first STI may be calculated based on a trigonometric function of the angle between the axis of the MIMO antenna of the fixed device (e.g., a base station) and another axis (e.g., the trigonometric function may include at least one of sine, cosine, tangent, arcsine, arccosine, arctangent, secant, cosecant, cotangent, or other trigonometric function). For example, the MIMO antenna may be a one-dimensional array of evenly spaced antennas, and the other axis may be a line formed by the antenna array, which may connect the center of the antenna array of the fixed device to the mobile device. The first STI may be the distance traveled by the mobile device in an orthogonal direction between time tk and time t (e.g., a projection of the distance traveled by the mobile device onto the orthogonal direction). The first STI may be calculated based on the aperture of the MIMO antenna. If there are L antennas evenly spaced at d intervals on a line, the aperture may be (L-1)d. The antennas do not need to be evenly spaced. The MIMO antenna does not need to have a one-dimensional configuration. For example, the MIMO antenna may include several antenna groups, each group forming a line. The lines may or may not be coplanar. In another example, the MIMO antennas may be in a two-dimensional or three-dimensional grid. In another example, the MIMO antennas may be in pseudo-random positions within a 2D or 3D region. The first STI may be calculated based on the distance between the center of the antenna array and the mobile device (e.g., at time tk). The second STI may be a location. The location of the mobile device (second STI) (or the location of the object) may be calculated based on the first STI. The second STI may be calculated as a point on a line parallel to the axis and at a distance from the axis equal to the first STI. The first STI and the second STI may be calculated at different rates, such as 1 Hz, 2 Hz, 3 Hz, 5 Hz, 10 Hz, 20 Hz, 30 Hz, 50 Hz, 100 Hz, 200 Hz, 300 Hz, 500 Hz, 1000 Hz, 1500 Hz, 2000 Hz, 3000 Hz, 5000 Hz, 10000 Hz, etc. For example, the first STI may be calculated at 100 Hz and the second STI may be calculated at 1 Hz.

[0302] An "object" may be a human, an animal, another device, a vehicle, a machine, a movable structure, a structure / material subjected to forces (e.g., wind, earthquake, vibration, impact, stress, tension, buoyancy, fluid flow, etc.), a robotic device, an AGV. In another embodiment, the "distance" may be a first STI that is calculated and used to calculate a second STI.

[0303] TR has been proven effective in accurately locating targets even in environments with severe multipath signals. By further investigating the resonance phenomenon in the spatial domain, a highly accurate indoor velocity estimation system has been developed. Inspired by this promising property, this disclosure demonstrates that massive MIMO can collect spatially focused beams similar to TR in the long-distance case. Therefore, a highly accurate target location method is developed.

[0304] This disclosure further discloses object velocity estimation using focused beams of multiple distributed base stations with future 5G massive MIMO antennas. Furthermore, based on the velocity estimation results, a novel object localization method is disclosed. Extensive numerical simulations show that this novel method can achieve at least sub-meter accuracy in some extreme environments, but centimeter accuracy in some ideal conditions.

[0305] In one embodiment, the disclosed method has low complexity because the calculation of the ACFS distribution is linearly proportional to the received data size. A method is disclosed for estimating the moving direction based on the velocity estimation by further using BS location information. While most existing techniques, such as DOA estimation methods, can only estimate the local direction of an object, the disclosed method can obtain the absolute direction. Furthermore, the disclosed method can achieve an accuracy of less than 2 degrees in the presence of NLOS components, outperforming benchmark methods.

[0306] In another embodiment, some lower layer details (e.g., mode, operation, transmission, measurement, function, information, feedback) of a method / system / device may enable an upper layer application (e.g., software) to perform wireless sensing. Wireless sensing enablement (WSE) is for enabling wireless sensing. Assume that an application (e.g., software application, mobile app, embedded software, firmware, etc.) wants to perform wireless sensing based on some wireless data (e.g., WSE measurements) acquired at some lower layer. In this disclosure, WSE refers to anything (e.g., mode, function, wireless transmission, measurement, operation, information, feedback) at the physical (PHY) layer, medium access (MAC) layer, or other layer below the application layer of some Type 1 heterogeneous wireless device (transmitter or Tx) or Type 2 heterogeneous wireless device (receiver or RX) or another device (e.g., local server, cloud server) that enables the application to perform wireless sensing.

[0307] Navigation systems are widely used in modern applications, where estimating moving speed and direction are two crucial steps. Instead of estimating moving direction using attitude sensors based on traditional devices such as accelerometers and magnetometers, this disclosure discloses a novel radio frequency (RF) signal-based moving direction sensing method by using a 5G massive multiple-input multiple-output (MIMO) system. Herein, we create the energy distribution of received signals in massive MIMO for both short-range and long-range cases. Herein, we demonstrate that the energy distribution at short range is highly related to the geometry of the antenna arrangement. In contrast, at long range, we find that the energy distribution is a sinc-shaped focused beam that does not change. Inspired by this observation, we disclose a novel method for estimating the speed of a moving object relative to a single base station. By considering both the speed estimation result and the geometric characteristics of the location between the object and surrounding base stations, the moving direction can be further determined. Finally, numerical simulations demonstrate that the disclosed RF-based method can achieve high accuracy, with a moving speed estimation error of less than 1.5 m / s and a moving direction estimation error within 2 degrees.

[0308] This disclosure exploits time-reversal resonating strength (TRRS), which is proven to be a consistent and location-independent focusing ball-shaped distribution around the receiver. Leveraging this observation, a method for tracking objects with centimeter-level accuracy is disclosed and validated through extensive experiments.

[0309] In this disclosure, a massive MIMO system utilizes multiple antennas to physically generate multiple signal components that act similarly to multipath in a scattering-rich environment. Furthermore, the incident signal generated by massive MIMO can only reach the receiver from the transmitter side. In the long-distance case, it can be proven that the autocorrelation function strength (ACFS) distribution of the received signal around the receiver exhibits a sinc-like beam in the spatial domain, which can provide directional information. By further utilizing the dense deployment of 5G massive MIMO base stations (BSs), a novel radio frequency (RF) signal-based mobile speed / direction estimation method is disclosed. Furthermore, the natural overlapping property of the received signal is utilized to significantly reduce the computational load.

[0310] In one embodiment, the ACFS distribution of the received signal of a massive MIMO communication system is derived for both short-range and long-range cases. For short-range cases, the ACFS distribution is closely related to the geometric parameters of the antenna array, while for long-range cases, it shows an invariant sinc-like beam.

[0311] In one embodiment, considering the practical long-distance case, a moving speed estimation algorithm is developed by using the aforementioned ACFS distribution, which achieves a high accuracy of speed estimation error of less than 1.5 m / s. Because the ACFS distribution of the received signal is stable, the computational complexity for calculating the ACFS is linearly proportional to the received data size, and the disclosed speed estimation algorithm also has a low complexity.

[0312] In one embodiment, a method for estimating the moving direction based on the velocity estimation and further using BS location information is disclosed. Numerical simulations show that the method is environment independent and the moving direction estimation error is less than 2 degrees, which is superior to benchmark methods.

[0313] In one embodiment, a signal model involves a base station (BS) equipped with a massive MIMO array of M antennas communicating with a receiver fixed to a moving object. In a typical downlink system, the BS transmits a probe signal, which is recorded by the receiver. Figure 17 shows the communication system setup, where "B" and "R" represent the base station center and receiver, respectively. H B indicates the altitude of the BS, and L BR indicates the horizontal distance between the BS and the receiver. e is the aperture of antenna A.

[0314] If the inter-element spacing is d, the aperture can be expressed as Ae = (M-1)d, where d = λ is equal to the wavelength of the transmitted signal. Next, for long distances, It is reasonable that TIFF0007779954000019.tif1034 holds. Furthermore, all transducers can be assumed to be arranged with omnidirectional antennas. In unbounded free space, the received signal at baseband can be expressed as TIFF0007779954000020.tif17148TIFF0007779954000021.tif20147The composite phase distortion of the mth propagation path, including the initial phase and phase errors due to inter-system interference, propagation attenuation, reflectors, etc. In general, this composite phase distortion can be assumed to be iid uniformly distributed over [-π,π) for all m=1,2,...,M.

[0315] TR is a signal processing technique that attempts to utilize the channel state information (CSI) embedded in multipath signals. Considering scattering environments such as indoors and urban areas, there are usually many obstacles between the transmitter and receiver, and non-line-of-sight (NLOS) multipath propagation is inevitable. Given a sufficiently wide bandwidth, these multipath components (MPCs) can be decomposed into different taps in discrete time. Here, the channel impulse response (CIR) from transmitter T to receiver R at the kth tap can be expressed as h(k;T→R). Typically, in wireless communication systems, receiver R first transmits a pilot impulse, which is then acquired by transmitter T. In that case, the CIR, h(k;R→T), can be easily estimated by analyzing the relationship between the signal collected by the transmitter and the original pilot impulse. Then, transmitter T calculates the inverse conjugate version of the CIR, i.e., h * (-k; R → T), where * is the complex conjugate operation. If the experimentally verified channel reciprocity holds, then the position R S The received signal at is given by: TIFF0007779954000022.tif23137Since the transmitter T is usually fixed, from now on, h(k;T→R S ) to h(k;R S ) Furthermore, the length of the CIR, L, is related to the channel condition and the bandwidth of the transmitted signal. From equation (14), R S = R and k = 0 only This means that all MPCs are coherently summed at a precise location R and a specific time instance. At other locations or timestamps, or at different times and locations, the energy of the received signal will be attenuated differently. This is the so-called spatiotemporal focusing in a TR system.

[0316] In one embodiment, k=0 can be fixed to primarily study the TR focusing effect in the spatial domain. In that case, the velocity and location of the object can be estimated. To further quantify the TR focusing strength (TRFS), the location R SThe normalized energy of the received signal at can be defined as TIFF0007779954000024.tif23152

[0317] From the above discussion of TRFS, we can easily conclude that abundant multipath signals are one of the necessary conditions for achieving the TRFS effect. In the field of signal processing, TR methods use a wide bandwidth to resolve and control the multipaths that naturally exist in scattering-rich environments. However, in outdoor environments such as roads, squares, or outdoor parking lots, there are usually not enough multipath signals. Furthermore, practical communication systems are bandwidth-limited. Thanks to the rapidly growing massive MIMO technology, we can seek a better alternative that can achieve similar performance to TR systems for outdoor localization problems. Intuitively, massive MIMO systems utilize multiple antennas to physically generate numerous signal components that play similar roles to multipaths in scattering-rich environments. Instead of calculating TRFS, we can make a small change by calculating the autocorrelation function (ACF) of the received signal. This is because, due to the automatic averaging process, ACF is more resistant to noise in practical systems. Unlike TRFS, which is a ball, the ACF distribution in massive MIMO systems is a focused beam. To distinguish it from the previous TRFS, we can call it the autocorrelation focusing strength (ACFS), which is derived as follows:

[0318] Referring again to equation (13), two different timestamps t0 and t S The positions of the moving object in r0 and r S In this case, r0 and r S The autocorrelation function (ACF) of the received energy between TIFF0007779954000025.tif48143When the SNR is high, the Gaussian noise n(t) does not affect the ACF distribution of the received signal because it is signal independent. However, as the SNR decreases, n(t) will affect the ACF distribution to some extent, which will be interpreted in the simulation part. TIFF0007779954000026.tif30152However, a similar approximation cannot be applied to the numerator of equation (16) because of the additional wavenumber coefficient k = 2π / λ. In TIFF0007779954000027.tif151365G communication systems, the carrier frequency fc (which can be 28 GHz) is usually very high, so λ=1 / fc is very small. In the following derivation process, the denominator part of equation (16) can be considered as a constant for long distances, so it can be omitted for simplicity.

[0319] Next, we consider equation (16) in two different cases: TIFF0007779954000028.tif20123

[0320] TIFF0007779954000029.tif19121

[0321] From Figure 17, the coordinates of the mth transmitting antenna element are (md / 2,0, H B ), we obtain the following equation: TIFF0007779954000030.tif34148

[0322] TIFF0007779954000031.tif28149

[0323] For long distances, L BR is usually H B This is more than 10 times the amount. TIFF0007779954000032.tif14140As a result, the three-dimensional signal propagation geometry shown in FIG. 18A can be simplified as shown in FIG. 18B. TIFF0007779954000033.tif20144

[0324] As shown in Figure 18B, r0 and r S The Euclidean distance between is p. However, an electromagnetic wave is not just a scaler, but a vector containing both a modulus and a direction. From antenna propagation theory, r S The difference in propagation paths that ultimately affects the phase of the received signal at r0 (which acts as a reference point) is TIFF0007779954000034.tif30126

[0325] TIFF0007779954000035.tif21144

[0326] Inserting equation (23) into equation (22), we get TIFF0007779954000036.tif21169

[0327] Referring to equations (17) and (18), TIFF0007779954000037.tif22170

[0328] therefore, TIFF0007779954000038.tif23119

[0329] From the definition in equation (19), Xm=md / 2 and d is very small (aperture A e = Md / 2 and L). As a result, the summation in equation (26) can be approximated by an integral expressed as TIFF0007779954000039.tif43131TIFF0007779954000040.tif22124

[0330] TIFF0007779954000041.tif48166

[0331] Similar to the analysis of (16), the denominator and the corresponding constant term in (29) can be omitted. In that case, the (i,m) pair in (29) can be obtained from TIFF0007779954000042.tif22159

[0332] By exchanging the subscripts i and m, we obtain the symmetric (m,i) pair of equation (30): TIFF0007779954000043.tif21132

[0333] The sum of equations (30) and (31) simplifies to: TIFF0007779954000044.tif25140

[0334] TIFF0007779954000045.tif43169

[0335] In that case, the expectation value of equation (32) can be reformulated as follows: TIFF0007779954000046.tif19131

[0336] TIFF0007779954000047.tif51160

[0337] Similar to equations (35) and (36), TIFF0007779954000048.tif38161

[0338] Combining equations (16), (28) and (37), TIFF0007779954000049.tif34133

[0339] Position r0 and position r SBy calculating the ACF of the received signal between r and r, we can obtain the autocorrelation focusing strength (ACFS), which is similar to the TRFS in the time-reversal method. To verify this conclusion, we can build a numerical simulation system using a massive MIMO antenna array with 50 elements. We also consider a 5G communication system with a carrier frequency of f = 28 GHz. In Figure 19, r is set as the center of a square. The ACFS distribution around r in the spatial domain is shown in Figure 19A, and Figure 19B specifically shows the ACFS distribution along the cross-beam direction. In Figure 19B, the numerically simulated ACF obtained using Equation (16) is shown with a solid red line, and the derived ACF obtained using Equation (38) is shown with a solid blue line. Figure 19 intuitively shows that the derived ACFS matches well with the theoretical ACFS in terms of peak and valley locations.

[0340] If the object continues to move, In this case, the effective aperture Ae in equation (38) needs to be changed to Aecosβ. Correspondingly, the distance L needs to be replaced by L / cosβ. This variable transformation is also consistent with the definition of effective aperture used in antenna propagation and electromagnetics. If interested, you can also refer to equation (27) for a better understanding of this variable transformation process. Then, we can introduce a velocity estimation / target location method based on the obtained ACFS.

[0341] This disclosure discloses a novel method for estimating object velocity by using ACFS derived for massive MIMO systems. For clarity of explanation, we can start by defining two directions: along-beam direction and across-beam direction, as shown in Figure 19A. Furthermore, Figure 19C shows the peak distance p and the corresponding travel time T, which is the time it takes for the object to travel from the reference position (t=0) to the first peak (time index t).

[0342] Referring again to equation (38), the shape of the ACFS distribution (shown in FIG. 19A) is In other words, once an initial point is selected, L can be considered a constant when calculating ACFS in the neighborhood around the selected initial point. Therefore, the only parameter that determines the ACFS distribution with respect to the initial point is Consider the special case where the target moves at a constant velocity v along the direction intersecting the beam and the receiver fixed on the target keeps recording the signal transmitted by the massive MIMO array at a constant sampling rate. In this case, the ACFS measured at the receiver is given by TIFF0007779954000053.tif37164This is very close to the result p=1.43m obtained from the simulation experiment.

[0343] Then, to estimate the velocity v, we can obtain the travel time t for the receiver to move from the initial point to the first peak, which corresponds to the first peak of the ACFS distribution. This can be achieved by searching for TIFF0007779954000054.tif1256. TIFF0007779954000055.tif14150This is a key step in the velocity estimation method.

[0344] To obtain more robust results, we first employ a local regression method to fit the ACCF distribution curve. This allows us to find statistical peaks. This preprocessing is necessary because, in practice, the true peak may be corrupted by noise or other distortions. Therefore, directly searching for the peak may result in unexpected time estimation errors and significantly degrade speed estimation performance. The numerical simulation in Figure 21 shows that when the signal is corrupted, it is difficult to accurately find the true peak because there are many glitches in the ACF of the corrupted signal. However, after local regression processing, the estimated peak is very close to the actual peak, demonstrating the effectiveness of local regression processing. Meanwhile, due to the ACF distribution given by Equation (38), the true peak should not be too close to the reference point (t = 0). In practice, the moving speed is naturally limited (it should not be excessively fast). This means that the distance between the base station and the receiver center does not change significantly between two adjacent measurements. As a result, we can use the last position to obtain a rough estimate of the peak distance p. This serves as a new constraint to eliminate the obvious false peaks shown in Figure 21.

[0345] In the former special case, velocity estimation results can be easily obtained by assuming that the object moves along a direction intersecting the beam, but this assumption usually does not hold in practice. TIFF0007779954000056.tif14124If we can estimate the velocity in the same way as in the special case 1) above, what we actually get is an estimate of v', not the actual velocity v. This is because the peak distance p defined in Figure 19C is Therefore, in this case, we only have an estimate of v'. To solve this problem, we can introduce another base station, as shown in Figure 22. Combining the velocity estimation results from these two base stations, we obtain TIFF0007779954000058.tif1486

[0346] However, we cannot obtain the values ​​of θ1 and θ2 from Equation (39) alone. Further analysis of the positions of the base station centers B1 and B2 and the initial position r0 provides a promising method for solving θ1 and θ2. Specifically, B1, B2 and r o From the triangle relation, we can obtain the new equation (40). TIFF0007779954000059.tif14108

[0347] Since we have already obtained the equations consisting of Equation (39) and Equation (40), it seems that the problem has been completely solved. However, as shown in Figure 23, the ambiguity of the angle becomes a new problem. In detail, from Equation (39), the ratio TIFF0007779954000060.tif1936 but cannot determine the sum θ1 + θ2, so for every (θ1, θ2) pair there exists a coupling pair (θ1', θ2'). In other words, without further prior information, TIFF0007779954000061.tif9139For example, if it can be assumed that (θ1 + θ2) = (50°, 70°), the following set of coupling equations may be obtained: TIFF0007779954000062.tif27159TIFF0007779954000063.tif27159

[0348] Then, by introducing one more base station, as shown in Figure 24, the angle ambiguity problem can be solved first. To better understand the function of the third base station, one particular case can be taken as an example. In Figure 24, it can be assumed that (θ1, θ2, θ3) = (50°, 70°, 30°) is the true value. According to the above process, the following results can be obtained by just using base station 1 and base station 2: TIFF0007779954000064.tif25128

[0349] Similarly, another pair of values ​​can be obtained using base station 1 and base station 3. TIFF0007779954000065.tif24127

[0350] The true value of θ1 is unique. In other words, the value of θ1 obtained from equation (43) and equation (44) must be the same. Therefore, by comparing these two pairs of results, the value of θ1 can be accurately selected. Sequentially, θ2 and θ3 can be further determined.

[0351] Once the initial values ​​of (θ1, θ2, θ3) are obtained, a new constraint can be obtained, so that the third base station can be omitted in the next processing. Specifically, due to natural limitations on the moving speed and high sampling rate of the device (embedded in the receiver), the direction of one particular moving object cannot be changed abruptly between two adjacent locations. Figure 25 intuitively illustrates the new constraint. If the values ​​(θ1, θ2) = (50°, 70°) are known, then: TIFF0007779954000066.tif15114TIFF0007779954000067.tif15114TIFF0007779954000068.tif15114By using only two base stations, the angle ambiguity can be effectively eliminated, and as a result, the actual velocity v can be estimated.

[0352] As shown in Figure 26A, even if the initial angle (θ1, θ2) is known, it cannot be distinguished from the opposite vertical angle pair (θ1', θ2') shown in blue. In this case, simply adding an additional base station B3 is not effective. Figure 26B provides an intuitive explanation for why there is another opposite vertical angle θ3' to θ3. It is clear that this is a geometric ambiguity problem that does not depend on the number of base stations. This problem is solved by considering a continuous movement process shown in Figure 27. The sequence r0, r1, ... r n is the true position of the moving object, then two adjacent positions r i-1 , r i The peak distance between As a result, as the object moves closer to or further away from the reference base station B1, p (i-1,i)The essential reason comes from the ACFS distribution expressed in equation (38), which shows that the peak distance p is proportional to the distance L between the center of the base station and the receiver. From this point of view, p (i-1,i) The fluctuation trend (increase or decrease) of serves as a code to determine the moving direction. Note that the Doppler frequency of the received signal may be another way to determine the moving direction, since it increases when the object approaches the base station and decreases when the object moves away from the base station. Then, the location of the object can be estimated by the following method.

[0353] Once the exact value of θ1, the peak distance p1 and the sign (approaching or receding) of the base station (take base station 1 shown in Figure 28 as an example) are known, the coordinates of the new position r1 in x'Oy' Cartesian coordinates can be calculated as follows: TIFF0007779954000070.tif25102

[0354] Next, by a simple coordinate system transformation (r x1 ,r y1 ) can be converted to the absolute coordinate system xOy. TIFF0007779954000071.tif18108

[0355] As a result, we can obtain the new position of r1 relative to the xOy coordinate system with the base station center B1 as the origin. Since the base station location information in the 5G system is a priori, the exact position of r1 can also be easily calculated. By updating r1 as the new initial point and repeating the above location process, we can obtain a new position sequence r2, r3, ... r that appropriately tracks the target. n can be obtained.

[0356] Figure 30 shows three consecutive adjacent points r0, r1 and r2 along the movement trajectory. r0 is the initial point and Δθ is the direction change angle. From equations (39) and (40), we can obtain the estimated values ​​of θ1, θ2 and the true velocity v. Therefore, TIFF0007779954000072.tif1237 can be obtained, and the distance between the object and the BS can be obtained as follows: TIFF0007779954000073.tif9125

[0357] TIFF0007779954000074.tif21164TIFF0007779954000075.tif44166

[0358] Equations (39) and (40) can then be updated. Finally, TIFF0007779954000076.tif14142 may be estimated, respectively. In one embodiment, assuming r0 is the initial position and the object moves from r0 to r1, the main steps of the massive MIMO algorithm according to the disclosed method can be summarized as the following steps:

[0359] Step 1: Obtain the initial location information of base stations B1, B2 and r0, and then calculate the initial distance Calculate TIFF0007779954000077.tif1474.

[0360] Step 2: Calculate θ1+θ2 according to the location information of B1, B2, and r0. Because "sin(·)" means that θ1+θ2 is the same whether it is an acute angle or an obtuse angle, a third base station is needed in the first iteration to determine whether θ1+θ2 is an acute angle or an obtuse angle. However, after the first iteration, the previous θ1+θ2 can be used as an auxiliary constraint to help determine whether θ1+θ2 is an acute angle or an obtuse angle, so the third BS is no longer needed.

[0361] Step 3: Calculate ACFS based on the received signal to obtain peak widths d1 and d2.

[0362] Step 4: When the object moves from r0 to r1, calculate θ1, θ2 and the absolute movement speed v using the following equations (A1) and (A2). TIFF0007779954000078.tif25169

[0363] Step 5: Distance traveled p=v T win Calculate T winis a preset parameter.

[0364] Step 6: Once p, θ1 and θ2 are obtained, the location of r0 can be estimated by TIFF0007779954000079.tif26106

[0365] Step 7: Using the following formula (A4), (r x1 , r y1 ) into the absolute coordinate system xOy to obtain the location of r1. TIFF0007779954000080.tif18100

[0366] Step 8: Go back to Step 1 and Step 2 Update TIFF0007779954000081.tif1474 and θ1+θ2 and continue tracking the object.

[0367] Simulations are performed to verify the velocity / location estimation performance of the disclosed method using a future 5G communication system. The carrier frequency is set as fc = 28 GHz. The distance between the base station and the receiver is assumed to be within 200 m. The inter-element distance is set as the wavelength λ of the signal.

[0368] To investigate the impact of the number of antennas on velocity / location estimation performance, we performed extensive Monte Carlo simulations with a fixed SNR of 10 dB to obtain the root mean square error (RMSE) of the corresponding estimation results, as shown in Figures 29A-29C. As the number of antennas increases, the estimation accuracy of both velocity and location improves. Specifically, when the number of antennas is less than 100, it may not perform well when the moving speed is too fast (e.g., v = 30 m / s in Figures 29A-29C). However, when the number of antennas is 100 or more, the disclosed system can locate the target within 0.3 m accuracy for various speeds. However, when M is greater than a certain threshold, i.e., M = 200, this improvement in approximation becomes less obvious. Note that when the number of antennas M approaches 400, the location estimation error is as small as 8 cm, which in other words indicates centimeter-level accuracy.

[0369] The disclosed method works particularly well when the number of antennas is greater than 50. Intuitively, it is not possible to collect enough signal components to observe the so-called TRFS phenomenon. Mathematically, if M is too small, the approximation in equation (27) does not hold. The disclosed outdoor tracking is a new candidate for outdoor object localization when 5G-based massive MIMO deployments are applicable and when GPS satellite signals are blocked by buildings in urban areas. The disclosed system can guide pedestrians to their destinations with centimeter-level accuracy in typical metropolitan areas where roads are densely populated with high-rise buildings that prevent GPS from functioning reliably. The disclosed system can also estimate a user's walking speed, which can be used for health monitoring. The collected walking information of pedestrians in a specific area can also be used to analyze geographical behavior and activity.

[0370] In one embodiment, some lower-layer details (e.g., mode, operation, transmission, measurement, function, information, feedback) of a method / system / device may enable an upper-layer application (e.g., software) to perform wireless sensing. WSE means enabling wireless sensing. Assume that an application (e.g., software application, mobile app, embedded software, firmware, etc.) wants to perform wireless sensing based on some wireless data (e.g., WSE measurements) acquired at some lower layer. In this disclosure, WSE refers to anything (e.g., mode, function, wireless transmission, measurement, operation, information, feedback) at the physical (PHY) layer, medium access (MAC) layer, or other layer below the application layer of some Type 1 heterogeneous wireless device (transmitter or Tx) or Type 2 heterogeneous wireless device (receiver or RX) or another device (e.g., local server, cloud server) that enables the application to perform wireless sensing.

[0371] WSE mode is a mode tha...

Claims

1. 1. A method for a certified wireless system, comprising: transmitting a wireless signal from a Type 1 device to a Type 2 device based on a protocol or a standardized protocol over a wireless multipath channel at a location, wherein the Type 1 device and the Type 2 device are heterogeneous wireless devices; the Type 2 device receiving the wireless signal based on the protocol or the standardized protocol; acquiring time-series channel information (TSCI) of the wireless multipath channel based on the received wireless signal based on the protocol or the standardized protocol, the TSCI including first channel information (CI) acquired by the Type 2 device at a first time during a test phase of wireless monitoring and second CI acquired by the Type 2 device at a second time during the test phase of wireless monitoring, the test phase being a period for testing for certification; With respect to the wireless monitoring-related tasks and certification tests, performing the certification test based on the TSCI for devices to be certified to determine whether each of the devices to be certified is a certified device or a non-certified device, wherein the devices to be certified are at least one of the Type 1 device, a module of the Type 1 device, an integrated circuit (IC) of the Type 1 device, the Type 2 device, a module of the Type 2 device, or an IC of the Type 2 device; To obtain at least one certified device, determining that each of the devices to obtain the at least one certification is a certified device based on a determination that a respective certification criterion associated with the device to obtain the certification has been satisfied, the respective certification criterion comprising: a similarity score between features of the first CI and the features of a predictor greater than a first threshold, the predictor being a prediction of the first CI based on the second CI, the first CI and the second CI being obtained at the first time and the second time, respectively, during the testing phase of the wireless monitoring; determining that the difference between the first time and the second time is less than a second threshold; and performing the task based on the TSCI using the at least one certified device.

2. 10. The method of claim 1, and associating an identification (ID) with at least one of the Type 1 device for which certification is to be obtained or the Type 2 device for which certification is to be obtained, the ID comprising: Name, number, alphanumeric ID, string of text, multiple numbers, symbol, file, database, database entry, index to entry, link to web page, link to storage device, MAC address, IP address, network address, network ID, domain ID, web ID, internet ID, mobile network ID, LAN ID, platform ID, software ID, software application ID, management ID, monitoring ID, hardware ID, device ID, device profile, hardware component ID, computer ID, processor ID, storage device ID, process ID, serial number, ID, class, class information, category, category information, performance information, capability information, policy information, pair ID of said type 1 device and said type 2 device, pair profile, link ID, link profile, antenna ID, antenna profile, system ID, User, Customer, Supervisor, Superuser, Administrator, Monitor, Service, Account, Password, Service Account, User Account, User Profile, User Name, User's Password, User Information, User ID, Service Provider, Service Profile, Manufacturer Information, Sales Channel, Vendor, Retailer, Logistics Channel, Content Channel, Apple ID, Amazon ID, Samsung ID, Google ID, Facebook ID, Microsoft ID, Company ID, Service ID, Service Provider ID, Service ID, Access ID, Hash of Other IDs, User Association, User Grouping, Account Permissions, User History, Task, Task ID, Task Information, Task Requirements, Users Associated with the Task, 10. The method of claim 10, further comprising at least one of: a location ID, a physical address, a physical location, a home, a household, an office, a business, a school, a warehouse, a store, a factory, a station, a stadium, a hall, a campus, a place, a site, a region, a zone, an area, a range, a proximity location, a neighborhood, a map, a map location, location-based information, a street, a city, a county, a state, a province, a district, a prefecture, a country, a continent, a zip code, a postal code, a GPS coordinate, other code, a phone number, a payment card information, a grouping, a rating, a category, or other ID.

3. 3. The method of claim 1 or 2, and exchanging information between the Type 1 device and the Type 2 device based on the protocol or one of the standardized protocols, the information comprising: metadata, device information, manufacturing information, model information, version information, registration information, and identification information of at least one of the Type 1 device, the Type 2 device, or a device communicatively connected to the Type 1 device or the Type 2 device; Classification, category, grouping, constraints, usage information, service provider information, service information, sales information, logistic information, System information, companion system information, capacity information, power information, compute information, processor information, storage information, Supported system information, supported task information, performance requirements, carrier frequency, frequency information, timing information, antenna information, location-based information, Software or firmware information, update information, hardware information, component information, The method includes at least one of network information, wireless network information, address information, access information, security information, encryption information, Internet information, and other information.

4. 4. The method according to any one of claims 1 to 3, The method further includes determining a set of agreed-upon settings between the Type 1 device and the Type 2 device based on the standardized protocol.

5. 5. The method according to any one of claims 1 to 4, receiving the wireless signal based on the IC of the Type 2 device; calculating the TSCI based on the IC; obtaining the TSCI from the IC, wherein the qualification criteria associated with the IC of the Type 2 device include: Estimation error requirements for each CI, an estimation error requirement for the TSCI; At least one of the following requirements calculated based on the TSCI: motion statistics, motion characteristics, intermediate analysis, or task analysis; The precision requirements for each CI, the accuracy requirements of the TSCI; The TSCI display requirements; format requirements of the TSCI; coding requirements for the TSCI; the real-time requirements of said TSCI; the buffering requirements of said TSCI; the memory requirements of the TSCI; stability requirements of the TSCI; a consistency requirement for the TSCI; a time consistency requirement of the TSCI; the timing requirements of the TSCI; correlation requirements for the TSCI; The TSCI outlier requirement; deviation requirements of the TSCI; the sensitivity requirement of the TSCI; The TSCI tail requirement; percentile requirements of said TSCI; quantile requirements of the TSCI; scalability requirements of the TSCI; an indicator conveying the availability of at least one of the CI, a group of recent CIs, and all of the TSCIs; or accuracy requirements for calculation of a plurality of successive CIs; monitoring movement of objects at the location based on the TSCI.

6. 6. The method according to claim 1, wherein each of the certification criteria comprises: the location requirements; a requirement for the number of antennas on said Type 1 device; or a requirement on the number of antennas of the Type 2 device.

7. 7. The method of claim 1, wherein each of the certification criteria includes a location requirement, the location requirement comprising: a location type requirement, including at least one of indoor, outdoor, semi-outdoor, underground, house, office, building, warehouse, laboratory, or special testing facility; a partition type requirement, including at least one of plaster wall, drywall, fiberboard, siding, plaster, wood, metal, vinyl, stucco, shingles, asphalt, brick, stone, masonry, concrete, cement, tile, ceramic tile, or glass; Location size requirements, including at least one of volume, area, width, length, height, depth, thickness, or layers; Structural requirements including at least one of furniture, support structures, columns, beams, tables, chairs, shelves, cupboards, or vehicles; a multipath richness requirement of said location; the state requirements of said location; movement requirements of said location; the constituent elements of said location; said location remains unchanged at least temporarily; said location is at least temporarily stable; said location is at least temporarily immobile; the location is temporarily free of moving objects; The requirements for the wireless multipath channel are: a bandwidth requirement including at least one of 10 MHz, 20 MHz, 30 MHz, 40 MHz, 50 MHz, 60 MHz, 70 MHz, 80 MHz, 100 MHz, 160 MHz, or 320 MHz; a carrier frequency requirement based on at least one of an ISM band, a mobile band, a mobile channel, 3G, 4G, LTE, 5G, 6G, 7G, a WiFi band or a WiFi channel centered approximately on at least one of 6.78 MHz, 13.56 MHz, 27.12 MHz, 40.68 MHz, 4.5 GHz, 33.93 MHz, 915 GHz, 2.45 GHz, 5 GHz, 5.8 GHz, 24.125 GHz, 61.25 GHz, 122.5 GHz, and 245 GHz; Standard compliance requirements including at least one of WLAN, WiFi, 802.11 standard, 802.15 standard, 802.20 standard, mobile communication standard, 3GPP standard, 3G, 4G, LTE, 5G, 6G, 7G, 8G, Bluetooth standard, standard using OFDM, standard including calculation of said CI; protocol requirements, network requirements, signaling requirements, signal handshaking requirements, Multiple access requirements, channel traffic requirements, channel availability requirements, frequency hopping requirements, data transmission requirements; The requirements for the radio signal are: the wireless signal includes a time series of probe signals (TSPS); protocol requirements, network requirements, signaling requirements, signal handshaking requirements, Multiple access requirements, channel traffic requirements, modulation requirements, frequency hopping requirements, data transmission requirements, a transmit power requirement for the TSPS; a probe frequency requirement for said TSPS; probe timing requirements of said TSPS; a sounding frequency requirement for the TSPS; sounding timing requirements of the TSPS; the timing requirements of the TSPS; the rapid firing timing requirements of said TSPS; the pulse timing requirements of the TSPS; the progressive timing requirements of the TSPS; the time-varying timing requirements of the TSPS; the timing jitter requirements of the TSPS; the broadcast requirements associated with each probe signal; the signaling requirements associated with each probe signal; the protocol requirements associated with each probe signal; the handshaking requirements associated with each probe signal; a requirement for a precursor signal to trigger said Type 1 device to transmit a probe signal; the requirement that the probe signal be a preceding signal that is an acknowledgment in the handshake; The requirement for a preceding signal for the probe signal to be the response in the handshake; Data field requirements of a probe signal that allows acquisition of CI by said Type 2 device; Header field requirements for a probe signal that causes the acquisition of CI by the Type 2 device; requirements for a data field of a packet of a probe signal that causes the acquisition of CI by the Type 2 device; or a control data field requirement of a probe signal that causes the acquisition of CI by the Type 2 device; The requirements for the Type 1 device are: placement requirements, installation requirements, processor requirements, memory requirements, Software requirements, System requirements, power requirements, interface requirements, transmission requirements, housing requirements, signaling requirements, environmental requirements, Antenna type requirements, Antenna count requirements, antenna gain requirements, antenna placement requirements, antenna radiation requirements, Antenna material requirements, or antenna structural requirements; The requirements for the Type 2 device are: placement requirements, installation requirements, processor requirements, memory requirements, Software requirements, System requirements, power requirements, interface requirements, transmission requirements, housing requirements, signaling requirements, environmental requirements, Antenna type requirements, Antenna count requirements, antenna gain requirements, antenna placement requirements, antenna radiation requirements, Antenna material requirements, and antenna structure requirements.

8. 8. The method according to claim 1, wherein the certification criteria are: The first CI and the second CI are similar; There is little variation between the first CI and the second CI; The first CI is in the vicinity of the second CI; a variation score between the first CI and the second CI is less than a threshold; a distance score between the first CI and the second CI is less than a threshold; the first CI and a predictor of the first CI based on the second CI are similar; there is little variation between the first CI and the predictor; the first CI being a neighborhood of the predictor; a similarity score between the first CI and the predictor is greater than a threshold; a variability score between the first CI and the predictor is less than a threshold; a distance score between the first CI and the predictor is less than a threshold; The characteristics of the first CI and the characteristics of the second CI are similar; the features of the first CI are in proximity to the features of the second CI; a similarity score between the feature of the first CI and the feature of the second CI is greater than a threshold; a variation score between the feature of the first CI and the feature of the second CI is less than a threshold; a distance score between the feature of the first CI and the feature of the second CI is less than a threshold; the predictor of the first CI based on the features of the first CI and the second CI is similar; the features of the first CI are neighborhoods of the features of the predictor; a variation score between the feature of the first CI and the feature of the predictor is less than a threshold; a distance score between the feature of the first CI and the feature of the predictor is less than a threshold; the first time and the second time are similar; the first time is near the second time; The first CI and the second CI are neighboring CIs of the TSCI; The method, wherein the first time and the second time further include at least one of a neighboring sampling time of the Type 2 device.

9. 9. The method of any one of claims 1 to 8, the wireless signal comprises a time series of probe signals (TSPS); a first probe signal transmitted by the Type 1 device at a first transmission time is associated with the first CI; a second probe signal transmitted by the Type 1 device at a second transmission time is associated with the second CI; The certification criteria are: the first transmission time and the second transmission time are similar; there is little variation between the first transmission time and the second transmission time; the first transmission time is close to the second transmission time; the first probe signal and the second probe signal are adjacent probe signals in a time sequence of probe signals; the first transmission time and the second transmission time are proximity transmission times of the Type 1 device; a difference between the first transmission time and the second transmission time is less than a threshold; two of the first transmission time, the first time, the second transmission time, or the second time are similar; There is little variation between the two times. The two times are close to each other, the difference between the two times is less than a threshold.

10. 10. The method of any one of claims 1 to 9, There are multiple CIs in the TSCI for a certain period of time, The certification criteria are: The plurality of CIs are similar; the plurality of CIs match; There is little variation between the multiple CIs; There are few outliers among the plurality of CIs; all of the plurality of CIs are in the vicinity; all of the plurality of CIs are in a dense cluster; the similarity scores of the plurality of CIs are greater than a threshold; the variation scores of the plurality of CIs are less than a threshold; the distance scores of the plurality of CIs are less than a threshold; the similarity score of the plurality of CIs is greater than a threshold for at least a certain percentage of the time period; the variability scores of the plurality of CIs are less than a threshold value for at least a certain percentage of the time period; the distance scores of the plurality of CIs are less than a threshold for at least a certain percentage of the time period; the similarity scores of the plurality of CIs are less than a threshold for a maximum percentage of the time period; the variability scores of the plurality of CIs are greater than a threshold for a maximum percentage of the time period; the distance scores of the plurality of CIs are greater than a threshold for a maximum percentage of the time period; the average pairwise similarity scores of the plurality of CIs is greater than a threshold; the median of the pairwise similarity scores of the plurality of CIs is greater than a threshold; the mode of the pairwise similarity scores of the plurality of CIs is greater than a threshold; a percentile value of the pairwise similarity scores of the plurality of CIs is greater than a threshold; the minimum pairwise similarity score of the plurality of CIs is greater than a threshold; a weighted average of the pairwise similarity scores of the plurality of CIs is greater than a threshold; the variance of the pairwise similarity scores of the plurality of CIs is less than a threshold; an average of the pairwise similarity scores of the plurality of CIs from the centroids of the plurality of CIs is greater than a threshold; the median of the pairwise similarity scores of the plurality of CIs from the centroid is greater than a threshold; the mode of the pairwise similarity scores of the plurality of CIs from the centroid is greater than a threshold; a percentile value of the pairwise similarity scores of the plurality of CIs from a centroid is greater than a threshold; the minimum pairwise similarity score of the plurality of CIs from the centroid is greater than a threshold; a weighted average of the pairwise similarity scores of the plurality of CIs from the centroid is greater than a threshold; the variance of the pairwise similarity scores of the plurality of CIs from a centroid is less than a threshold; the variance of the plurality of CIs is less than a threshold; the average distance of the plurality of CIs from the centroid is less than a threshold; the central moments of the plurality of CIs are less than a threshold; the kurtosis of the plurality of CIs is less than a threshold; a tailedness measure of the plurality of CIs is less than a threshold; the measurements of the plurality of outliers are less than a threshold; The characteristics of each of the plurality of CIs are similar; the characteristics of each of the plurality of CIs match; There is little variation in the characteristics among the plurality of CIs. There are few outliers of the feature among the plurality of CIs. the features of the plurality of CIs are adjacent; The features of the plurality of CIs are in dense clusters, the similarity scores of the features of the plurality of CIs are greater than a threshold; the variability scores of the features of the plurality of CIs are less than a threshold; the distance scores of the features of the plurality of CIs are less than a threshold; the feature similarity scores of the plurality of CIs are greater than a threshold for at least a certain percentage of the time period; a variability score for the feature of the plurality of CIs is less than a threshold value for at least a certain percentage of the time period; a distance score of the feature of the plurality of CIs is less than a threshold for at least a certain percentage of the time period; the feature similarity scores of the plurality of CIs are less than a threshold for a maximum percentage of the time period; the variability scores of the features of the plurality of CIs are greater than a threshold value for a maximum percentage of the time period; the distance scores of the features of the plurality of CIs are greater than a threshold for a maximum percentage of the time period; an average pairwise similarity score of the features of the plurality of CIs is greater than a threshold; the median of the pairwise similarity scores of the features of the plurality of CIs is greater than a threshold; the mode of the pairwise similarity scores of the features of the plurality of CIs is greater than a threshold; a percentile value of the pairwise similarity scores of the features of the plurality of CIs is greater than a threshold; a minimum pairwise similarity score of the features of the plurality of CIs is greater than a threshold; a weighted average of the pairwise similarity scores of the features of the plurality of CIs is greater than a threshold; the variance of the pairwise similarity scores of the features of the plurality of CIs is less than a threshold; an average of the pairwise similarity scores of the features of the plurality of CIs from the centroids of the plurality of CIs is greater than a threshold; the median of the pairwise similarity scores of the features of the plurality of CIs from a centroid is greater than a threshold; the mode of the pairwise similarity scores of the features of the plurality of CIs from the centroid is greater than a threshold; a percentile value of the pairwise similarity scores of the features of the plurality of CIs from a centroid is greater than a threshold; the minimum pairwise similarity score of the features of the plurality of CIs from a centroid is greater than a threshold; a weighted average of the pairwise similarity scores of the features of the plurality of CIs from a centroid is greater than a threshold; the variance of the pairwise similarity scores of the features of the plurality of CIs from a centroid is less than a threshold; the variance of the features of the plurality of CIs is less than a threshold; an average distance of the features of the plurality of CIs from a centroid is less than a threshold; the central moments of the features of the plurality of CIs are less than a threshold; the kurtosis of the features of the plurality of CIs is less than a threshold; the tail length measure of the feature of the plurality of CIs is less than a threshold; or an outlier measure of the feature of the plurality of CIs is less than a threshold.

11. 11. The method of any one of claims 1 to 10, monitoring a movement of an object at the location based on the TSCI, wherein the wireless multipath channel is affected by the movement of the object at the location; calculating characteristics of the object's movement based on the TSCI, the characteristics including at least one of: frequency of repetitive movement, frequency characteristics, vital characteristics, respiratory rate, heart rate, frequency spectrum, period of repetitive movement, temporal characteristics, temporal profile, time, timing, start time, end time, duration, history, trend, prediction, type of movement, movement characteristics, movement intensity, movement measure, movement classification, identification, presence, proximity, count, number of people, position, geometry, speed, velocity, displacement, distance, range, direction, angle, acceleration, rotational speed, rotational characteristics, gait cycle of the object, gesture, transient behavior of the object, transient movement, change, change in the movement, change in frequency, change in period, change in gait cycle, event, sudden movement, and fall event; calculating a current characteristic of the motion of the object based on at least one of the TSCI and a past characteristic of the motion of the object; calculating the current characteristic of the motion of the object based on at least one of the current window of the TSCI and the past characteristic; calculating the historical characteristics of the motion of the object based on at least one of the TSCI historical windows; or generating a presentation related to the task in a user interface (UI) of a user device, wherein the task is performed based on the characteristics of the movement of the object, the task comprising: Object detection, presence detection, proximity detection, object recognition, activity recognition, object verification, object counting, Daily activity monitor, health monitor, vital signs monitor, health condition monitor, baby monitor, elderly monitor, sleep monitor, sleep stage monitor, walking monitor, exercise monitor, Tool detection, tool recognition, tool verification, Patient detection, patient monitor, patient verification, Machine detection, machine recognition, machine verification, Human detection, human recognition, human verification, Baby detection, baby recognition, baby verification, Human breathing detection, human breathing recognition, human breathing estimation, human breathing verification, Human heartbeat detection, human heartbeat recognition, human heartbeat estimation, human heartbeat verification, Fall detection, fall recognition, fall estimation, fall verification, Emotion detection, emotion recognition, emotion estimation, emotion verification, Motion detection, motion degree estimation, motion recognition, motion estimation, motion verification, Periodic motion detection, Periodic motion recognition, Periodic motion estimation, Periodic motion verification, Repeated motion detection, Repeated motion recognition, Repeated motion estimation, Repeated motion verification, Steady motion detection, Steady motion recognition, Steady motion estimation, Steady motion verification, Cyclostationary motion detection, cyclostationary motion recognition, cyclostationary motion estimation, cyclostationary motion verification, Instantaneous motion detection, instantaneous motion recognition, instantaneous motion estimation, instantaneous motion verification, Trend detection, trend recognition, trend estimation, trend verification, Breath detection, breath recognition, breath estimation, breath verification, Human biometric detection, Human biometric recognition, Human biometric estimation, Human biometric verification, Environmental information detection, environmental information recognition, environmental information estimation, environmental information verification, Gait detection, gait recognition, gait estimation, gait verification, Gesture detection, gesture recognition, gesture estimation, gesture verification, Machine learning, supervised learning, unsupervised learning, semi-supervised learning, clustering, Feature extraction, feature training, Principal component analysis, eigenvalue decomposition, frequency decomposition, time decomposition, time-frequency decomposition, fractional decomposition, other decompositions, Training, discriminative training, supervised learning, unsupervised learning, semi-supervised learning, neural networks, Sudden movement detection, fall detection, danger detection, life-threatening detection, regular movement detection, steady movement detection, periodic steady movement detection, Intrusion detection, suspicious activity detection, security, safety monitoring, Navigation, Guidance, Map-Based Processing, Map-Based Correction, Irregularity Detection, Localization, Indoor Sensing, Tracking, Multiple Object Tracking, Indoor Tracking, Indoor Positioning, Indoor Navigation, Energy management, power transmission, wireless power transmission, object counting, car tracking in parking lots, geometric estimation, augmented reality, A method comprising at least one of wireless communication, data communication, signal broadcasting, networking, coordination, management, encryption, protection, or cloud computing.

12. 12. The method of any one of claims 1 to 11, acquiring a second TSCI of a second wireless multipath channel of the location based on a second wireless signal transmitted from a second Type 1 device and received by a second Type 2 device, wherein one of the second Type 1 device and the second Type 2 device is a certified device and the other device is a device seeking certification; conducting a second qualification test based on the second TSCI; determining that the device to obtain the certification is a certified device based on a determination that second certification criteria associated with the second certification test are satisfied.

13. 13. The method of claim 12, associating the certified wireless system database with at least one of the device to obtain the certification, the certified device between the second Type 1 device and the second Type 2 device, the location, the second wireless signal, the second wireless multipath channel, or at least one task associated with at least one of the second certification test or the second certification standard; Certifying the device to receive said certification as a certified device; and certifying a device that is to obtain the certification as the certified device with respect to at least one associated task of the second certification test or the second certification standard.

14. 13. The method of claim 12, The method further includes obtaining a third TSCI of a third wireless multipath channel at the location based on a third wireless signal transmitted between the device to obtain the certification determined to be the certified device and an additional certified device.

15. 15. The method of claim 14, the certified device is one of the Type 1 device or the Type 2 device; The method, wherein the additional certified device is the certified device between the second Type 1 device and the second Type 2 device.

16. 13. The method of claim 12, receiving a request related to a device for which said certification is to be obtained; determining, based on the TSCI, that the location is in appropriate test conditions for the second certification test, where both the Type 1 device and the Type 2 device are certified; adjusting the second Type 1 device to transmit the second wireless signal on the second wireless multipath channel at the location and adjusting the second Type 2 device to receive the second wireless signal; The request is a request from a device intending to obtain said certification to said certified device; a request from the device for which the certification is to be obtained to at least one of the Type 1 device, the Type 2 device, a processor of the Type 2 device, or a memory of the Type 2 device; a request from the device to obtain said certification to a server communicatively coupled to said certified device; a request for certification of the device for which said certification is to be obtained; A requirement that the device to receive said certification be certified as a certified Type 1 device; a request to certify the device to be certified as a certified Type 1 device for at least one second task; A requirement that the device seeking said certification be certified as a certified Type 2 device; a request to certify the device to be certified as a certified Type 2 device for the at least one second task; A requirement that the device seeking such certification be certified as both a certified Type 1 device and a certified Type 2 device; a request that the device to obtain said certification be certified as both a certified Type 1 device and a certified Type 2 device for said at least one second task; a request that the at least one second task be initiated on the device for which the certification is to be obtained; a request for the device to obtain said certification to participate in said at least one second task; a request to conduct said second certification test for the device to obtain said certification; a request to perform the second certification test related to at least one second task on the device to be certified; A request to add the device to be certified to a group of certified devices; a request to add the device to be certified to a group of certified devices associated with at least one second task; a request to add the device to be certified to a group of certified devices associated with at least one of Type 1 devices or Type 2 devices associated with the at least one second task; a request to transmit and a request to receive the second wireless signal at the location for at least one of the second qualification test or the at least one second task; requesting use of the second wireless multipath channel at the location for at least one of the second qualification test or the at least one second task; a request to perform transmission and reception of the second wireless signal on the second wireless multipath channel for at least one of the second qualification test or the at least one second task; or another request; The suitable test conditions are: no detectable motion is detected in the TSCI-based motion detection test; a low level of motion in the TSCI-based motion detection test; No detectable change is detected in the TSCI-based change detection test; In the steady-state test based on the TSCI, the wireless multipath channel is in a steady state; The wireless multipath channel is stable in the TSCI-based stability test; the channel traffic of the wireless multipath channel is low; the traffic noise level of the wireless multipath channel is low; a movement of a target object associated with at least one of the second qualification test or a task associated with the second qualification test is detected based on the TSCI; The movement of the target object is predicted based on the TSCI. The movement of the target object is monitored normally based on the TSCI. A repeatable object movement of the target is detected based on the TSCI; or and another condition of the channel based on the TSCI.

17. 17. The method of any one of claims 1 to 16, Obtaining a second TSCI of a second wireless multipath channel of the location based on a second wireless signal transmitted from a second Type 1 device and received by a second Type 2 device, both of which are devices seeking certification; and conducting a second qualification test based on the second TSCI; and determining that both of the devices intended to obtain certification are certified based on a determination that second certification criteria associated with the second certification test are satisfied.

18. 1. A certified radio system comprising: a Type 1 device configured to transmit a wireless signal based on a protocol or a standardized protocol to a Type 2 device over a wireless multipath channel at a location, wherein the Type 1 device and the Type 2 device are heterogeneous wireless devices; and The Type 2 device, receiving the wireless signal based on the protocol or the standardized protocol; a Type 2 device configured to acquire time-series channel information (TSCI) of the wireless multipath channel based on the received wireless signal in accordance with the protocol or the standardized protocol, the TSCI including first channel information (CI) acquired by the Type 2 device at a first time during a test phase of wireless monitoring and second CI acquired by the Type 2 device at a second time during the test phase of wireless monitoring; an authorized device communicatively connected to the Type 2 device, performing a certification test on devices to be certified for at least one certification to determine whether each of the devices to be certified for at least one certification is a certified device or a non-certified device for a task related to wireless monitoring based on the TSCI, wherein the devices to be certified are at least one of the Type 1 device, a module of the Type 1 device, an integrated circuit (IC) of the Type 1 device, the Type 2 device, a module of the Type 2 device, or an IC of the Type 2 device; configured to determine that each of the devices intended to obtain the at least one certification is a certified device based on a determination that the device has satisfied a respective certification criterion associated with the device intended to obtain the certification, the respective certification criterion comprising: a similarity score between features of the first CI and the features of a predictor is greater than a first threshold, the predictor is a prediction of the first CI based on the second CI, the first CI and the second CI being obtained at a first time and a second time, respectively, during a testing phase of the wireless monitoring; and and a certified device, wherein the difference between the first time and the second time is less than a second threshold, and the task is performed based on the TSCI using at least one certified device.

19. 20. The system of claim 18, the wireless signal is received by the Type 2 device based on the IC of the Type 2 device; the TSCI is calculated based on the IC of the Type 2 device; the TSCI is obtained from the IC of the Type 2 device; The qualification criteria associated with the IC of the Type 2 device include: Estimation error requirements for each CI, an estimation error requirement for the TSCI; At least one requirement of motion statistics, motion characteristics, intermediate analysis or task analysis calculated based on the TSCI; The precision requirements for each CI, the accuracy requirements of the TSCI; The TSCI display requirements; format requirements of the TSCI; coding requirements for the TSCI; the real-time requirements of said TSCI; the buffering requirements of said TSCI; the memory requirements of the TSCI; stability requirements of the TSCI; a consistency requirement for the TSCI; a time consistency requirement of the TSCI; the timing requirements of the TSCI; correlation requirements for the TSCI; The TSCI outlier requirement; deviation requirements of the TSCI; the sensitivity requirement of the TSCI; The TSCI tail requirement; percentile requirements of said TSCI; quantile requirements of the TSCI; scalability requirements of the TSCI; an indicator conveying the availability of at least one of the TSCIs, a group of recent CIs, or all of the TSCIs; or at least one of accuracy requirements for the calculation of a plurality of successive CIs; A system that monitors movement of objects at the location based on the TSCI.

20. 20. A system according to claim 18 or 19, comprising: The certification criteria are: The first CI and the second CI are similar; There is little variation between the first CI and the second CI; The first CI is in the vicinity of the second CI; a variation score between the first CI and the second CI is less than a threshold; a distance score between the first CI and the second CI is less than a threshold; the first CI and a predictor of the first CI based on the second CI are similar; there is little variation between the first CI and the predictor; the first CI being a neighborhood of the predictor; a similarity score between the first CI and the predictor is greater than a threshold; a variability score between the first CI and the predictor is less than a threshold; a distance score between the first CI and the predictor is less than a threshold; The characteristics of the first CI and the characteristics of the second CI are similar; the features of the first CI are in proximity to the features of the second CI; a similarity score between the feature of the first CI and the feature of the second CI is greater than a threshold; a variation score between the feature of the first CI and the feature of the second CI is less than a threshold; a distance score between the feature of the first CI and the feature of the second CI is less than a threshold; The features of the first CI and the features of the predictor of the first CI based on the second CI are similar; the features of the first CI are neighborhoods of the features of the predictor; a variation score between the feature of the first CI and the feature of the predictor is less than a threshold; a distance score between the feature of the first CI and the feature of the predictor is less than a threshold; the first time and the second time are similar; the first time is near the second time; The first CI and the second CI are neighboring CIs of the TSCI, or The system, wherein the first time and the second time further include at least one of a neighboring sampling time of the Type 2 device.

Images