Method, apparatus and system for object tracking and navigation

Abstract

A method, apparatus and system for object tracking are disclosed. The method detects at least one time-series channel information (CI) of a wireless multipath channel affected by a current movement of an object tracked within a location from wireless signals transmitted over the wireless multipath channel between a Type 1 heterogeneous wireless device at a first position within the location and a Type 2 heterogeneous wireless device at a second position within the location. The method also includes continuously calculating a distance of the current movement of the object based on a similarity score between a pair of temporally adjacent CIs of at least one TSCI associated with the current movement of the object, and monitoring the spatiotemporal information of the object based on the distance and on at least one of the at least one TSCI, a time parameter associated with the current movement, and past spatiotemporal information of the object.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application incorporates by reference the entire disclosure of each of the following cases and claims priority to each of the following: (a) U.S. Provisional Patent Application No. 62 / 593,826, entitled "Method, Apparatus and System for Object Tracking and Navigation," filed December 1, 2017. (b) U.S. Provisional Patent Application No. 62 / 678,207, entitled "Method, Apparatus and System for Object Tracking and Motion Monitoring," filed May 30, 2018. (c) U.S. Patent Application Serial No. 15 / 861,422, entitled "Method, Apparatus, Server, and System for Time Reversal Technology," filed January 3, 2018. (d) U.S. Patent Application No. 15 / 873,806, entitled "Method, Apparatus, and System for Object Tracking and Navigation," filed January 17, 2018. (e) U.S. Patent Application No. 16 / 101,444, entitled "Method, Apparatus and System for Wireless Motion Monitoring," filed August 11, 2018. (f) U.S. Provisional Application No. 62 / 734,224, entitled "Method, Apparatus and System for Wireless Sleep Monitoring," filed September 20, 2018. (g) U.S. Provisional Application No. 62 / 744,093, entitled "Method, Apparatus and System for Wireless Proximity and Presence Monitor," filed October 10, 2018. (h) U.S. Provisional Patent Application No. 62 / 753,017, entitled "Method, Apparatus and System for Person Identification Based on Human Wireless Biometrics," filed October 30, 2018. (i) U.S. Patent Application Serial No. 16 / 200,608, entitled "Method, Apparatus, Server and System for Vital Signs Detection and Monitoring," filed November 26, 2018. (j) U.S. Patent Application Serial No. 16 / 200,616, entitled "Method, Apparatus, Server and System for Real-Time Vital Signs Detection and Monitoring," filed November 26, 2018. (k) U.S. Patent Application Serial No. 16 / 203,299, entitled "Apparatus, System, and Method for Wireless Signal-Based Event Recognition," filed November 28, 2018. (l) U.S. Patent Application Serial No. 16 / 203,317, entitled "Devices, Systems, and Methods for Wireless Signal-Based Fall Detection," filed November 28, 2018.

[0002] The present teachings relate generally to object tracking, and more particularly to object tracking and monitoring in richly cluttered environments based on time reversal techniques. [Background technology]

[0003] Indoor location-based services are becoming increasingly important today with respect to object tracking and monitoring. One common approach is to use dead reckoning to estimate the position of a moving object in real time. Typically, the direction and distance traveled are estimated by an inertial measurement unit (IMU). However, the performance of distance traveled in dead reckoning-based approaches is far from satisfactory, which is the main reason why such indoor navigation systems have not yet become widespread. Estimating the speed of a moving object in an indoor environment that can support location-based services is also an open problem, and satisfactory results have not yet emerged. The Doppler effect has been widely applied to various speed estimation systems using sound waves, microwaves, or laser light. However, low speeds, such as the walking speed of a human, are very difficult to estimate using Doppler shift, especially using electromagnetic (EM) waves. This is because the maximum Doppler shift is approximately Δ f = v / cf0, where f0 is the carrier frequency of the transmitted signal, c is the speed of light, and v is the human walking speed. For a normal human walking speed of v = 5.0 km / h and f0 = 5.8 GHz, Δ fis approximately 26.85 Hz, and it is extremely difficult to estimate this small quantity with high accuracy. In addition, these methods require line-of-sight (LOS) conditions and do not work well in complex indoor environments with rich multipath reflections.

[0004] Because direct-path signals are disrupted by multipath signals in indoor environments, and the arrival time (or Doppler shift) of direct-path signals cannot be accurately estimated, most existing speed estimation methods that work well in outdoor environments cannot provide satisfactory performance in indoor environments. Researchers have focused on estimating the maximum Doppler frequency that can be used to estimate mobile speed. Various methods have been proposed, including the level-crossing rate method, covariance-based methods, and wavelet-based methods. However, these estimators do not provide satisfactory results because the statistics used in these estimators have large variances and are location-dependent in real-world scenarios. For example, the accuracy of one existing speed estimation method can only distinguish between a mobile terminal moving at a fast speed (over 30 km / h) and a slow speed (under 5 km / h).

[0005] Another type of indoor speed estimation method based on the traditional pedestrian dead reckoning algorithm is to use an accelerometer to detect steps and estimate stride length. However, pedestrians often have different stride lengths, which can vary by up to 40% at the same speed and up to 50% at different speeds for the same person. Therefore, calibration is required to obtain an average stride length for different individuals, which is impractical in practical applications and therefore has not been widely adopted. Currently, object motion detection is becoming increasingly important. For example, for security and / or management purposes, a user may want to detect the movement of any object in the user's home, a supermarket manager may want to detect the movement of any object in the supermarket, or a hospital nurse may want to detect the movement of any patient in the hospital.

[0006] Existing systems and methods for detecting object movement fail to provide sufficient accuracy and often lead to false alarms. Existing approaches include passive infrared (PIR), active infrared (AIR), and ultrasonic-based approaches. PIR sensors are the most widely used motion sensors in home security systems, detecting human movement by sensing the difference between the heat emitted by moving people and the background heat. However, PIR sensor-based solutions are prone to false alarms due to their sensitivity to environmental changes such as hot / cold air currents and sunlight. They are easily failed by blocking body heat emissions (such as wearing an insulated full-body suit). They also have limited range and require line of sight (LOS), necessitating the use of multiple devices. In AIR-based approaches, an IR emitter transmits an IR beam that is received by an IR receiver. When the beam is interrupted, motion is detected. However, this type of approach is easily visible using a conventional camera or IR detection mechanism, has limited range, and requires line of sight. Ultrasonic sensors detect human movement by sending ultrasonic waves into space and measuring their return velocity, detecting any frequency changes. However, such approaches are also failed by wearing anechoic clothing. Also, ultrasound cannot penetrate solid objects such as furniture or boxes, creating gaps in the detection field. Even slow movements by a thief may not trigger an alarm. Therefore, existing systems and methods for detecting object movement are not entirely satisfactory. Summary of the Invention

[0007] The present teachings generally relate to object tracking. More specifically, the present disclosure relates to time-reversal based object tracking and monitoring in rich cluttered environments, either indoors or in outdoor locations with barriers, such as urban metropolitan areas, closed environments, underground environments, parking lots, warehouses, gardens, squares, forests, caves, valleys, etc.

[0008] In one embodiment, a method for an object tracking system includes acquiring one or more 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. At least one or more time-series channel information (TSCIs) are extracted from wireless signals transmitted over the wireless multipath channel between a type 1 heterogeneous wireless device at a first position within a location and a type 2 heterogeneous wireless device at a second position within the location. The wireless multipath channel is affected by a current movement of an object within the location. The method also determines spatiotemporal information of the object based on at least one of the at least one TSCI, a time parameter associated with the current movement, and past spatiotemporal information (info) of the object. The one or more TSCIs are preprocessed. The spatiotemporal information (info) includes location, horizontal location, vertical location, length, area, volume, volume, distance, direction, displacement, speed, velocity, acceleration, rotational velocity, rotational acceleration, gait cycle, motion type, motion classification, motion characteristics, sudden motion, transient motion, periodic motion, period of periodic motion, frequency of periodic motion, transient motion, time trend, time stamp, duration, time window, sliding time window, history, frequency trend, spatiotemporal trend, spatiotemporal change, event and / or other information, etc. Computations associated with the method are shared among the processor, the Type 1 heterogeneous wireless device, and the Type 2 heterogeneous wireless device.

[0009] The distance of the object's current movement may be determined (e.g., by the object tracking server, the processor, the type 1 heterogeneous device, the type 2 heterogeneous device, and / or another device) based on one or more time series of the CIs. An estimated direction of the object's current movement may be obtained. Spatiotemporal information of the object may be determined based on the distance and / or the estimated direction of the object's current movement. At least one similarity score may be calculated (e.g., by the object tracking server, the processor, the type 1 heterogeneous device, the type 2 heterogeneous device, and / or another device). Each similarity score may be based on a pair of temporally adjacent CIs (e.g., 111A, 111B) of the TSCIs associated with the object's current movement. A characteristic similarity score may be calculated based on the at least one similarity score. The distance of the object's current movement may be determined based on comparing the characteristic similarity score with a reference decay curve. The spatiotemporal information of the object may be determined based on the distance of the object's current movement.

[0010] At least one latest CI may be determined (e.g., by the object tracking server, the processor, the Type 1 heterogeneous device, the Type 2 heterogeneous device, and / or another device). Each latest CI (e.g., 111A, 111B) may be the latest in one of the one or more TSCIs. One or more time series similarity scores may be calculated. Each similarity score may be calculated based on two CIs of a particular time series associated with the similarity score. The two CIs may be the latest CI and a temporally adjacent CI within a time window associated with the current movement of the object. At least one curve may be calculated. Each curve may be determined based on the similarity scores of the time series. At least one feature point of the curve may be identified. Spatiotemporal information of the object may be determined / calculated based on the at least one feature point. The at least one feature point may include a local maximum, a local minimum, a first local maximum, a second local maximum, another local maximum, a first local minimum, a second local minimum, another local minimum, a zero crossing, a first zero crossing, a second zero crossing, another zero crossing, a point having a predetermined relationship with the second feature point, and / or another feature point. At least one second similarity score may be calculated (e.g., by the object tracking server, the processor, the type 1 heterogeneous device, the type 2 heterogeneous device, and / or another device). Each similarity score may be calculated based on an initial CI and a current CI. The initial CI may be temporally close to the start of the current movement. The current CI may be temporally close to the end of the current movement or the current time. A characteristic second similarity score may be determined based on the at least one second similarity score. If the characteristic second similarity score is greater than (or "greater than," "less than," or "less than") the threshold, the object may be determined to be stationary, and the current motion may be determined to be zero motion, small motion, and / or negligible motion.

[0011] Initial spatiotemporal information of the object at an initial time before the object's current movement may be determined (e.g., by the object tracking server, the processor, the type 1 heterogeneous device, the type 2 heterogeneous device, and / or the like) based on other time series of other CIs associated with the object's previous movement. Another time series of another CI may be extracted from a second wireless signal transmitted through a wireless multipath channel between a second type 1 heterogeneous device at a third position (e.g., 408) within the location and a second type 2 heterogeneous device at a fourth position within the location. The wireless multipath channel may be affected by the object's previous movement. The initial spatiotemporal information of the object at the initial time may be determined to be a known location associated with the second type 1 heterogeneous device and / or the second type 2 heterogeneous device. A time period may be determined. Within the time period, corresponding CIs of essentially all (e.g., all but a few, or all but outliers) of other time series of other CIs may be greater than the first adaptive threshold and less than the second adaptive threshold. The initial time may be determined as a feature point of the time period. The spatiotemporal information of the object may be determined based on the initial spatiotemporal information of the object. The characteristic points of the period may be the midpoint, quartile points, percentile points, start points, end points, points at which other CIs (e.g., 111A, 111B) are local maxima, points at which other CIs are local minima, points at which other CIs have particular properties, and / or other characteristic points.

[0012] At an initial point in time during the object's previous movement, one of the second type 1 heterogeneous device and the second type 2 heterogeneous device may be spatially close to the object during the object's previous movement and move with the object. At the initial point in time, the other of the second type 1 heterogeneous device and the second type 2 heterogeneous device may have a directional antenna. Initial spatiotemporal information of the object at the initial point in time may be determined (e.g., by the object tracking server, the processor, the type 1 heterogeneous device, the type 2 heterogeneous device, and / or another device) to be a known location associated with the other of the second type 1 heterogeneous device and the second type 2 heterogeneous device that has a directional antenna.

[0013] One of the first Type 1 heterogeneous device and the first Type 2 heterogeneous device may be spatially proximate to the object and move with the object during movement of the object. One of the first Type 1 heterogeneous device and the first Type 2 heterogeneous device may be communicatively coupled to the network server and / or to a local device that may be communicatively coupled to the network server. The local device may be a smartphone, a smart device, a smart speaker, a smart watch, smart glasses, a smart clock, a smart TV, a smart oven, a smart refrigerator, a smart air conditioner, a smart device, a chair, a smart table, a smart accessory, a smart utility, a smart appliance, a smart machine, a smart vehicle, an Internet of Things (IoT) device, an Internet-enabled device, a computer, a portable computer, a tablet, a smart house, a smart office, a smart building, a smart parking lot, a smart system, and / or other device.

[0014] One or more time series of power information (PI) may be calculated (e.g., by the object tracking server, a processor, a type 1 heterogeneous device, a type 2 heterogeneous device, and / or another device) based on one or more TSCI. Each PI may be associated with a CI. The PI may be a complex number. The real part of the power information (PI) may be calculated based on the magnitude, magnitude squared, phase, real part, imaginary part, and other functions of the CI. A first function of the one or more time series of PI may be calculated. The first function may include calculation of a function such as an autocorrelation function, a squared autocorrelation function, an autocovariance function, a squared autocovariance function, an inner product, an autocorrelation-like function, and / or a covariance function.

[0015] In another embodiment, the object tracking server includes a processor, a memory communicatively connected to the processor, and a set of instructions stored in the memory. When the set of instructions is executed by the processor using the memory, the object tracking server is configured to acquire one or more TSCIs of wireless multipath. The at least one TSCI is extracted from a wireless signal transmitted over a wireless multipath channel between a type 1 heterogeneous wireless device at a first position within the location and a type 2 heterogeneous wireless device at a second position (e.g., 406) within the location. The wireless multipath channel is affected by a current movement of an object within the location. The object tracking server determines the at least one TSCI based on at least one of the at least one TSCI, a time parameter associated with the current movement, and past spatiotemporal information (info) of the object. The one or more TSCIs are preprocessed.

[0016] In yet another embodiment, a system of object tracking servers includes a Type 1 heterogeneous wireless device at a first location within the venue, a Type 2 heterogeneous wireless device at a second location (e.g., 406) within the venue connectable with the Type 1 heterogeneous device over a wireless multipath channel, and / or an object tracking server, the object tracking server comprising a processor, a memory communicatively coupled to the processor, and a set of instructions stored in the memory.

[0017] The object tracking server, the type 1 heterogeneous device, and the type 2 heterogeneous device are configured to acquire at least one TSCI of a wireless multipath channel. The at least one TSCI is extracted from a wireless signal transmitted between the type 1 heterogeneous device and the type 2 heterogeneous wireless device over the wireless multipath channel. The wireless multipath channel is affected by a current movement of an object within the location. Spatiotemporal information of the object is determined based on the at least one TSCI, a time parameter associated with the current movement, and / or past spatiotemporal information of the object.

[0018] Other concepts relate to software for implementing the present teachings for object tracking and monitoring (including periodic object motion monitoring) based on time reversal techniques in sufficiently scattering environments. Other novel features are 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 and the accompanying drawings, or may be learned by the manufacture or operation of the embodiments. The novel features of the present teachings may be realized or attained by practice or use of various aspects of the methodologies, instrumentalities and combinations described in the detailed embodiments discussed below.

[0019] The present teachings generally relate to object motion detection and monitoring. More specifically, the present disclosure relates to wireless CI-based object tracking and monitoring in clutter-rich environments, such as indoor environments or outdoor locations with barriers, such as urban metropolitan areas, closed environments, underground environments, parking lots, warehouses, gardens, squares, forests, caves, valleys, etc.

[0020] In one embodiment, a system is disclosed having at least one processor and a memory having stored therein a set of instructions for detecting object movement within a location, the system comprising: a first wireless device configured to transmit a wireless signal over a wireless multipath channel impacted by object movement within the location; a second wireless device of a different type from the first wireless device configured to receive the wireless signal over the wireless multipath channel impacted by object movement within the location and to obtain a time series correlation (TSCI) of the wireless multipath channel based on the wireless signal; and a motion detector configured to detect object movement within the location based on motion information related to the object movement, wherein the motion information associated with the first and second wireless devices is calculated based on the TSCI by at least one of the motion detector and the second wireless device.

[0021] In yet another embodiment, a wireless monitoring system method is disclosed. The wireless monitoring system includes at least one asynchronous heterogeneous Type 1 wireless device and at least one asynchronous heterogeneous Type 2 wireless device. The wireless monitoring system includes at least one pair of Type 1 and Type 2 devices. Each pair includes one of the at least one asynchronous heterogeneous Type 1 wireless device and one of the at least one asynchronous heterogeneous Type 2 wireless device. Each of the at least one Type 2 wireless device is in a respective specific pair of at least one Type 1 and Type 2 device. Each Type 2 wireless device is associated with at least one respective specific Type 1 wireless device via at least one respective specific pair of Type 1 and Type 2 devices. For each Type 2 wireless device and for each of at least one respective specific pair of Type 1 and Type 2 devices comprising a Type 2 wireless device, the method includes asynchronously obtaining at least one respective TSCI of a wireless multipath channel impacted by object movement within a location, wherein the at least one respective TSCI (TSCI) associated with the particular pair of Type 1 and Type 2 wireless devices is asynchronously extracted from a respective asynchronous heterogeneous wireless signal, and the respective asynchronous heterogeneous wireless signal is asynchronously transmitted from a respective one of the at least one respective specific Type 1 wireless device to at least one asynchronous heterogeneous Type 2 wireless device over the wireless multipath channel using a respective first heterogeneous processor, a respective first heterogeneous memory, and a respective first heterogeneous instruction set of the Type 1 wireless device.The method further includes at least one of using a third processor, a third memory, and a third instruction set of the Type 3 device to individually and asynchronously monitor movement of objects within a location based on TSCIs associated with a pair of Type 1 and Type 2 devices that includes a particular Type 2 device; jointly and asynchronously monitoring movement of objects based on TSCIs associated with any of at least one pair of Type 1 and Type 2 devices associated with a particular Type 2 wireless device; jointly and asynchronously monitoring movement of objects based on TSCIs associated with any of at least one pair of Type 1 and Type 2 devices associated with one of at least one respective particular Type 1 wireless devices; and collectively and asynchronously monitoring movement of objects based on TSCIs associated with any of at least one pair of Type 1 and Type 2 devices.

[0022] In yet another embodiment, a specific asynchronous heterogeneous Type 2 device of a wireless monitoring system is disclosed. The wireless monitoring system includes at least one asynchronous heterogeneous Type 1 device and at least one asynchronous heterogeneous Type 2 device. The wireless monitoring system includes at least one pair of Type 1 and Type 2 devices, each pair including at least one Type 1 device and at least one Type 2 device. The specific asynchronous heterogeneous Type 2 device is at least one specific pair of Type 1 and Type 2 devices of the wireless monitoring system. The specific Type 2 device is grouped with at least one specific Type 1 device through at least one specific pairing of Type 1 and Type 2 devices.a second heterogeneous processor communicatively coupled to the second heterogeneous processor; and a second heterogeneous memory communicatively coupled to the second heterogeneous processor, and a second heterogeneous instruction set stored in the second heterogeneous memory, which, when executed, causes the second heterogeneous processor to asynchronously protect at least one respective set of CIs (TSCIs) of the wireless multipath channel for each of at least one particular pairing of Type 1 and Type 2 devices, and to protect at least one respective set of CIs (TSCIs) of the wireless multipath channel for each of at least one particular pairing of Type 1 and Type 2 devices, and to protect at least one respective set of CIs (TSCIs) of the wireless multipath channel for each of at least one particular pairing of Type 1 and Type 2 devices. and a second heterogeneous instruction set stored in a second heterogeneous memory, the TSCIs being asynchronously derived from the respective asynchronous heterogeneous radio signals asynchronously received by the radio circuit, wherein the movement of the mass within the site is monitored individually and asynchronously based on TSCIs associated with pairings of Type 1 and Type 2 devices that include a particular Type 2 device using a third processor, third memory, and third instruction set of the Type 3 device; the movement of the mass is monitored jointly and asynchronously based on TSCIs associated with any pairings of Type 1 and Type 2 devices that include one of the at least one particular Type 1 device; and the movement of the mass is monitored collectively and asynchronously based on TSCIs associated with any pairings of Type 1 and Type 2 devices.

[0023] In yet another embodiment, a specific asynchronous heterogeneous Type 1 device of a wireless monitoring system is disclosed. The wireless monitoring system includes at least one asynchronous heterogeneous Type 1 device and at least one asynchronous heterogeneous Type 2 device. The wireless monitoring system includes at least one combination of Type 1 and Type 2 devices, each combination including at least one Type 1 device and at least one Type 2 device. The specific asynchronous heterogeneous Type 1 device is at least one specific combination of Type 1 and Type 2 devices of the wireless monitoring system. The specific Type 1 device is paired with at least one specific Type 2 device via at least one specific combination of Type 1 and Type 2 devices. The specific asynchronous heterogeneous Type 1 device includes a radio circuit, a first processor communicatively coupled to the radio circuit, a first memory communicatively coupled to the first processor, and a first instruction set stored in the first memory. The first set of instructions, when executed, causes the first processor to use the radio circuitry to asynchronously transmit asynchronous heterogeneous radio signals from the specific type 1 device to at least one specific type 2 device over a wireless multipath channel affected by material movement within the site, and for each of the at least one specific type 2 device, at least one series of CIs (TSCIs) of the wireless multipath channel collected from the asynchronous heterogeneous radio signals transmitted by the radio circuitry is fetched by the type 2 device using a second processor, a second memory, and a second set of instructions of the type 2 device.The movement of material within the site may include at least one of: using a third processor, a third memory, and a third instruction set of the Type 3 device, the movement of material within the site may be tracked individually and asynchronously based on TSCIs associated with any combination of Type 1 and Type 2 devices, including a specific Type 1 device; the movement of material may be tracked jointly and asynchronously based on TSCIs associated with any combination of Type 1 and Type 2 devices, including one of at least one specific Type 2 device; and the movement of material may be tracked collectively and asynchronously based on TSCIs associated with any combination of Type 1 and Type 2 devices.

[0024] In yet another embodiment, a particular asynchronous heterogeneous Type 3 device of a wireless monitoring system is disclosed. The wireless monitoring system includes at least one asynchronous heterogeneous Type 1 device and at least one asynchronous heterogeneous Type 2 device. The wireless monitoring system includes at least one doublet of Type 1 and Type 2 devices, each doublet including one of the at least one asynchronous heterogeneous Type 1 device and one of the at least one asynchronous heterogeneous Type 2 device. The particular asynchronous heterogeneous Type 3 device includes a third processor communicatively coupled to at least one of the at least one asynchronous heterogeneous Type 1 device and the at least one asynchronous heterogeneous Type 2 device, a third memory communicatively coupled to the third processor, and a third instruction set stored in the third memory.The third instruction set, when executed, causes the third processor, for each Type 2 device and for each of at least one doublet of Type 1 and Type 2 devices that includes the Type 2 device, to asynchronously receive, using a respective second processor, a respective second memory, and a respective second instruction set of the Type 2 device, at least one respective TSCI (TSCI) of a wireless multipath channel affected by movement of an item at a location that is asynchronously received by the Type 2 device, wherein the at least one respective TSCI associated with each doublet of Type 1 and Type 2 devices is transmitted over the wireless multipath channel from a respective Type 1 device of the respective doublet to the at least one asynchronous heterogeneous Type 2 device using a respective first processor, a respective first memory, and a respective first instruction set of the respective Type 1 device. and causing the third processor to at least one of individually and asynchronously track the movement of an item at a location based on TSCIs associated with a particular doublet of Type 1 and Type 2 devices, including a particular Type 2 device and a particular Type 1 device; jointly and asynchronously track the movement of an item based on TSCIs associated with any of at least one doublet of Type 1 and Type 2 devices associated with a particular Type 2 device; jointly and asynchronously track the movement of an item based on TSCIs associated with any of at least one doublet of Type 1 and Type 2 devices associated with a particular Type 1 device; and collectively and asynchronously track the movement of an item based on TSCIs associated with any of at least one doublet of Type 1 and Type 2 devices. [Brief explanation of the drawings]

[0025] [Figure 1]1 illustrates an exemplary environment for determining a first position of an object using wireless signals (eg, 140), in accordance with one embodiment of the present teachings.

[0026] [Figure 2] A diagram showing multiple links each formed by one antenna of a wireless transmitter (e.g., 108) having at least one antenna and one antenna of a wireless receiver (e.g., 109) having at least one antenna, according to one embodiment of the present teachings.

[0027] [Figure 3] A diagram illustrating a period associated with at least one time instance (e.g., 116) used to determine an initial time and at least one associated CI (e.g., 111A, 111B) used to determine an initial position, according to one embodiment.

[0028] [Figure 4A] FIG. 10 illustrates a first device moving between a second device and a third device at an angle less than 90 degrees, with an initial direction, initial position, and initial time (e.g., 116) of the first device determined, according to one embodiment.

[0029] [Figure 4B] 1 illustrates a first device moving between a second device and a third device at a 90 degree angle, with an initial direction, initial position, and initial time (e.g., 116) of the first device determined, according to one embodiment.

[0030] [Figure 4C] 1 illustrates a first device moving between a second device and a third device at an angle greater than 90 degrees, with an initial direction, initial position, and initial time (e.g., 116) of the first device determined, according to one embodiment.

[0031] [Figure 5] 1 illustrates an exemplary method for object tracking, according to one embodiment of the present invention.

[0032] [Figure 6] 1 illustrates an example of spatiotemporal information according to one embodiment of the present invention.

[0033] [Figure 7] 1 illustrates an example of a task performed based on spatiotemporal information, according to one embodiment of the present teachings.

[0034] [Figure 8] 1 illustrates another exemplary method for object tracking, according to one embodiment of the present invention.

[0035] [Figure 9] 1 illustrates an exemplary presentation of spatiotemporal information at a location, according to an embodiment of the present invention, and an exemplary scenario in which object motion is detected based on spatiotemporal information within a location, according to an embodiment of the present invention.

[0036] [Figure 10] 10 illustrates an illustration of antenna matching based on velocity estimation, according to one embodiment of the present invention.

[0037] [Figure 11] 3 is a flow chart illustrating a method for antenna matching based on velocity estimation, in accordance with one embodiment of the present invention.

[0038] [Figure 12] 1 illustrates an illustration of antenna matching based on velocity estimation, according to one embodiment of the present invention.

[0039] [Figure 13] 1 shows an exemplary flow chart of antenna matching based on velocity estimation.

[0040] [Figure 14] 1 illustrates a flowchart of an exemplary method for motion detection, according to one embodiment of the present invention.

[0041] [Figure 15]2 illustrates an exemplary block diagram of a first wireless device according to one embodiment of the present invention.

[0042] [Figure 16] 3 illustrates a flowchart of an exemplary method performed by a first wireless device, according to one embodiment of the present invention.

[0043] [Figure 17] 1 illustrates an exemplary block diagram of a second wireless device, in accordance with one embodiment of the present invention.

[0044] [Figure 18] 4 illustrates a flowchart of an exemplary method performed by a second wireless device, according to one embodiment of the present invention.

[0045] [Figure 19] 1 illustrates an exemplary motion detector, according to one embodiment of the present invention.

[0046] [Figure 20] 3 illustrates an exemplary motion detection algorithm, according to one embodiment of the present invention.

[0047] [Figure 21] 1 illustrates an exemplary method for object motion detection, according to one embodiment of the present invention.

[0048] [Figure 22] 1 shows an exemplary flowchart for a method for object motion localization, according to one embodiment of the present invention.

[0049] [Figure 23] 1 illustrates an exemplary topology of an origin satellite, where one dummy origin and one dummy bot act as a satellite pair, and multiple tracker bots listen on the same predetermined MAC address, according to one embodiment of the present invention.

[0050] [Figure 24]1 illustrates an exemplary topology of an origin satellite, where one dummy bot acts as a satellite standalone device and multiple tracker bots listen on the same globally unique MAC address, according to one embodiment of the present invention.

[0051] [Figure 25] FIG. 2 illustrates an exemplary flow for estimating location information of an indoor moving object according to one embodiment of the present invention.

[0052] [Figure 26] FIG. 1 illustrates an exemplary flow for estimating position information of moving objects in a room using a particle filter-based approach, according to one embodiment of the present invention.

[0053] [Figure 27] FIG. 2 illustrates an exemplary flow of handover between multiple origin satellites according to one embodiment of the present invention.

[0054] [Figure 28] FIG. 2 illustrates an exemplary flow of a handover algorithm between multiple origin satellites, according to one embodiment of the present invention.

[0055] [Figure 29] 1 illustrates an exemplary scheme for extraction and maximization of a respiratory signal, according to one embodiment of the present invention.

[0056] [Figure 30] 1 illustrates an exemplary respiration signal based on real-world measurements, according to one embodiment of the present invention.

[0057] [Figure 31] 1 illustrates the benefits of the disclosed scheme for extraction and maximization of a respiratory signal, according to one embodiment of the present invention.

[0058] [Figure 32A] , [Figure 32B]1 illustrates a comparison between a wake state and a sleep state according to one embodiment of the present invention.

[0059] [Figure 33A] , [Figure 33B] 1 illustrates respiration rate performance for different sleep stages according to an embodiment of the present invention.

[0060] [Figure 34] 1 illustrates an exemplary network environment for sleep monitoring, according to one embodiment of the present invention.

[0061] [Figure 35] 1 illustrates an exemplary algorithm design for sleep monitoring, according to one embodiment of the present invention.

[0062] [Figure 36] 1 illustrates an exemplary interior bird's-eye view of a motor vehicle for seat occupancy detection and people counting according to an embodiment of the present invention.

[0063] [Figure 37A] , [Figure 37B] , [Figure 37C] , [Figure 37D] 1 illustrates the change in CI due to different seat occupancy situations in a vehicle, according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0064] In the following detailed description, numerous specific details are set forth by way of example to provide a thorough understanding of the relevant teachings. However, it will be apparent to those skilled in the art that the present teachings may be practiced without such details. In other instances, well-known methods, procedures, components, and / or circuits are described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present teachings. System / Interconnect

[0065] In one embodiment, the present teachings disclose 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 (channel) can be obtained using a processor, a memory communicatively coupled to the processor, and an instruction set stored in the memory. The time series of CI (TSCI) can be extracted from wireless signals (signals) transmitted between a Type 1 (Type 1) heterogeneous wireless device and a Type 2 (Type 2) heterogeneous wireless device through the channel within a location. The channel may be affected by the movement of objects within the location. Characteristics and / or spatiotemporal information (e.g., motion information) of the objects and / or their motion can be monitored based on the TSCI. A task can be performed based on the characteristics and / or spatiotemporal information (e.g., motion information). A presentation related to the task can be generated in a user interface (UI) on a user's device. The TSCI can be preprocessed.

[0066] A Type 1 device may include at least one heterogeneous wireless transmitter. A Type 2 device may include at least one heterogeneous wireless receiver. The Type 1 device and the Type 2 device may be the same device. Any device may have a processor, a memory communicatively coupled to the processor, and an instruction set stored in the memory for execution by the processor. Some processors, memories, and instruction sets may cooperate. There may be multiple Type 1 devices interacting 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 width / size and / or time shift. Wireless signals sent by multiple Type 1 devices may be synchronous and / or coexistent. Multiple Type 1 / Type 2 devices may operate independently and / or cooperatively. Type 1 and / or Type 2 devices may have / include / be 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). The operation of a device may be based on the operation, state, internal state, storage, processor, memory output, physical location, computational resources, or network of another device. Various devices may communicate directly and / or through another device / server / cloud server. Devices may be associated with one or more users with associated settings. Settings may be selected once, pre-programmed, and / or changed over time. When one or more Type 1 devices interact with one or more Type 2 devices, any processing (e.g., time domain, frequency domain) may differ for different devices.Processing may be based on location, orientation, direction, role, user-related characteristics, settings, configuration, available resources, available bandwidth, network connectivity, hardware, software, processor, co-processor, memory, battery life, available power, antenna, antenna type, antenna directionality / unidirectionality, power settings, and / or other parameters / characteristics of the device.

[0067] A wireless receiver (e.g., a Type 2 device) can receive a signal and / or another signal from a wireless transmitter (e.g., a Type 1 device). A wireless receiver can receive another signal from another wireless transmitter (e.g., a second Type 1 device). A wireless transmitter can transmit a signal and / or another signal to another wireless receiver (e.g., a second Type 2 device). A wireless transmitter, a wireless receiver, another wireless receiver, and / or another wireless transmitter can move with the object and / or another object. Another object can be tracked. Another object can be attached.

[0068] 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 switch / establish a wireless coupling from a Type 2 device to a second Type 2 device at another location within the location. Similarly, the Type 2 device may switch / establish a wireless coupling from a Type 1 device to a second Type 1 device at yet another location within the location. The switching may be controlled by a server, a processor, a Type 1 device, a Type 2 device, and / or another device. The radio used before and after the switching may be different. A second wireless signal (second signal) may be transmitted between the Type 1 device and the second Type 2 device (or between the Type 2 device and the second Type 1 device) over the channel. A second TSCI of the channel may be extracted from the second signal. The second signal may be the first signal. Object features, spatiotemporal information, and / or another quantity may be monitored based on the second TSCI. The Type 1 device and the Type 2 device may be identical.

[0069] The signal and / or another signal may have data embedded therein. The signal may be a series of probe signals. The probe signal may have data embedded therein. The probe signal may be replaced by a data signal. The wireless receiver, the wireless transmitter, the another wireless receiver, and / or the another wireless transmitter may be associated with at least one processor, a memory communicatively coupled to the respective processor, and / or a respective instruction set stored in the memory, which, when executed, causes the processor to perform any and / or all steps necessary to determine spatiotemporal information (e.g., motion information), initial spatiotemporal information, initial time, direction, instantaneous location, instantaneous angle, and / or speed of the object. The processor, memory, and / or instruction set may be associated with the Type 1 heterogeneous wireless transceiver, one of the at least one Type 2 heterogeneous wireless transceiver, the object, a device associated with the object, another device associated with the location, a cloud server, and / or another server.

[0070] A Type 1 device can transmit a signal in a broadcast manner to at least one Type 2 device(s) over a channel within a location. The signal is transmitted without the Type 1 device establishing a wireless connection with any Type 2 device. A Type 1 device can transmit to a specific media access control (MAC) address that is common to two or more Type 2 devices. Each Type 2 device can adjust its MAC address to the specific MAC address.

[0071] A particular MAC address may be associated with a location. The association may be recorded in an association table of an association server. A location may be identified by a Type 1 device, a Type 2 device, and / or another device based on a particular MAC address, a series of probe signals, and / or at least one TSCI extracted from the probe signals. For example, a Type 2 device may be moved to a new location within a location (e.g., from another location). A Type 1 device may be newly set up within a location such that the Type 1 and Type 2 devices are unaware of each other. During setup, the Type 1 device may be instructed / induced / caused / controlled (e.g., using a dummy receiver, using a hardware pin configuration / connection, using a stored configuration, using a local configuration, using a remote configuration, using a downloaded configuration, or using a server) to send a series of probe signals to a particular MAC address. Upon powering on, the Type 2 device may scan for probe signals according to a table of MAC addresses (e.g., various MAC addresses used in various locations such as homes, offices, enclosures, floors, skyscrapers, stores, airports, malls, stadiums, halls, stations, subways, lots, areas, zones, regions, districts, cities, countries, continents, etc.) that may be used to broadcast at various locations (e.g., stored in a designated source, server, cloud server). If the Type 2 device detects a probe signal transmitted to a particular MAC address, the Type 2 device can use the table to identify a location based on the MAC address. The location of the Type 2 device within the location can be calculated based on the particular MAC address, the series 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.

[0072] The specific MAC address may change over time. The change may be in accordance with a timetable, a rule, a policy, a mode, a condition, a situation, and / or a change. The specific MAC address may be selected based on MAC address availability, a preselection list, a collision pattern, a traffic pattern, data traffic between the Type 1 device and another device, available bandwidth, a random selection, and / or a MAC address switching plan. The specific MAC address may be the MAC address of a second wireless device (e.g., a dummy receiver or a receiver acting as a dummy receiver). The Type 1 device may transmit a probe signal on a channel selected from a set of channels. At least one CI for the selected channel may be obtained by each Type 2 device from the probe signal transmitted on the selected channel. The selected channel may change over time. The change may be in accordance with a timetable, a rule, a policy, a mode, a condition, a situation, and / or a change. The selected channel may be selected based on channel availability, random selection, a pre-selection list, co-channel interference, inter-channel interference, channel traffic patterns, data traffic between the Type 1 device and another device, the available bandwidth associated with the channel, security criteria, channel switching plans, criteria, and / or considerations. Information about the specific MAC address and / or selected channel may be communicated between the Type 1 device and a server over a network. Information about the specific MAC address and / or selected channel may also be communicated between the Type 2 device and a server over another network. A Type 2 device may communicate information about the specific MAC address and / or selected channel to another Type 2 device (e.g., via a mesh network, Bluetooth, WiFi, etc.). The specific MAC address and / or selected channel may be selected by a server. The specific MAC address and / or selected channel may be signaled by the Type 1 device, the Type 2 device, and / or the server over an announcement channel. Any information may be pre-processed before communication.

[0073] A wireless connection between a Type 1 device and another wireless device may be established (e.g., using a signal handshake). The Type 1 device may 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 may reply by sending a second handshake signal (e.g., an instruction or a clear to send CTS) to the Type 1 device and triggering the Type 1 device to send a signal (e.g., a series of probe signals) to multiple Type 2 devices in a broadcast manner without establishing a connection with any Type 2 devices. The second handshake signal may be a response or acknowledgment (e.g., an ACK) to the first handshake signal. The second handshake signal may include data having location information and / or information of the Type 1 device.

[0074] The other device may be a dummy device with the purpose (e.g., primary purpose, secondary purpose) of establishing a wireless connection with the Type-1 device, receiving the first signal, and / or transmitting the second signal. The other device may be physically attached to the Type-1 device. In another example, the other device may send a third handshake signal to the Type-1 device, triggering the Type-1 device to broadcast a signal (e.g., a series of probe signals) to multiple Type-2 devices without establishing a connection with any Type-2 devices. The Type-1 device may reply to the third special signal by sending a fourth handshake signal to the other device. The other device may be used to trigger and broadcast to two or more Type-1 devices. Triggering may be sequential, partially sequential, partially parallel, or fully parallel. The other device may have two or more radio circuits for triggering multiple transmitters in parallel. Parallel triggering may also be achieved by using at least one additional device to perform triggering in parallel with the other device (as the other device does). The other device may not communicate (or may suspend communication) with the Type-1 device after establishing a connection with the Type-1 device. Suspended communication may be resumed. The other device may enter an inactive mode, a hibernation mode, a sleep mode, a standby mode, a low power mode, an OFF mode, and / or a power down mode after establishing a connection with the Type 1 device. The other device may have a specific MAC address, which causes the Type 1 device to send a signal 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 designated source, and / or a fourth processor associated with the other device. The first and second processors may be coordinated with each other.

[0075] A first series of probe signals may be transmitted by a first antenna of the Type 1 device to at least one first Type 2 device via a first channel in a first location. A second series of probe signals may be transmitted by a second antenna of the Type 1 device to at least one second Type 2 device via a second channel in a second location. The first series and the second series 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 between the Type 1 device and any Type 2 device. The first and second antennas may be the same or not.

[0076] The two locations may have different sizes, shapes, and multipath characteristics. The first and second locations may overlap. The respective proximity areas around 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 2.4 GHz WiFi and the second may be 5 GHz WiFi. Or, both may be 2.4 GHz WiFi but have different channel numbers, SSID names, and / or WiFi settings. Each Type 2 device may obtain at least one TSCI from each series of probe signals, and the CI may be for 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) may be identical. The first and second series of probe signals may be synchronous / asynchronous. The probe signals may be transmitted along with data or may be replaced by data signals. The first and second antennas may be the same or different.

[0077] The first series of probe signals may be transmitted at a first rate (e.g., 30 Hz). The second series of probe signals may be transmitted at a second rate (e.g., 200 Hz). The first and second rates may be the same or different. The first and / or second rates may change over time. The change may be according to a timetable, rule, policy, mode, condition, situation, and / or change. Any rate may change 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 on a first channel. The second series of probe signals may be transmitted on 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 changes may be in accordance with timetables, rules, policies, modes, conditions, circumstances, and / or changes.

[0078] A Type 1 device and another device may be controlled and / or coordinated, may be physically attached, or may be / be within a common device. They may be controlled / connected by a common data processor or connected to a common bus interconnect / network / LAN / Bluetooth network / BLE network / 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), household, house, 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., transmit a respective signal (e.g., a respective series of probe signals) to a respective set of Type 2 devices). Each respective Type 2 device selects a Type 1 device as a signal source from among all Type 1 devices. Each Type 2 device may select a Type 1 device as a signal source asynchronously. At least one TSCI can be obtained by each respective Type 2 device from a respective series of probe signals from the Type 1 device, the CI being of a channel between the Type 2 device and the Type 1 device.

[0079] Each Type 2 device selects a Type 1 device as a signal source from among all Type 1 devices based on the Type 1 / Type 2 device's identification (ID), 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, account, access information, parameters, characteristics, and / or signal strength information (e.g., associated with the Type 1 device and / or each Type 2 receiver). Initially, a Type 1 device may be a signal source for the initial set of respective Type 2 devices (i.e., the Type 1 device transmits a respective signal (series of probe signals) to the initial set of respective Type 2 devices). Each initial respective Type 2 device selects a Type 1 device as a signal source from among all Type 1 devices.

[0080] A signal source of a particular Type 2 device (Type 1 device) may be changed if (1) the time interval between two adjacent probe signals (e.g., between the current probe signal and the immediately preceding probe signal, or between the next probe signal and the current probe signal) received from the current signal source of the Type 2 device exceeds a first threshold, (2) the signal strength associated with the current signal source of the Type 2 device is less than a second threshold, (3) the processed signal strength associated with the current signal source of the Type 2 device is less than a third threshold, and the signal strength has been 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 linear filter, and / or (4) the signal strength (or the processed signal strength) associated with the current signal source of the Type 2 device is less than a fourth threshold for a significant percentage (e.g., 70%, 80%, 90%, etc.) of the recent time window. The percentage may exceed a fifth threshold. The first, second, third, fourth, and / or fifth thresholds may change over time. Condition (1) may occur when the Type 1 device and the Type 2 device gradually move away from each other, causing some probe signals from the Type 1 device to become too weak and not be received by the Type 2 device. Conditions (2)-(4) may occur when the two devices move away from each other, causing the signal strength to become very weak. The signal source of a Type 2 device may not change if other Type 1 devices have signal strengths weaker than the current signal source by a factor (e.g., 1, 1.1, 1.2, or 1.5). When the signal source changes, a 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. A list of available Type 1 devices may be initialized and maintained by each Type 2 device. The list may be updated by examining the signal strength and / or processed signal strength associated with each set of Type 1 devices.

[0081] The 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 / states, tasks to be performed by the Type 2 device, signal strengths of the first and second series, and / or other considerations. The series of probe signals may be transmitted at a regular rate (e.g., 200 Hz). The series of probe signals may be scheduled at regular intervals (e.g., 0.01 seconds at 100 Hz), although each probe signal may experience brief 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. The change may be according to a timetable (e.g., once per hour), a rule, a policy, a mode, a condition, and / or a change (e.g., whenever some event occurs). For example, the rate may be normally 100Hz, but may be changed to 1000Hz in situations where necessary, or to 1Hz in low power / standby situations. The probe signal may be transmitted in bursts.

[0082] The probe signal rate may vary based on the task performed by the Type 1 or Type 2 device (e.g., a task may require 100 Hz normally, 1000 Hz for 20 seconds instantaneously). In one embodiment, the transmitter (Type 1 device), receiver (Type 2 device), and associated task may be adaptively associated with a class (e.g., low priority, high priority, emergency, critical, normal, privileged, non-subscription, subscription, payment, and / or non-payment classes). The (transmitter's) rate may be adjusted for some classes (e.g., high priority class). If the needs of a class change, the rate can be changed. If the receiver's power is very low, the rate can be reduced to reduce the receiver's power consumption to respond to the probe signal. In one embodiment, the probe signal can be used to wirelessly transmit power to the receiver (Type 2 device), and the rate can be adjusted to control the amount of power transmitted to the receiver. The rate can be changed by (or based on) the server, the Type 1 device, and / or the Type 2 device. Control signals can be communicated between them. The server can monitor, track, forecast, and / or anticipate the needs of and / or tasks performed by Type 2 devices and can control Type 1 devices to change rates. The server can schedule rate changes according to a timetable. The server can detect emergencies and change rates immediately. The server can detect evolving conditions and gradually adjust rates.

[0083] The features and / or spatiotemporal information (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 collectively based on any TSCIs associated with any Type 1 device and any Type 2 device. Any joint monitoring may be associated with a user, user account, profile, household, location map, and / or user history, etc.

[0084] 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, modulation, wireless standards, coding, encryption, payload characteristics, networks, network IDs, SSIDs, network characteristics, network settings, and / or network parameters, etc. The two channels may be associated with different types of wireless systems (e.g., two of WiFi, LTE, LTE-A, 2.5G, 3G, 3.5G, 4G, greater than 4G, 5G, 6G, 7G, 802.11 systems, 802.15 systems, 802.16 systems, mesh networks, Zigbee, WiMax, Bluetooth, BLE, RFID, UWB, microwave systems, radar-like systems, etc.). For example, one channel is WiFi and the other is LTE. The two channels may be associated with the same type of wireless system but in 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 in the 5 GHz band with a bandwidth of 40 MHz and an SSID of "StarBud hotspot." The two channels may be different channels within the same network (e.g., the "StarBud hotspot" network).

[0085] In one embodiment, the wireless monitoring system may include training a classifier for multiple events within a location based on training TSCIs associated with the multiple events. For each of multiple known events occurring within the location during a respective training period associated with the known events, a respective training wireless signal (e.g., a respective series of training probe signals) may be transmitted by an antenna of the first type-1 heterogeneous wireless device to at least one first type-2 heterogeneous wireless device via a wireless multipath channel within the location during the respective training period using a processor, memory, and instruction set of the first type-1 device. At least one respective time series of training CIs (training TSCIs) may be asynchronously obtained by each of the at least one first type-2 device from the (respective) training signal. The CIs may be CIs of channels between the first type-2 device and the first type-1 device during the training period associated with the known events. The at least one training TSCI may be preprocessed.

[0086] For a current event occurring within a current time period within the location, a current wireless signal (e.g., a series of current probe signals) may be transmitted by an antenna of the second type-1 heterogeneous wireless device to at least one second type-2 heterogeneous wireless device via a channel within a current time period associated with the current event within the location using a processor, memory, and instruction set of the second type-1 device. At least one time series of current CI (current TSCI) may be asynchronously derived by each of the at least one second type-2 device from the current signal (e.g., the series of current probe signals). The CI may be a CI of a channel between the second type-2 device and the second type-1 device within a current time period associated with the current event. The at least one current TSCI may be preprocessed.

[0087] The classifier may be applied to classify at least one current TSCI obtained from a series of current probe signals by at least one second Type 2 device, to classify at least one portion of a particular current TSCI, and / or to classify a combination of at least one portion of a particular current TSCI with another portion of another TSCI. The classifier may also be applied to associate a current event with a known event, an unknown event, and / or another event. Each TSCI may include at least one CI, each associated with a respective timestamp. Two TSCIs associated with two Type 2 devices may differ in start time, duration, stop time, amount of CIs, sampling frequency, and sampling frequency. The CIs may have different characteristics. The first and second Type 1 devices may be the same location within a location. They may be the same device. At least one second Type 2 device (or these locations) may be a permutation of at least one first Type 2 device (or these locations). A specific second Type 2 device and a specific first Type 2 device may be the same device.

[0088] The subset of first Type 2 devices and the subset of second Type 2 devices may be identical. 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 the subset of at least one first Type 2 device may be a permutation of a subset of at least one second Type 2 device. At least one second Type 2 device and / or the subset of at least one second Type 2 device may be a permutation of a subset of at least one first Type 2 device. At least one second Type 2 device and / or the subset of at least one second Type 2 device may be in the same respective location as a subset of at least one first Type 2 device. At least one first Type 2 device and / or the subset of at least one first Type 2 device may be in the same respective location as a subset of at least one second Type 2 device.

[0089] The antenna of the Type 1 device and the antenna of the second Type 1 device may be at the same location within the location. The antenna(s) of the at least one second Type 2 device and / or the antenna(s) of a subset of the at least one second Type 2 device may be at the same respective location as the respective antenna(s) of the subset of the at least one first Type 2 device. The antenna(s) of the at least one first Type 2 device and / or the antenna(s) of the subset of the at least one first Type 2 device may be at the same respective location(s) as the respective antenna(s) of the subset of the at least one second Type 2 device.

[0090] 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 the items of the first section and the items of the second section may be calculated. The first section may include a first segment of the first TSCI having a first start time / end time and / or another segment 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 of the second TSCI having a second start time and a second end time, as well as another segment of the processed second TSCI. The processed second TSCI may be the second TSCI processed by the second operation.

[0091] The first operation and / or the second operation may include subsampling, resampling, interpolation, filtering, transformation, feature extraction, preprocessing, and / or another operation. A first item in the first section may be mapped to a second item in the second section. A first item in the first section may also be mapped to another item in the second section. Another item in the first section may also be mapped to a second item in the 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: a first item of a first section of a first TSCI; another item of the first TSCI; a timestamp of the first item; a time difference of the first item; a time derivative of the first item; an adjacent timestamp of the first item; another timestamp associated with the first item; a second item of a second section of a second TSCI; another item of the second TSCI; a timestamp of the second item; a time difference of the second item; a time derivative of the second item; an adjacent timestamp of the second item; and another timestamp associated with the second item may satisfy at least one constraint.

[0092] One constraint may be that the difference between the timestamp of the first item and the timestamp of the second item may be upper bounded by an adaptive upper threshold and lower bounded by an adaptive lower threshold. The first section may be the entire first TSCI. The second section may be the entire second TSCI. The first duration may be equal to the second duration. The section duration of the TSCI may be adaptively determined. A provisional section of the TSCI may be calculated. A start time and an end time of a section (e.g., a provisional section) may be determined. The section may be determined by removing the beginning and end of the provisional section. The beginning of the provisional section may be determined as follows: Again, the item in the provisional section with increasing timestamp may be considered the current item, one item at a time. At each iteration, at least one activity measure may be calculated and / or considered. At least one activity measure may be associated with at least one of a current item associated with a current timestamp, a past item in the provisional section having a timestamp not greater than the current timestamp, and / or a future item in the provisional section having a timestamp not less than the current timestamp. If at least one criterion associated with the at least one activity measure is met, the current item may be added to an initial portion of the provisional section.

[0093] At least one criterion associated with the activity measure is: (a) the activity measure is less than an adaptive upper threshold; (b) the activity measure is greater than an adaptive lower threshold; (c) for at least a predetermined amount of consecutive timestamps, the activity measure is continuously less than the adaptive upper threshold; (d) for at least another predetermined amount of consecutive timestamps, the activity measure is continuously greater than the adaptive lower threshold; (e) for at least a predetermined percentage of at least a predetermined amount of consecutive timestamps, the activity measure is continuously less than the adaptive upper threshold; or (f) for at least another predetermined percentage of another predetermined amount of consecutive timestamps, the activity measure is continuously greater than the adaptive lower threshold. (g) another activity measure associated with another timestamp associated with the current timestamp is less than another adaptive upper threshold and greater than another adaptive lower threshold; (h) at least one activity measure associated with at least one respective timestamp associated with the current timestamp is less than a respective upper threshold and greater than a respective lower threshold; (i) a percentage of timestamps in the set of timestamps associated with the current timestamp that are less than a respective upper threshold and greater than a respective lower threshold exceed a threshold; and (j) another criterion.

[0094] The activity measure associated with an item at time T1 may include at least one of: (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 over time; T2 may be updated periodically; T2 may be the beginning of a period and T1 may be a sliding time within the period); and (4) a fourth function of the item at time T1 and another item.

[0095] 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,...)) having 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 some 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 can be (XY)^2 when abs(XY) is less than a threshold T, and (XY)+a when abs(XY) is greater than a threshold T. Alternatively, the function can be constant when abs(XY) is greater than T. The function can also be bounded by slowly increasing the function when abs(Xy) is greater than T so that outliers do not seriously affect the results. Another example of a function can be (abs(X / Y)-a), where a=1. In this way, when X=Y (i.e., no change or no 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. And, when X is less than Y, (X / Y) is less than 1, and the function is negative. In another embodiment, both arguments X and Y can be n-tuples, such that 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_i-x_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-tuple of X and Y (1≦i≦n), e.g., the component index of x_1 is i=1 and the component index of x_2 is i=2.The function may include a component-wise sum of at least one further function of the following: x_i, y_i, (x_i-y_i), (y_i-x_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 X and Y in the n-tuple. For example, the function may be of the form sum_{i=1}^n(abs(x_i / y_i)-1) / n, or sum_{i=1}^n w_i×(abs(x_i / y_i)-1), where w_i is the weight of component i.

[0096] The map may be calculated using dynamic time warping (DTW). The DTW may include constraints on at least one of the map, the first TSCI item, the second TSCI item, the first duration, the second duration, the first section, and / or the second section. Assume that in the map, the i-th domain item is mapped to the j-th range item. The constraint may be on the allowable combinations of i and j (constraints on the relationship between i and j). A mismatch cost between a first section of a first duration of the first TSCI and a second section of a second duration of the second TSCI may be calculated. The first and second sections may be aligned, thereby establishing a map including two or more links between the first item of the first TSCI and the second item of the second TSCI. For each link, one of the first items having a first timestamp may be associated with one of the second items having a second timestamp. A mismatch cost between the matched first section and the matched second section may be calculated, and the mismatch cost may include a function of an item-like cost between the first item and the second item related by a particular link of the map and a link-like cost associated with the particular link of the map.

[0097] The matched first section and the matched 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 an inner product, an inner-product-like quantity, a correlation-based quantity, a covariance-based quantity, a discriminant score, a distance, a Euclidean distance, an absolute distance, an Lk distance (e.g., L1, L2, ...), a weighted distance, a distance-like quantity, and / or another similarity value between the first vector and the second vector. The mismatch cost may be normalized by the respective vector lengths.

[0098] A parameter derived from the 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 can be modeled with a statistical distribution. At least one of a scale parameter, a location parameter, and / or another parameter of the statistical distribution can be estimated. The first section of the first duration of the first TSCI can be a sliding section of the first TSCI. The second section of the second duration of the second TSCI can be a sliding section of the second TSCI. A first sliding window can be applied to the first TSCI, and a corresponding second sliding window can be applied to the second TSCI. The first sliding window of the first TSCI and the corresponding second sliding window of the second TSCI can be aligned. A mismatch cost between the aligned first sliding window of the first TSCI and the aligned second sliding window of the second TSCI can be calculated. The current event may be associated with at least one of a known event, an unknown event, and / or another event based on a mismatch cost.

[0099] The classifier may be applied to at least one of each first section of the first duration of the first TSCI and / or each second section of the second duration of the second TSCI to obtain at least one provisional classification result. Each provisional classification result may be associated with the respective first section and the respective second section. The current event may be associated with at least one of a known event, an unknown event, and / or another event based on mismatch costs. The current event may be associated with at least one of a known event, an unknown event, and / or another event based on the maximum number of corresponding provisional classification results in two or more sections of the first TSCI and in more sessions of the second TSCI. For example, if the mismatch costs point to a specific known event N consecutive times (e.g., N=10), the current event may be associated with the specific known event. In another example, if the percentage of mismatch costs within the N consecutive N immediately preceding the specific known event that point to the specific known event exceeds a certain threshold (e.g., >80%), the current event may be associated with the specific known event.

[0100] In another example, the current event may be associated with a known event that minimizes the mismatch cost most of the time within the time period. The current event may be associated with a known event that minimizes the overall mismatch cost, where the overall mismatch cost 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 the minimum of another overall cost. If no known event achieves a mismatch cost less than a first threshold T1 for a sufficient percentage of at least one first section, the current event may be associated with an "unknown event." If no event achieves an overall mismatch cost less than a second threshold T2, the current event may also be associated with an "unknown event." 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 costs and additional mismatch costs associated with at least one additional section of the first TSCI and at least one additional section of the second TSCI. The known events may include at least one of a door close event, a door open event, a window close event, a window open 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 human presence event, a human absence event, a popular presence event, and / or a popular absence event. A projection for each CI may be trained using a dimension 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 linear discriminant analysis, 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 the at least one event and / or the current TSCI for the classifier. The classifier for the at least one event may be trained based on the projection and the training TSCI associated with the at least one event. The at least one current TSCI may be classified 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 linear discriminant analysis, vector quantization, supervised learning, unsupervised learning, self-organizing maps, autoencoders, neural networks, deep neural networks, and / or other methods. A classifier for 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. 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 values ​​including the magnitude of the corresponding complex value. Each training TSCI may be weighted in training the projections. The projections may include two or more projected components.The projection may include at least one most significant projected component. The projection may include at least one projected component that may be informative to the classifier. Channel / Channel information / Location / Spatial-temporal information / Motion / Object

[0101] Channel information (CI) may be associated with / include signal strength, signal amplitude, signal phase, received signal strength indicator (RSSI), channel state information (CSI), channel impulse response (CIR), channel frequency response (CFR), channel characteristics, channel filter response, timestamp, auxiliary information, data, metadata, user data, account data, access data, security data, session data, status data, monitor data, household data, identification information (ID), device data, network data, neighborhood data, environmental data, real-time data, sensor data, stored data, encrypted data, compressed data, protected data, and / or other channel information. CI may be associated with information associated with frequency bands, frequency signatures, frequency phases, frequency amplitudes, frequency trends, frequency features, frequency-like characteristics, time-domain elements, frequency-domain elements, time-frequency domain elements, orthogonal decomposition features, and / or non-orthogonal decomposition features of signals over the channel.

[0102] A CI may also be associated with information associated with a signal's duration, time signature, time stamp, time amplitude, time phase, time trend, and / or time characteristics. A CI may be associated with information associated with a signal's time-frequency division, signature, amplitude, phase, trend, and / or characteristics. A CI may be associated with a signal's decomposition. A CI may be associated with information associated with a signal's direction, angle of arrival (AoA), directional antenna angle, and / or phase through a channel. A CI may be associated with a signal's attenuation pattern through a channel. Each CI may be associated with a Type 1 device and a Type 2 device. Each CI may be associated with a Type 1 device's antenna and a Type 2 device's antenna. A CI may be obtained from communication hardware capable of providing a CI. The communications hardware may be a WiFi-enabled chip / IC (integrated circuit), a chip compliant with 802.11 or 802.16 or another wireless / wireless standard, a next-generation WiFi-enabled chip, an LTE-enabled chip, a 5G-enabled chip, a 6G / 7G / 8G-enabled chip, a Bluetooth-enabled chip, a Bluetooth low power (BLE)-enabled chip, a UWB chip, another communications chip (e.g., Zigbee, WiMax, mesh network), etc. The communications hardware calculates the CI and stores the CI in a buffer memory to make the CI available for extraction. The CI may include data related to channel state information (CSI) and / or at least one matrix. The at least one matrix may be used for channel equalization, beamforming, etc.

[0103] A channel may be associated with a location. Attenuation may be due to signal propagation within the location, signal propagation / reflection / refraction / diffraction through / in / at the air (e.g., the air of the location), and refractive media / reflective surfaces (e.g., walls, doors, furniture, obstacles, and / or barriers). Attenuation may be due to reflections off surfaces and obstacles (e.g., reflective surfaces, obstacles), such as floors, ceilings, furniture, built-ins, objects, people, and pets. Each CI may be associated with a timestamp. Each CI may include N components (e.g., N frequency-domain components of the CFR, N time-domain components of the CIR, or N decomposed components). Each component may be associated with a component index. Each component may be a real, imaginary, or complex number, magnitude, phase, flag, and / or set. Each CI may include a vector or matrix of complex numbers, a set of mixed quantities, and / or a multi-dimensional collection of at least one complex number.

[0104] 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 respective component time series is associated with a respective component index. Object motion characteristics / spatiotemporal information may be monitored based on the component time series.

[0105] A feature for each component of the TSCI of the component-feature time series may be calculated. The feature for each component may be a scalar (e.g., energy) or a function with a domain and range (e.g., autocorrelation function, transform, inverse transform). The feature / spatiotemporal information of the object's motion may be monitored based on the feature for each component. The overall feature of the TSCI may be calculated based on the feature for each component of each component time series of the TSCI. The overall feature may be a weighted average of the feature for each component. The feature / spatiotemporal information of the object's motion may be monitored based on the overall feature.

[0106] The Type 1 and Type 2 devices may support WiFi, WiMax, 3G / exceeding 3G, 4G / exceeding 4G, LTE, 5G, 6G, 7G, Bluetooth, BLE, Zigbee, UWB, mesh networks, proprietary wireless systems, IEEE 802.11 standards, 802.15 standards, 802.16 standards, 3GPP standards, and / or another wireless system. 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. At least one Type 2 transceiver may transmit their respective signals simultaneously using the common wireless system and / or common wireless channel. A Type 1 transceiver may transmit signals to at least one Type 2 transceiver using the common wireless system and / or common wireless channel. A Type 1 device may temporarily function as a Type 2 device, and vice versa. A device may simultaneously function as both a Type 1 device (wireless transmitter) and a Type 2 device (wireless receiver). Each Type 1 and Type 2 device may have at least one transmit / 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.

[0107] At least one TSCI may correspond to various antenna pairs between the Type 1 device and the 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 may be performed over the antenna links. The averaging or weighted averaging may be over at least one TSCI. The averaging may optionally be performed over a subset of at least one TSCI corresponding to a subset of antenna pairs. The timestamps of some CIs of a TSCI may be irregular and may be corrected, such that the corrected timestamps of the time-corrected CIs may be uniformly spaced in time. In the case of multiple Type 1 devices and / or multiple Type 2 devices, the corrected timestamps may relate to the same or different clocks. An original timestamp associated with each of the CIs may be determined. The original timestamps may not be uniformly spaced in time. The original timestamps of all CIs in a particular portion of a particular TSCI in the current sliding time window may be modified such that the modified timestamps of the time-modified CIs are uniformly spaced in time.

[0108] The features and / or spatiotemporal information (e.g., movement information) may include location, location coordinates, location change, location (e.g., initial location, new location), location on map, height, horizontal location, vertical location, distance, displacement, speed, acceleration, rotation speed, rotation acceleration, angle of movement, orientation, direction of movement, rotation, path, deformation, translation, shrinkage, expansion, gait, gait cycle, head movement, repetitive movement, cyclic movement, pseudo-cyclic movement, impulsive movement, sudden movement, fall movement, transient movement, behavior, transient behavior, movement period, movement frequency, time trend, temporal profile, temporal features, occurrence, change, Change in frequency, change in timing, change in gait cycle, timing, start time, end time, duration, movement history, movement type, movement classification, frequency, frequency spectrum, frequency features, presence, absence, proximity, approach, retreat, object identity, object configuration, head movement rate, head movement direction, mouth-related rate, eye-related rate, respiratory rate, heart rate, hand movement rate, hand movement direction, leg movement, body movement, gait rate, hand movement rate, position features, features associated with object movement, tool movement, machine movement, complex movement, and / or combination of multiple motions, events, motion statistics, motion magnitude, motion phase, similarity score, distance score, Euclidean distance, weighted distance, L_1 norm for k>2, L_2 norm, L_k norm, statistical distance, correlation, autocorrelation, covariance, autocovariance, cross-covariance, dot product, cross product, motion signal transformation, motion features, motion presence, motion absence, motion localization, motion identification, motion recognition, object presence, object absence, object entry, object exit, object change, motion cycle, motion count, gait cycle, motion rhythm, change Shape movement, gesture, handwriting, head movement, mouth movement, heart movement, visceral movement, movement tendency, size, length, area, volume, capacity, shape, form, tag, start location, end location, start amount, end amount, event, fall event, security event, accident event, home event, office event, factory event, warehouse event, manufacturing event, assembly line event, maintenance event, automobile-related event, navigation event, tracking event, door event, door open event, door close event, window event, window open event,The information may include a window close event, a repeatable event, a one-time event, a consumption amount, an unconsumption amount, a state, a physical state, a health state, a happiness state, an emotional state, a mental state, another event, and / or another information. The processor shares the computational workload with the Type 1 heterogeneous wireless device and the Type 2 heterogeneous wireless device.

[0109] 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 soundbar, a set-top box, an access point, a router, a repeater, a remote control, a speaker, a fan, a refrigerator, a microwave, an oven, a coffee maker, a kettle, a kitchen 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, a switch, a power adapter, a computer, a dongle, a computer peripheral, an electronic pad, a sofa, a tile, an accessory, a smart home device, a smart vehicle device, a smart office device, a smart building device, a smart manufacturing device, a smart watch, a smart glasses, a smart clock, a smart TV, a smart oven, a smart refrigerator, a smart air conditioner, a smart chair, a smart table, a smart accessory, a smart utility, a smart appliance, a smart machine, a smart vehicle, an Internet of Things (IoT) device, an Internet-enabled device, a computer, a portable computer, a tablet, a smart house, a smart office, a smart building, a smart parking lot, a smart system, and / or another device. Each Type 1 device may be associated with a respective identity (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 of 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, on a server, in the cloud, locally, remotely, permanently, temporarily) and retrieved.The ID may be associated with at least one record, account, user, household, address, phone number, social security number, customer number, another ID, 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 the Type 1 device and / or the Type 2 device for registration, initialization, communication, identification, verification, detection, recognition, authentication, access control, cloud access, networking, social networking, logging, recording, cataloging components, tagging, association, pairing, transactions, electronic transactions, and / or intellectual property control.

[0110] Objects include people, passengers, children, elderly people, babies, sleeping babies, babies in vehicles, patients, workers, high value workers, experts, specialists, waiters, mall customers, travelers at airports / train stations / bus terminals / ship terminals, factory / mall / supermarket / office / workplace staff / laborers / customer service representatives, sewer / ventilation system / lift shaft service personnel, lifts in lift shafts, elevators, inmates, people to be tracked / monitored, animals, plants, creatures, pets, dogs, cats, smartphones, phone accessories, computers, tablets, portable computers, dongles, computing accessories, network devices, WiFi devices, IoT devices, smart watches, smart glasses, smart devices, speakers, keys, smart keys, wallets, purses, handbags, backpacks, goods, cargo, luggage, equipment, motors, machinery, air conditioners, fans, air conditioning equipment, built-in lights, movable lights The object may be a light, a television, a camera, audio and / or video equipment, stationary, monitoring equipment, a part, a sign, a tool, a cart, a ticket, a parking pass, a local exchange pass, an airline ticket, a credit card, a plastic card, an access card, food packaging, kitchen utensils, a table, a chair, cleaning equipment / tools, a vehicle, an automobile, an automobile in a parking facility, goods in a warehouse / store / supermarket / distribution center, a boat, a bicycle, an airplane, a drone, a remote controlled automobile / plane / boat, a robot, a manufacturing device, an assembly line, materials / unfinished parts / robots / wagons / transports in a factory, an object being tracked in an airport / shopping mart / supermarket, a non-object, the absence of an object, the presence of an object, an object with form, an object that changes form, an object without form, a fluid mass, a liquid mass, a gas / smoke mass, a fire, a flame, an electromagnetic (EM) source, an EM medium, and / or another object.

[0111] The object itself may be communicatively coupled to several networks, such as WiFi, MiFi, 3G / 4G / LTE / 5G, Bluetooth, BLE, WiMax, Zigbee, mesh networks, ad-hoc networks, and / or other networks. The object itself may be bulky and equipped with AC power, but may be moved during installation, cleaning, maintenance, repair, etc. It may also be mounted on a movable platform, such as a lift, pad, mobile platform, elevator, conveyor belt, robot, drone, forklift, automobile, boat, vehicle, etc. The object may have multiple parts, each of which performs a different motion. For example, the object may be a person walking forward. While walking, the left and right hands may move in different directions with different instantaneous speeds, accelerations, motions, etc.

[0112] 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 can move with the object and / or another object (e.g., in a previous motion, a current motion, and / or a future motion). They can be communicatively coupled to one or more nearby devices. They can transmit TSCIs and / or information associated with the TSCIs to nearby devices and / or to each other. They can have nearby devices. The wireless transmitter and / or wireless receiver can be part of a small (e.g., coin-sized, cigarette pack-sized, or smaller), lightweight, portable device. The portable device can be wirelessly coupled to the nearby devices.

[0113] The nearby devices may be smartphones, iPhones, Android phones, smart devices, smart appliances, smart vehicles, smart gadgets, smart TVs, smart refrigerators, smart speakers, smart watches, smart glasses, smart pads, iPads, computers, wearable computers, notebook computers, gateways, etc. The nearby devices may be connected to cloud servers, local servers, and / or other servers via the Internet, wired Internet connections, and / or wireless Internet connections. The nearby devices may be portable. The portable device, nearby devices, local servers, and / or cloud servers may share computation and / or storage for tasks (e.g., obtaining TSCI, determining object features / spatiotemporal information associated with object motion, calculating time series power information, determining / calculating specific functions, searching for local extrema, classification, identifying specific values ​​of time offset, noise removal, processing, simplification, cleaning, wireless smart sensing tasks, CI extraction from signals, switching, segmentation, trajectory estimation, map processing, correction, correction adjustment, adjustment, map correction, error detection, boundary hitting check, thresholding, etc.) and information (e.g., TSCI).

[0114] The nearby device may or may not move with the object. The nearby device may be portable / non-portable / mobile / non-mobile. The nearby device may use battery power, solar power, 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 device may be a smart device, a network-enabled device, a device with a connection to WiFi / 3G / 4G / 5G / 6G / Zigbee / Bluetooth / ad-hoc network / other network, a smart speaker, a smart watch, a smart clock, a smart appliance, a smart machine, a smart piece of equipment, a smart tool, a smart vehicle, an internet-of-thing (IoT) device, an internet-enabled device, a computer, a portable computer, a tablet, and another device.

[0115] A nearby device and / or at least one processor, wireless transmitter, another wireless receiver, another wireless transmitter, and / or a cloud server (in the cloud) associated with the wireless receiver may determine initial spatiotemporal information of the object. Two or more of them may jointly determine the initial spatiotemporal information. Two or more of them may share intermediate information in determining the initial spatiotemporal information (e.g., initial location). In one embodiment, the wireless transmitter (e.g., a Type 1 device or a tracker bot) may move with the object. The wireless transmitter may transmit a signal to the wireless receiver (e.g., a Type 2 device or an origin registrar) or determine the object's initial spatiotemporal information (e.g., initial location). The wireless transmitter may also transmit a signal and / or another signal to another wireless receiver (e.g., another Type 2 device or another origin registrar) to monitor the object's movement (spatiotemporal information). The wireless receiver may 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 may be known. In another example, the wireless receiver (e.g., a Type 2 device or a tracker bot) can move with the object. The wireless receiver may receive a signal transmitted from a wireless transmitter (e.g., a Type 1 device or an origin registrar) to determine the object's initial spatiotemporal information (e.g., an initial position). The wireless transmitter may also receive a signal and / or another signal from another wireless transmitter (e.g., another Type 1 device or another origin registrar) to monitor the object's current movement (e.g., spatiotemporal information). 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) to monitor the object's movement. The location of the wireless transmitter and / or another wireless transmitter may be known.

[0116] Locations include rooms, homes, offices, workplaces, corridors, walkways, lifts, lift shafts, escalators, elevators, sewer systems, ventilation systems, stairs, assembly areas, ducts, air ducts, pipes, tubes, enclosed structures, semi-enclosed structures, enclosed areas, areas with at least one wall, plant, machinery, engines, structures made of wood, structures made of glass, structures made of metal, structures with walls, structures with doors, structures with gaps, structures with reflective surfaces, structures with fluids, buildings, rooftops, stores, factories, assembly lines, hotel rooms, museums, classrooms, schools schools, universities, government buildings, warehouses, garages, malls, airports, train stations, bus terminals, hubs, transportation hubs, shipping terminals, government facilities, public facilities, schools, universities, entertainment facilities, recreational facilities, hospitals, nursing homes, care facilities, community centers, stadiums, playgrounds, parks, fields, sports facilities, swimming facilities, tracks and / or fields, basketball courts, tennis courts, soccer stadiums, baseball stadiums, gymnasiums, auditoriums, garages, shopping marts, malls, supermarkets, manufacturing facilities, parking facilities, buildings Construction site, mining facility, transportation facility, highway, road, valley, forest, tree, land, terrain, cubicle, patio, ground, path, amusement park, urban area, countryside, suburban area, metropolitan area, garden, square, plaza, music hall, downtown facility, over-air facility, semi-open facility, enclosed area, railroad platform, train station, distribution center, warehouse, store, distribution center, storage facility, underground facility, space (e.g., ground, outer space) facility, floating facility, cave, tunnel facility, indoor facility, outdoor facility, outdoor facility with some walls / doors / reflective barriers, open The space may be an open facility, a semi-open facility, an automobile, truck, bus, van, container, ship / boat, submersible, train, tram, airplane, vehicle, mobile home, cavern, tunnel, pipe, channel, metropolitan area, downtown area with relatively tall buildings, valley, well, duct, pathway, gas line, oil line, water pipe, interconnecting network of pathways / alleys / roads / tubes / cavities / caves / pipe-like structures / airspaces / fluid areas, human body, animal body, body cavities, organs, bones, teeth, soft tissue, hard tissue, rigid tissue, non-hard tissue, blood vessels / body fluid ducts, trachea, air ducts, chambers, etc.A location can be indoors or outdoors. A location can include both inside and outside a space. For example, a location can include both inside and outside a building. For example, a location can be a one-story or multi-story building, and part of the building can be underground. The shape of the building can be, for example, circular, square, rectangular, triangular, or irregular. These are just examples. The present disclosure can be used to detect events in other types of locations or spaces.

[0117] The wireless transmitter (e.g., Type 1 device) and / or wireless receiver (e.g., Type 2 device) may be embedded within a portable device (e.g., a module or a device with a module) that can move with the object (e.g., in a previous motion and / or a current motion). The portable device may have a wired connection (e.g., via USB, micro USB, Firewire, HDMI, serial port, parallel port, and other connectors) and / or a wireless connection (e.g., via Bluetooth, Bluetooth Low Energy, etc.).The portable device may be communicatively coupled to the object using Bluetooth Low Energy (BLE), WiFi, LTE, ZigBee, etc. The portable device may be a lightweight device. The portable may be powered by a battery, a rechargeable battery, and / or an AC power source. The portable device may be ultra-small (e.g., sub-millimeter scale 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 bulky (e.g., heavy equipment installed). 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 glasses, smart mirrors, smart antennas, smart batteries, smart lighting, smart pens, smart rings, smart doors, smart windows, smart clocks, small batteries, smart wallets, smart belts, smart handbags, smart clothing / apparel, smart ornaments, smart packaging, smart paper / books / magazines / posters / printed materials / signs / displays / illuminated systems / lighting systems, smart keys / tools, smart bracelets / chains / necklaces / wearables / accessories, smart pads / cushions, smart tiles / blocks / bricks / building materials / other materials, smart trash cans / waste containers, smart food carriages / storage, smart balls / racquets, smart chairs / sofas / beds, smart shoes / footwear / carpets / mats / shoe racks, smart gloves / mittens / rings / handwearware), smart hats / headwear / makeup / stickers / tattoos, smart mirrors, smart toys, smart pills, smart kitchen appliances, smart bottles / food containers, smart tools, smart devices, IoT devices, WiFi-enabled devices, network-enabled devices, 3G / 4G / 5G / 6G-enabled devices, embedded devices, implantable devices, air conditioners, refrigerators, heaters, fireplaces, furniture, ovens, cooking devices, televisions / set-top boxes The portable device may be a TV, DVD player, audio player, video player, remote control, hi-fi, audio device, speaker, lamp / lighting, wall, door, window, roof, roof tile / shingle / structure / attic structure / device / feature / equipment / built-in furniture, lawn mower / garden tool / garage tool, trash can / container, 20-foot / 40-foot container, storage container, factory / manufacturing / production device, repair tool, fluid container, machine, installed machine, vehicle, cart, wagon, warehouse vehicle, automobile, bicycle, motorcycle, boat, large ship, airplane, basket / box / bag / bucket / container, smart plate / cup / bowl / pot / mat / kitchen utensil / kitchen tool / kitchen device / kitchen accessory / cabinet / table / chair / tile / lighting / water pipe / faucet / gas range / oven / dishwasher, etc. The portable device may have a battery that may be replaceable, non-replaceable, rechargeable, and / or non-rechargeable. The portable device may be wirelessly charged. 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 where payment is required. The portable device may have an identification (ID), as described above.

[0118] 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 sick person) falling, rolling, hesitating, pausing, or impact (e.g., a person hitting a punching bag, door, window, bed, chair, table, desk, cabinet, box, another person, animal, bird, fly, table, chair, ball, bowling ball, tennis ball, football, soccer ball, baseball, basketball, volleyball, etc.), two-body actions (e.g., a person releasing a balloon, catching a fish, molding clay, writing a document, typing on a computer, etc.), a car moving in a garage, a person walking around an airport / mall / government building / office / etc. with a smartphone, or an autonomous mobile object / machine moving around (e.g., a vacuum cleaner, utility vehicle, automobile, drone, self-driving car, etc.).

[0119] Tasks or wireless smart sensing tasks include object detection, presence detection, object recognition, object verification, tool detection, tool recognition, tool verification, machine detection, machine recognition, machine verification, human detection, human recognition, human verification, infant detection, infant recognition, infant verification, human breathing detection, motion detection, motion estimation, motion verification, periodic motion detection, periodic motion estimation, periodic motion verification, steady motion detection, steady motion estimation, steady motion verification, periodic steady motion detection, periodic steady motion estimation, periodic steady motion verification, transient motion detection, transient motion estimation, transient motion verification, trend detection, trend estimation, trend verification, breathing detection, breathing estimation, breathing estimation, human biometric detection, human biometric estimation, human biometric verification, environmental information detection, environmental information estimation, environmental information verification, gait detection, gait estimation, gait verification, gesture detection, gesture estimation, gesture verification, machine learning, supervised learning, unsupervised learning, semi-supervised learning, The tasks may include filtering, 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, steady motion detection, periodic steady motion detection, intrusion detection, suspicious motion detection, security, safety monitoring, navigation, guidance, map processing, map correction, irregularity detection, localization, tracking, multiple object tracking, indoor tracking, indoor location, indoor navigation, power transmission, wireless power transmission, object counting, vehicle tracking in a parking garage, patient detection, patient monitoring, patient verification, 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, an edge server, a cloud server, and / or another device.

[0120] The first portion of tasks may include at least one of pre-processing, signal conditioning, signal processing, post-processing, 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, 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, speed detection, speed estimation, object localization, object tracking, navigation, acceleration estimation, acceleration detection, fall detection, change detection, intruder detection, infant detection, infant monitor, patient monitor, object recognition, wireless power transfer, and / or wireless charging.

[0121] The second portion of the tasks may include a smart home task, a smart office task, a smart building task, a smart factory task (e.g., manufacturing using machines or an assembly line), a smart Internet of Things (IoT) task, a smart system task, a smart home action, a smart office action, a smart building action, a smart manufacturing action (e.g., moving materials / components / raw materials to a machine / assembly line), an IoT action, a smart system action, turning on lights, turning off lights, controlling lights, playing a sound clip, playing a sound clip in at least one of the rooms, areas, and / or locations, playing at least one sound clip of a welcome, greeting, farewell, first message, and / or second message associated with the first portion of the task, turning on appliances, turning off appliances, controlling appliances in at least one of the rooms, areas, and / or locations, The control may include at least one of: turning on an electrical system, turning off an electrical system, controlling an electrical system in at least one of the 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 heating device, a stove, an entertainment system, a door, a fence, a window, a garage, a computer system, a network device, a network system, a home appliance, an office equipment, 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.

[0122] Tasks include: detect when the user comes home, detect when the user leaves, detect when the user moves from room to room, detect / control / lock / unlock / open / close / partially open windows / doors / garage doors / blinds / curtains / panels / solar panels / shades, 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 upstairs / downstairs, going out / coming home, in the toilet, etc.), monitor / detect user / pet location, automatically do something when detected, automatically do something for the user when detected, turn lights on / off / dim, turn music / radio / home entertainment system on / off, TV / HiFi / Examples of such actions include turning on / off / adjusting / controlling a set-top box (STB) / home entertainment system / smart speaker / smart device, turning on / off / adjusting an air conditioning system, turning on / off / adjusting a ventilation system, turning on / off / adjusting a heating system, adjusting / controlling curtains / light shades, turning on / off / starting a computer, turning on / off / preheating / controlling a coffee maker / kettle, turning on / off / controlling / preheating a cooker / oven / microwave / other cooking device, checking / adjusting a temperature, checking the weather forecast, checking a phone message box, checking email, performing a system check, controlling / adjusting a system, checking / controlling / preparing / disarming a security system / baby monitor, checking / controlling a refrigerator, making reports (e.g., through a speaker such as Google Home, Amazon Echo, on a display / screen, via a web page / email / messaging system / notification system, etc.).

[0123] For example, when a user arrives home in their car, the task may be to automatically detect that the user or their car is approaching, open the garage door as soon as it is detected, turn on the driveway / garage lights as the user approaches the garage, turn on the air conditioner / heater / fan, etc.When a user enters the home, tasks may automatically include turning on entrance lights, turning off driveway / garage lights, playing a greeting message to welcome the user, turning on music, turning on the radio to the user's preferred radio news channel, opening curtains / blinds, monitoring the user's mood, adjusting the lighting and sound environment depending on the user's mood or current / impending events in the user's daily calendar (e.g., the user is having dinner with his girlfriend in an hour, so turn on romantic lighting and music), warming up food in the microwave that the user prepared that morning, running a diagnostic check of all systems in the home, checking the weather forecast for tomorrow's work, checking news of the user's interest, checking the user's calendar and to-do list for reminders, checking the phone answering system / messaging system / email and verbally reporting using a dialogue system / speech synthesis (e.g., For example, a task could be reminding a user's mother about her birthday and calling her (e.g., using audible tools such as speakers / HiFi / speech synthesis / voice / voice / music / song / sound field / background sound field / dialogue system, etc.; using visual tools such as TV / entertainment system / computer / notebook / smartpad / display / lighting / color / brightness / pattern / symbol; using haptic tools / virtual reality tools / gestures / tools; using smart devices / appliances / utensils / furniture / built-in furniture; using web tools / servers / cloud servers / fog servers / edge servers / home networks / mesh networks; using messaging tools / notification tools / communication tools / scheduling tools / email; using a user interface / GUI; using scents / smells / aromas / tastes; using neural tools / neurological tools; using a combination, etc.); reminding a user's mother about her birthday and calling her (e.g., using a tool for reminding as described above); preparing a report and performing the report (e.g., using a tool for reminding as described above); and preemptively turning on an air conditioner / heater / ventilation system or preemptively adjusting the temperature setting of a smart thermostat.When the user moves from the entrance 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 entrance lights behind the user, turn on the TV and set-top box, set the TV to the user's favorite channel, adjust the appliances according to the user's preferences and conditions / states (e.g., adjust the lights and select / play music to create a romantic atmosphere), etc.

[0124] Another example could be: When a user wakes up in the morning, the tasks could be to detect that the user is moving in the bedroom, open the blinds / curtains, open the windows, turn off the alarm clock, adjust the indoor temperature from a night temperature profile to a day temperature profile, turn on the bedroom lights, turn on the restroom lights as the user approaches the restroom, check a wireless or streaming channel to play the morning news, turn on the coffee maker and preheat water, turn off the security system, etc. When a user is walking from the bedroom to the kitchen, the tasks could be to turn on the kitchen and hallway lights, turn off the bedroom and restroom 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 and open the kitchen window to let in fresh air, unlock the back door so the user can check out the backyard, adjust the kitchen temperature setting, etc. Another example could be: When a user leaves home for work, the tasks may be to detect the user's departure, play farewell and / or goodbye messages, open / close the garage door, turn on / off garage and driveway lights, turn off / dim lights to save energy (in case the user forgets), close / lock all windows / doors (in case the user forgets), turn off appliances (especially stoves, ovens, microwaves), turn on / arm the home security system to protect the home against intruders, adjust the air conditioning / heating / ventilation system to an "away" profile to save energy, send alerts / reports / updates to the user's smartphone, etc.

[0125] Motion includes no motion, static motion, motionless motion, deterministic motion, transient motion, tipping motion, repetitive motion, periodic motion, pseudo-periodic motion, periodic motion associated with breathing, periodic motion associated with heartbeat, periodic motion associated with living things, periodic motion associated with machines, periodic motion associated with man-made objects, periodic motion associated with nature, complex motion with 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, periodic stationary random motion, non-stationary random motion, periodic autocorrelation function (PCF). Stationary random motion with an ACF, random motion with a periodic ACF over a period of time, random motion that is pseudo-stationary over a period of time, random motion with a pseudo-periodic component of the instantaneous ACF over a period of 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, changes in electromagnetic signatures, subsurface motion, earthquake motion, plant motion, animal motion, human motion, normal motion, abnormal motion, dangerous motion, warning motion, suspicious motion, rain, fire, flood, tsunami, explosion, collision, imminent collision, human body motion, head motion, face motion, eye motion, mouth motion, tongue motion, neck motion, finger motion The motions may include at least one of: movements, hand movements, arm movements, shoulder movements, body movements, chest movements, abdominal movements, hip movements, leg movements, foot movements, body joint movements, knee movements, elbow movements, upper body movements, lower body movements, skin movements, subcutaneous movements, subcutaneous tissue movements, blood vessel movements, intravenous movements, organ movements, heart movements, lung movements, stomach movements, intestinal movements, bowel movements, eating movements, breathing movements, facial expressions, eye expressions, mouth expressions, speaking movements, singing movements, eating movements, gestures, hand gestures, arm gestures, keystrokes, typing strokes, user interface gestures, human-machine interaction, walking, dancing movements, coordinated movements, and / or coordinated body movements.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 medium 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 sub-60 GHz radio, and / or another radio.

[0126] A heterogeneous IC may comprise a processor, a memory communicatively coupled to the processor, and an instruction set 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, other programmable logic device, discrete logic, and / or a combination. The heterogeneous ICs are used in broadband networks, wireless networks, mobile networks, mesh networks, cellular networks, wireless local area networks (WLANs), wide area networks (WANs), and metropolitan area networks (MANs), WLAN standards, WiFi, LTE, 802.11 standards, 802.11a, 802.11b, 802.11g, 802.11n, 802.11ac, 802.11ad, 802.11af, 802.11ah, 802.11ax, 802.11ay, mesh network standards, 802.15 standards, 802.16 standards, cellular network standards, 3G, 3.5G, 4G, beyond 4G, 4.5G, 5G, 6G, 7G, 8G, 9G, Bluetooth, It may support Low-Energy (BLE), Zigbee, WiMax, and / or other wireless network protocols.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 (CPU), 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), electrically programmable ROM (EPROM), electrically erasable programmable ROM (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 media known in the art. The instruction sets (machine-executable code) corresponding to the method steps may be directly embedded in hardware, software, firmware, or any combination thereof. The instruction set may be embedded, pre-loaded, loaded at startup, loaded on-the-fly, loaded on demand, pre-installed, installed, and / or downloaded. The presentation may be audiovisual, graphical (e.g., using a GUI), textual, symbolic, or mechanical. Basic Calculations

[0127] The computational workload associated with the method is shared among the processor, the Type 1 heterogeneous wireless device, the Type 2 heterogeneous wireless device, a local server, a cloud server, and another processor. Operations, pre-processing, processing, and / or post-processing can be applied to the data (e.g., TSCI, autocorrelation). The operations can be pre-processing, processing, and / or post-processing. The pre-processing, processing, and / or post-processing can be operations. The operations can include pre-processing, processing, post-processing, computation of operand functions, filtering, linear filtering, nonlinear filtering, folding, grouping, energy consumption, low-pass filtering, bandpass filtering, high-pass filtering, median filtering, rank filtering, quartile filtering, percentile filtering, mode filtering, finite impulse response (FIR) filtering, infinite impulse response (IIR) filtering, moving average (MA) filtering, autoregressive (AR) filtering, autoregressive moving average ... averaging (ARMA) filtering, selective filtering, adaptive filtering, interpolation, decimation, subsampling, upsampling, resampling, time correction, time base correction, phase correction, magnitude correction, phase cleaning, magnitude cleaning, matched filtering, enhancement, restoration, noise removal, smoothing, signal conditioning, enhancement, restoration, spectral analysis, linear transforms, nonlinear transforms, frequency transforms, inverse frequency transforms, Fourier transforms, wavelet transforms, Laplace transforms, Hilbert transforms, Hadamard transforms, trigonometric transforms, sine transforms, cosine transforms, power of two transforms, sparse transforms, graph-based transforms, graph signal processing, fast transforms, transforms combined with zero padding, cyclic padding, padding, zero padding, feature extraction, decomposition, projection, orthogonal projection, non-orthogonal projection, over-complete projectionprojection), eigenvalue decomposition, singular value decomposition (SVD), principal component analysis (PCA), 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, local maximization, local minimization, cost function optimization, neural networks, recognition, labeling, training, clustering, machine learning, supervised learning, unsupervised learning, semi-supervised learning, comparison with another TSCI, similarity score calculation, quantization, vector quantization, matching pursuit, compression, encryption, coding, storage, transmission, normalization, time normalization, frequency domain normalization, classification, clustering, labeling, tagging, learning, detection, estimation, learning network, matching The operations may include ping, remapping, expanding, storing, retrieving, transmitting, receiving, representing, merging, combining, splitting, tracking, monitoring, matched filtering, Kalman filtering, particle filters, interpolation, extrapolation, importance sampling, Monte Carlo sampling, compressed sensing, representing, merging, combining, splitting, scrambling, error protection, forward error correction, do nothing, time-varying processing, adjusted average, weighted average, arithmetic average, geometric average, harmonic average, average over selected frequencies, average over antenna links, logical operations, permutation, combination, sorting, AND, OR, XOR, union, logical and, 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. The operations may be applied jointly to multiple time series or functions.

[0128] Functions (e.g., operand functions) include scalar function, vector function, discrete function, continuous function, polynomial function, feature, characteristic, magnitude, phase, exponential function, logarithmic function, trigonometric function, transcendental function, boolean function, linear function, algebraic function, nonlinear function, piecewise linear function, real function, complex function, vector-valued function, inverse function, derivative of a function, integral of a function, circular function, function of another function, one-to-one function, one-to-many function, many-to-one function, many-to-many function, zero crossing, absolute value function, indicator function, mean, mode, median, range, statistics, variance, arithmetic mean, geometric mean, harmonic mean, trimmed mean, percentile, square, cube, root , power, sine, cosine, tangent, cotangent, secant, cosecant, elliptic functions, parabolic functions, hyperbolic functions, game functions, zeta functions, absolute value, thresholding, 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), stochastic functions, probability functions, random functions, ergodic functions, stationary functions, deterministic functions, periodic 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 transform, sparse transform, projection, decomposition, principal component analysis (PCA), independent component analysis (ICA), neural network, feature extraction, moving function, moving window function for adjacent items in time series, filtering function, convolution, mean function, variance function, statistical function, short-time transform, discrete transform, 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, sparse approximation, arbitrary decomposition, graph-based processing, graph-based transformation, graph signal processing , classification, labeling, learning, machine learning, detection, estimation, feature extraction, learning network, feature extraction, noise removal, signal enhancement, coding, encryption, mapping, remapping, vector quantization, low pass filtering, high pass filtering, bandpass filtering, matched filtering, Kalman filtering, pre-processing, post-processing, particle filter, FIR filtering, IIR filtering, autoregressive (AR) filtering, adaptive filtering, first derivative, higher order derivative, integration, zero crossing, smoothing, median filtering, mode filtering,These may include sampling, random sampling, resampling functions, downsampling, upsampling, interpolation, extrapolation, importance sampling, Monte Carlo sampling, compressed sensing, statistics, short-term statistics, long-term statistics, mean, variance, autocorrelation function, cross-correlation, moment generating function, time averaging, weighted average, special functions, Bessel functions, error functions, complementary error functions, beta functions, gamma functions, entire functions, Gaussian functions, Poisson functions, etc.

[0129] Machine learning, training, discriminative training, deep learning, neural networks, continuous-time processing, distributed computing, distributed storage, and acceleration using GPUs / DSPs / coprocessors / multicores / multiprocessing may be applied to the steps (or each step) of the present disclosure. Frequency transforms may include Fourier transforms, Laplace transforms, Hadamard transforms, Hilbert transforms, sine transforms, cosine transforms, trigonometric transforms, wavelet transforms, integer transforms, power-of-two transforms, zero-padding and combinations of transforms, Fourier transforms with zero padding, and / or other transforms. Fast and / or approximate versions of transforms may be implemented. Transforms may be implemented using floating-point and / or fixed-point arithmetic. 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, zero-padding and a combination of transforms, an inverse Fourier transform with zero padding, and / or another transform. Fast and / or approximate versions of the transforms may be implemented. The transforms may be implemented using floating-point and / or fixed-point arithmetic.

[0130] Quantities from the TSCI can be calculated, including motion, location, map coordinates, height, speed, acceleration, angle of motion, rotation, size, volume, time trend, time profile, periodic motion, frequency, transient, breathing, walking, action, event, suspicious event, dangerous event, alarm event, warning, belief, proximity, collision, power, signal, signal power, signal strength, received signal strength indicator. The statistics may include at least one of: RSSI (Radio Frequency Indicator), signal amplitude, signal phase, signal frequency component, signal frequency band component, channel state information (CSI), map, time, frequency, time frequency, decomposition, orthogonal decomposition, non-orthogonal decomposition, tracking, respiration, heart rate, biometrics, infant, 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, impact event, cyclostationary, environment, vibration, material, surface, 3D, 2D, local, global, presence, and / or another. Sliding Window Algorithm

[0131] The sliding time window may have a time-varying window width. It may be initially small to allow for rapid acquisition and may increase over time to a steady-state size. The steady-state size may relate to the frequency, repetitive motion, transient motion, and / or spatiotemporal information being monitored. Even in the steady state, the window size may be adaptively changed based on battery life, power consumption, available computational power, changes in the amount of target, the nature of the motion being monitored, etc. The time shift between two sliding time windows at adjacent time instances may be constant / variable / locally adaptive over time. If a shorter time shift is used, updates of any monitor may become more frequent, which may be used for fast-changing conditions, object motion, and / or objects. A longer time shift may be used for slower conditions, object motion, and / or objects. The window width / size and / or time shift may be changed according to user request / selection. The time shift may be changed automatically (e.g., controlled by a processor / computer / server / cloud server) and / or adaptively.

[0132] At least one characteristic of the 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). The at least one feature of the function can include 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 local extremum with a negative time offset, a first local extremum with a negative time offset, an nth local extremum with a negative time offset, a constrained (with argument in constraint) maximum, a minimum, a constrained maximum, a constrained minimum, a constrained extremum, a slope, a derivative, a higher order derivative, a maximum slope, a minimum slope, a local maximum slope, a local maximum slope with a positive time offset, a local minimum slope, a constrained maximum slope, a constrained minimum slope, a maximum higher order derivative, a minimum higher order derivative, a constrained 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 constrained zero crossing, a zero crossing of a slope, a zero crossing of a higher order derivative, and / or another feature. At least one argument of the function can be identified that is associated with at least one characteristic of the function, and some quantity (e.g., spatiotemporal information of the object) can be determined based on the at least one argument of the function.

[0133] The features (e.g., features of the movement of an object within a location) may include at least one of instantaneous features, short-term features, repeating features, recurring features, history, incremental features, fluctuation features, deviation features, phase, magnitude, time features, frequency features, time-frequency features, decomposition features, orthogonal decomposition features, non-orthogonal decomposition features, deterministic features, stochastic features, probabilistic features, autocorrelation function (ACF), mean, variance, statistics, duration, timing, trend, periodic features, long-term features, history features, average features, current features, past features, future features, predicted features, location, distance, height, speed, direction, velocity, acceleration, change in acceleration, angle, angular velocity, angular velocity, angular acceleration of the object, change in angular acceleration, object orientation, 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 characteristics of the object.

[0134] At least one local maximum and at least one local minimum of the function can be identified. At least one local signal-to-noise-ratio-like (SNR-like) parameter can be calculated for each pair of adjacent local maxima and minima. The SNR-like parameter can be a function (e.g., linear, logarithmic, exponential, monotonic) of the fraction of the quantity (e.g., power, magnitude, etc.) of the local maximum that is the same quantity as the local minimum. It can also be a function of the difference between the quantity of the local maximum and the quantity of the local minimum that is the same quantity.

[0135] Significant local peaks can be identified or selected. Each significant local peak can be a local maximum of an SNR-like parameter greater than a threshold T1 and / or a local maximum of an amplitude greater than a threshold T2. A persistence-based approach can be used to identify / calculate at least one local minimum and at least one local minimum in the frequency domain. A set of selected significant local peaks can be selected from the set of identified significant local peaks based on a selection criterion. Object features / spatiotemporal information can be calculated based on the selected significant local peaks and a set of frequency values ​​associated with the set of selected significant local peaks.

[0136] In one embodiment, the selection criteria may always correspond to selecting the strongest peak within a range. While the strongest peak may be selected, peaks that are not selected may still be significant (fairly strong). Significant peaks that are not selected may be stored and / or monitored as "saved" peaks for use in future selection within a future sliding time window. As an example, a particular peak (at a particular frequency) may appear uniformly over time. Initially, it may be significant but not selected (because other peaks may be strong). However, later on, the peak may become stronger and more dominant and may be selected. Once "selected," it may regress in time to become "selected" earlier if it was significant but not selected. In such a case, the regressed peak may replace a peak that was previously selected earlier. The replaced peak may be relatively weak or a peak that appears isolated in time (i.e., appears for only a short time in time). In another embodiment, the selection criteria may not correspond to selecting the strongest peak within a range. Instead, one may consider not only the "strength" of a peak but also the "signature" of the peak, which may be a peak that has occurred in the past, especially a peak that has been identified for a long time. For example, if a finite state machine (FSM) is used, the peak(s) may be selected based on the state of the FSM. The decision threshold may be adaptively calculated based on the state of the FSM.

[0137] Similarity scores and / or component similarity scores can be calculated (e.g., by a server, a processor, a Type 1 device, a Type 2 device, a local server, a cloud server, and / or another device) based on pairs of temporally adjacent CIs of a TSCI. The pairs can be from the same sliding window or two different sliding windows. Similarity scores can also be based on pairs of temporally adjacent or non-adjacent CIs from two different TSCIs. The similarity scores and / or component similarity scores can be / can include time reversal resonating strength (TRRS), correlation, cross-correlation, autocorrelation, covariance, cross-covariance, autocovariance, dot product of two vectors, distance score, norm, metric, statistical feature, discriminant score, metric, neural network, deep learning network, machine learning, training, discrimination, weighted averaging, preprocessing, denoising, signal conditioning, filtering, time correction, timing compensation, phase offset compensation, transform, component-wise operation, feature extraction, finite state machine, and / or another score. The features and / or spatiotemporal information can be determined / calculated based on the similarity score. Any threshold can be predetermined, adaptively determined, and / or determined by a finite state machine. The adaptive determination can be based on time, space, location, antenna, path, link, condition, battery life, remaining battery life, available power, available computational resources, available network bandwidth, etc. A threshold can be determined to apply to a test statistic to distinguish between two events (or two conditions, or two situations, or two states) A ​​and B. Data (e.g., CI, channel state information (CSI)) can be collected under A and / or B in a training situation. The test statistic can be calculated based on the data. The distribution of the test statistic under A can be compared to the distribution of the test statistic under B, and the threshold can be selected based on some criteria.Criteria may include maximum likelihood (ML), maximum aposterior 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. Thresholds may be adjusted to obtain different sensitivities to A, B, and / or other events / conditions / situations / states. Threshold adjustments may be automatic, semi-automatic, and / or manual. Threshold adjustments may be applied once, occasionally, frequently, periodically, infrequently, occasionally, and / or on demand. Threshold adjustments may be adaptive. Threshold adjustments may depend on the object, object's movement / location / direction / action within / at / at the location, object characteristics / spatiotemporal information / size / properties / habits / habits / behavior, location, features / fitted furniture / furniture / barriers / materials / machines / creatures / things / objects / boundaries / surfaces / mediums, map, map constraints, event / state / situation / condition, time, timing, duration, current state, past history, user, and / or personal preference, etc.

[0138] A stopping criterion for the iterative algorithm may be that the change in the current parameter (e.g., offset value) during the update during the iteration is less than a threshold value. The threshold value may be 0.5, 1, 1.5, 2, or another number. The threshold adjustment may be adaptive and may change as the iterations progress. For the offset value, the adaptive threshold may be determined based on the task, a specific value of the first time, 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. A local extremum may be determined as a corresponding extremum of the regression function in the regression window. The local extremum 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 sets of associated regression function values ​​associated with a set of time offset values ​​may be within a range from a corresponding extremum of the regression function within the regression window.

[0139] Searching for local extrema includes robust search, minimization, maximization, 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, calculus 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 linear programming, cone programming, integer programming, The methods may include quadratic programming, fractional programming, numerical analysis, simplex algorithms, iterative methods, gradient descent, subgradient methods, coordinate descent, conjugate gradient methods, Newton's algorithm, sequential quadratic programming, interior point methods, ellipsoid methods, reduced gradient methods, quasi-Newton methods, simultaneous perturbation stochastic approximation, interpolation methods, pattern search methods, line search, non-differentiable optimization, genetic algorithms, evolutionary algorithms, dynamic relaxation, hill climbing, particle swarm optimization, gravitational search algorithms, simulated annealing, mimetic algorithms, differential evolution, dynamic relaxation, stochastic tunneling, tabu search, reactive search optimization, curve fitting, least squares methods, simulation-based optimization, calculus of variations, and / or deformation. The search for local extrema may be associated with an objective function, a loss function, a cost function, a utility function, a fitness function, an energy function, and / or an energy function. The regression may be performed using a regression function to fit the sampled data (e.g., CI, features of CI, components of CI) or another function (e.g., an autocorrelation function) within a regression window. In at least one iteration, the length of the regression window and / or the location of the regression window may be varied. The regression function may be a linear function, a quadratic function, a cubic function, a polynomial function, and / or another function.

[0140] The regression analysis may be absolute error, squared error, higher order error (e.g., third order, fourth order, etc.), robust error (e.g., squared error if the error magnitude is small, absolute error if the error magnitude is large, or Type I error if the error magnitude is small, or Type II error if the error magnitude is large), another error, a weighted sum of absolute errors (e.g., for a wireless transmitter with multiple antennas and a wireless receiver with multiple antennas, each pair of transmitter antenna and receiver antenna forms a link. Errors associated with different links may have different weights. One possibility is that some links and / or some components with large noise may have smaller or larger weights.), weighted sum-of-squares error, weighted sum of higher order errors, weighted sum of robust errors, weighted sum of another error, absolute cost, squared cost, higher order cost, robust cost, another cost, weighted sum of absolute costs, weighted sum-of-squares cost, weighted sum of higher order costs, weighted sum of robust costs, and / or weighted sum of another cost.

[0141] The determined regression error may 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. 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 current time offset updated during the iteration. Local extrema may be searched for based on a quantity including the difference between two different errors (e.g., the difference between the absolute error and the squared error). Each of the two different errors may 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. The quantity may be compared to an F-distribution, a central F-distribution, another statistical distribution, a threshold, a threshold associated with a probability, a threshold associated with a probability 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.

[0142] The regression window may be determined based on at least one of the following conditions: the motion of the object, a quantity associated with the object, at least one feature and / or spatiotemporal information of the object associated with the motion of the object, an estimated location of a local extremum, a noise feature, an estimated noise feature, an F-distribution, a central F-distribution, another statistical distribution, a threshold, a preset threshold, a threshold associated with a probability, a threshold associated with a desired probability, a threshold associated with a probability of finding a false peak, a threshold associated with an F-distribution, a threshold associated with a central F-distribution, a threshold associated with another statistical distribution, a condition that the quantity at the window center is a maximum within the regression window, a condition that the quantity at the window center is a maximum within the regression window, a condition that for a particular value of a first time within the regression window, another regression window, there is only one of the local extrema of a particular function, and / or another condition. The width of the regression window may be determined based on the particular local extremum to be searched for. The local extrema include a first local maximum, a second local maximum, a higher local maximum, a first local maximum with a positive time offset, a second local maximum with a positive time offset, a higher 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 higher local maximum with a negative time offset, a first local minimum, a second local minimum, a higher local minimum, a first local minimum with a positive time offset, a second local minimum with a positive time offset, a The local extrema may include a higher local minimum with a negative time offset value, a first local minimum with a negative time offset value, a second local minimum with a negative time offset value, a higher local minimum with a negative time offset value, a first local extrema, a second local extrema, a higher local extrema, a first local extrema with a positive time offset value, a second local extrema with a positive time offset value, a higher local extrema with a positive time offset value, a first local extrema with a negative time offset value, a second local extrema with a negative time offset value, and / or a higher local extrema with a negative time offset value.

[0143] The current parameters (e.g., time offset value) may be initialized based on a target value, a target profile, a trend, a past trend, a current trend, a target speed, a speed profile, a target speed profile, a past speed trend, an object motion, at least one feature and / or spatiotemporal information of the object associated with the object motion, a position quantity of the object, an initial velocity of the object associated with the object motion, a predefined value, an initial width, a duration of the regression window, a value based on the carrier frequency of the signal, a bandwidth of the signal, an antenna quantity associated with the channel, a noise characteristic, and / or an adaptive value. The current time offset may be the center, left, right, and / or another fixed relative location of the regression window.

[0144] In the presentation, the information may be displayed on a map of the location. Information may include location, zone, region, area, modified location, approximate location, location relative to a map of locations, location relative to a segmentation of locations, direction, path, path relative to a map and / or segmentation, path (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 adjusted, where the time window duration may be adaptively adjusted with respect to speed, acceleration, etc.), path history, approximate area / zone along the path, history / summary of past locations, history of past locations of interest, frequently visited areas, customer traffic, crowd distribution, crowd behavior, crowd control information, speed, acceleration, movement statistics, breathing rate, heart rate, presence / absence of movement, presence / absence of people, presence / absence of vital signs, gestures, gesture control (control of devices using gestures), location-based gesture control, location-based action information, identification (ID) of objects involved (e.g., pet, person, autonomous machine / device, vehicle, drone, car, boat, bicycle, autonomous vehicle, f The information may include: a machine with a fan, an air conditioner, a TV, a machine with moving parts), user (e.g., person) identification, user information, user location / speed / acceleration / direction / movement / gesture / gesture control / motion tracking, user ID, user activity, user status, user sleep / rest characteristics, user emotional state, user vital signs, location environment information, location weather information, earthquake, explosion, storm, rain, fire, temperature, collision, impact, vibration, events, door open event, door close event, window open event, window close event, fall event, burning event, freezing event, water-related event, wind-related event, air movement event, accident event, pseudo-periodic event (e.g., running on a treadmill, jumping up and down, skipping rope, somersaults, etc.), crowd event, vehicle event, user gesture (e.g., hand gesture, arm gesture, foot gesture, leg gesture, body gesture, head gesture, face gesture, mouth gesture, eye gesture, etc.).

[0145] Location may be two-dimensional (e.g., by two-dimensional coordinates), three-dimensional (e.g., by three-dimensional coordinates). Location may be relative (e.g., with respect to a map), or relational (e.g., midpoint between point A and point B, around a corner, on the stairs, on a table, on the ceiling, on the floor, on a sofa, near point A, distance R from point A, within radius R from point A, etc.). Location may be displayed in Cartesian coordinates, polar coordinates, and / or another representation. Information (e.g., location) may be marked with at least one symbol. Symbols may change over time. Symbols may flash and / or vibrate, with or without changing color / intensity. Size may change over time. Orientation of a symbol may change over time. The symbols may be numbers reflecting momentary quantities (e.g., user's vital signs / breathing rate / heart rate / gesture / condition / status / action / movement, temperature, network traffic, network connectivity, device / machine status, remaining device power, device status, etc.). The rate of change, size, orientation, color, intensity, and / or symbol may reflect the respective movement. Information may be displayed visually and / or explained verbally (e.g., using pre-recorded audio or speech synthesis). Information may be written in text. Information may also be displayed in mechanical ways (e.g., animated gadgets, moving parts).

[0146] The user interface (UI) device may be a smartphone (e.g., iPhone, Android phone), tablet (e.g., iPad), laptop (e.g., notebook computer), personal computer (PC), device with a graphical user interface (GUI), smart speaker, device with voice / audio / speaker capabilities, virtual reality (VR) device, augmented reality (AR) device, smart car, in-car display, voice assistant, in-car voice assistant, etc. The map may be two-dimensional, three-dimensional, and / or higher dimensional (e.g., time-varying 2D / 3D map). Walls, windows, doors, entrances, exits, and restricted areas may be marked on the map. The map may include a floor plan of the facility. The map may have one or more layers (overlays). The map may be a maintenance map including water pipes, gas pipes, wiring, cabling, air ducts, crawl spaces, ceiling layout, and / or underground layout. The location may be divided into multiple zones, e.g., bedroom, living room, storage room, hallway, kitchen, dining room, lobby, garage, first floor, second floor, restroom, office, conference room, reception area, various office areas, various warehouse areas, various facility areas, etc. The segments / zones / areas may be displayed in a map. Different regions may be color-coded. Different regions may be presented with characteristics (e.g., color, brightness, color depth, texture, animation, flashing, flash rate, etc.). The logical segmentation of the location may be performed using at least one heterogeneous Type 2 device, or server, or cloud server, etc.

[0147] This is an example of the disclosed system, apparatus, and method. Stephan and his family want to install the disclosed wireless motion detection system to detect motion in their 2,000-square-foot, two-story townhouse in Seattle, Washington. Because his house has two floors, Stephan has decided to use one Type 2 device (labeled A) and two Type 1 devices (labeled B and C) on the first floor. The first floor of his house has three main rooms: the kitchen, dining room, and living room, which are aligned in a straight line, with the dining room in the middle. The kitchen and living room are at opposite ends of the house. He has placed 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. By arranging the devices in this way, he is using the motion detection system to practically divide the first floor into three zones: the dining room, the living room, and the kitchen. When motion is detected by the AB and AC pairs, the system will analyze the motion information and associate the motion with one of three zones.

[0148] When Stephan and his family go away for the weekend (e.g., to go camping during a long weekend vacation), Stephan turns on the motion detection system using a mobile phone app (e.g., an Android phone app or an iPhone app). When the system detects motion, an alert signal is sent to Stephan (e.g., an SMS text message, email, push message to the mobile phone app, etc.). If Stephan pays a monthly fee (e.g., $10 / month), a service company (e.g., a security company) receives the alert signal via a wired network (e.g., broadband) or wireless network (e.g., home WiFi, LTE, 3G, 2.5G, etc.) and implements security procedures for Stephan (e.g., calling him to confirm the problem, sending someone to check the house, contacting the police on Stephan's behalf, etc.). Stephan loves his elderly mother and is concerned about her well-being when she is home alone. When his mother is alone and the rest of the family is out (e.g., going to work or shopping, or going on vacation), Stephan turns on the motion detection system using his mobile phone app to make sure his mother is okay. He then uses the mobile app to monitor his mother around the house. When Stephan uses the mobile app to see his mother moving between three areas of the house following her daily routine, Stephan knows his mother is doing well. Stephan is grateful that the motion detection system can help him monitor his mother's well-being while he is out.

[0149] On a typical day, his mother would wake up around 7:00 AM. She would spend about 20 minutes making breakfast in the kitchen. Then, she would eat breakfast in the dining room for about 30 minutes. Then, she would do her daily exercise in the living room and then sit on the living room couch and watch her favorite TV show. The motion detection system allows Stephan to see the timing of movement in each of the three areas of the house. If the movement matches the daily routine, Stephan knows that his mother must be doing well. However, if the movement pattern becomes abnormal (e.g., no movement until 10:00 AM, or she spends too long in the kitchen, or she remains motionless for too long), Stephan would suspect something is wrong and would call his mother to check on her. Stephan could also have someone (e.g., family, neighbors, paid personnel, friends, social workers, service providers) check on his mother.

[0150] One day, Stephan wants to relocate his Type 2 device. He simply unplugs the device from the original AC power plug and plugs it into another AC power plug. He's pleased that our wireless motion detection system is plug-and-play and relocating it doesn't affect the system's operation. It works immediately after powering it on. Later, Stephan is convinced that our wireless motion detection system can truly detect motion with high accuracy and very low alarms, and he can truly monitor motion on the first floor using the mobile app. He decides to install a similar setup (i.e., one Type 2 device and two Type 1 devices) on the second floor to monitor the bedrooms on the second floor. Once again, he finds that system setup is extremely simple; he simply plugs the Type 2 and Type 1 devices into the AC power plugs on the second floor. 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 / second floor can interact with all Type 1 devices on both the first and second floors. Stephen will be pleased to learn that doubling his investment in Type 1 and Type 2 devices will more than double the capacity of the combined system.

[0151] According to various embodiments, each CI (CI) may include at least one of channel state information (CSI), frequency domain CSI, frequency domain CSI associated with at least one subband, time domain CSI, CSI in a domain, channel impulse response (CIR), channel frequency response (CFR), channel characteristics, channel filter response, CSI of a wireless multipath channel, information of the wireless multipath channel, timestamp, aiding information, data, metadata, user data, account data, access data, security data, session data, status data, monitor data, household data, identification information (ID), device data, network data, neighborhood 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 has hardware components (e.g., a wireless transmitter / receiver with an antenna, analog circuitry, a power supply, a processor, a memory, etc.) and corresponding software components. According to various embodiments of the present teachings, the disclosed system includes a bot (referred to as a Type 1 device) and an origin (referred to as a Type 2 device) for detecting and monitoring vital signs, each device comprising a transceiver, a processor, and a memory.

[0152] The disclosed system is applicable in many cases. In one embodiment, a Type 1 device (transmitter) can be a small WiFi-enabled device placed on a table. It can 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 embodiment, a Type 2 device (receiver) can be a WiFi-enabled device placed on a table. It can 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 can be placed in or near a conference room to count people. Type 1 and Type 2 devices can be part of a health monitoring system for elderly people, monitoring their daily activities and any signs of symptoms (e.g., dementia, Alzheimer's disease). Type 1 and Type 2 devices can be used in baby monitors to monitor the vital signs (breathing) of live infants. Type 1 and Type 2 devices can be placed in bedrooms to monitor sleep quality and sleep apnea. Type 1 and Type 2 devices can be placed in cars to monitor the health of passengers and drivers, detect driver sleep, and detect infants left inside the car. Type 1 and Type 2 devices can be used in logistics to prevent human trafficking by monitoring any humans hiding in trucks and containers. Type 1 and Type 2 devices can be deployed by emergency responders in disaster areas to search for victims trapped in rubble. Type 1 and Type 2 devices can be deployed in an area to detect the breathing of any intruders. There are many applications for non-wearable wireless respiratory monitors.

[0153] The hardware modules can be configured to include either a Type 1 transceiver or a Type 2 transceiver. The hardware modules may be sold to or used by various brands to design, build, and sell end products. Products using the disclosed systems and / or methods may be home / office security products, sleep monitor products, WiFi products, mesh products, TVs, STBs, entertainment systems, HiFi, speakers, home appliances, lamps, stoves, ovens, microwaves, tables, chairs, beds, shelves, tools, kitchen utensils, torches, vacuum cleaners, smoke detectors, sofas, pianos, fans, doors, windows, door / window handles, locks, smoke detectors, car accessories, computing devices, office devices, air conditioners, heaters, pipes, connectors, monitor cameras, access points, computing devices, mobile devices, LTE devices, 3G / 4G / 5G / 6G devices, gaming devices, glasses, glass panels, VR goggles, necklaces, watches, waistbands, belts, wallets, pens, hats, wearables, implantable devices, tags, parking tickets, smartphones, etc.

[0154] The summary may include at least one of an analysis, a selected time window, subsampling, a transformation, a projection, etc. The display may include at least one of a monthly chart, a weekly chart, a daily chart, a simplified chart, a detailed chart, a cross-sectional chart, a small form factor chart, a large form factor chart, a color-coded chart, a comparison chart, a summary chart, an animation, a web chart, an audio announcement, and another depiction of cyclical features of recurring motion.

[0155] A Type 1 device and / or a Type 2 device may be an antenna, a device having an antenna, a device having an interface that attaches / connects / links to an antenna, 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 networking capabilities, a device having a wireless integrated circuit (IC), a Wi-Fi device, a device having a Wi-Fi chip (e.g., 802.11a / b / g / n / ac / ax compliant, etc.), a Wi-Fi access point (AP), a Wi-Fi client, a Wi-Fi router, a Wi-Fi repeater, a Wi-Fi hub, a Wi-Fi 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 / 3.5G / 4G / LTE / 5G / 6G / 7G, etc.), a cellular device, a mobile network base station, a mobile network hub, mobile network enabled device, LTE device, device having an LTE module, mobile module (e.g., circuit board having a mobile enabled chip (IC) such as a Wi-Fi chip, LTE chip, BLE chip), device having a Wi-Fi chip (IC), LTE chip, BLE chip, 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 having processor / memory / instruction set, smart clock, smart stationery, smart pen, smart user interface, smart paper, smart mat, smart camera, smart television (TV), set-top box, smart microphone, smart speaker, smart refrigerator, smart oven, smart machine, smartphone, smart wallet, smart furniture, smart door, smart window, smart ceiling, smart floor, smart wall, smart table, smart chair, smart bed, smart nightstand,The type 1 device and / or type 2 device may be a smart air conditioner, smart heater, smart pipe, smart duct, smart cable, smart carpet, smart decoration, smart gadget, smart USB device, smart plug, smart dongle, smart lamp / lighting, smart tile, smart ornament, smart bottle, vehicle, smart car, smart AGV, drone, smart robot, laptop, tablet, computer, hard disk, network card, smart instrument, smart racket, smart ball, smart shoe, smart wearable, smart clothing, smart glasses, smart hat, smart necklace, smart food, smart pill, small device that moves within a living organism (e.g., within blood vessels, lymphatic system, digestive system, etc.). 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, an edge server, a local server, and / or storage.

[0156] In one embodiment, a method / apparatus / system for an object tracking system includes acquiring one or more TSCIs (e.g., 111A, 111B of FIG. 1) of a wireless multipath channel (e.g., 144 of FIG. 1) using a processor (e.g., 100A, 100B), a memory (e.g., 102A, 102B of FIG. 1) communicatively coupled to the processor (e.g., 100A, 100B), and a set of instructions (e.g., 104A, 104B of FIG. 1) stored in the memory (e.g., 102A, 102B). One or more TSCIs (e.g., 300, 111A, 111B) are extracted / derived / obtained via a wireless multipath channel (e.g., 144) from type 1 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B in FIG. 1) located at a first location (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) within a location (e.g., 142 in FIG. 1) and type 2 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B in FIG. 1) located at a second location (114, 115B, 115A, 117, 119, 406, 408, 410, 414) within the location (e.g., 142 in FIG. 1). A wireless multipath channel (e.g., 144) is affected by the current motion (e.g., 120) of an object within a location (e.g., 142). The method / apparatus / system also determines spatiotemporal information (e.g., position, speed, velocity, acceleration, periodic motion, time trend, transient motion, period, etc.) of the object (e.g., 112) based on at least one TSCI (e.g., 300, 111A, 111B), time parameters associated with the current motion (e.g., 120), and / or past spatiotemporal information of the object (e.g., 112). One TSCI (e.g., 300, 111A, 111B) is preprocessed.

[0157] Type 1 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and Type 2 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be different. Different Type 1 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be different. Different Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be different. They may have different quantities of antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F). These antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F) may be different. They may use different circuit boards, different chips, and / or different logic, with different power consumption and / or different price. Each may have different additional and / or upgraded functionality. They may be from different manufacturers, different factories, and / or different brands. They may share some common components (e.g., hardware, antennas, circuits, processors (e.g., 100A, 100B), memory (e.g., 102A, 102B), storage, networks), some common software (e.g., in the devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), in the cloud, etc.), and / or some common modules.

[0158] A distance (e.g., 122) of the current movement (e.g., 120) of the object (e.g., 112) may be determined based on one or more time series of CIs (e.g., 300, 111A, 111B) (e.g., by the object tracking server, the processor (e.g., 100A, 100B), the Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices). An estimated direction (e.g., 123) of the current movement (e.g., 120) of the object (e.g., 112) may be obtained. Spatiotemporal information of an object (eg, 112) may be determined based on the distance (eg, 122) and / or estimated direction (eg, 123) of the current motion (eg, 120) of the object (eg, 112).

[0159] At least one similarity score may be calculated (e.g., by the object tracking server, the processor (e.g., 100A, 100B), the Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices). Each similarity score may be based on a pair of temporally adjacent CIs (e.g., 111A, 111B) of the TSCI (e.g., 300) associated with the current movement (e.g., 120) of the object (e.g., 112). A characteristic similarity score (e.g., a representative similarity score such as the mean, median, mode, weighted average, centroid, percentile, quartile, trimmed mean, etc.) may be calculated based on the at least one similarity score. A distance (e.g., 122) of a current movement (e.g., 120) of an object (e.g., 112) may be determined based on comparing the feature similarity score with a reference decay curve (e.g., a curve generated by a mathematical formula, a curve learned by some learning algorithm, or a combination of both). Spatiotemporal information of the object (e.g., 112) may be determined based on the distance (e.g., 122) of the current movement (e.g., 120) of the object (e.g., 112). At least one latest CI (e.g., 111A, 111B) may be determined (e.g., by the object tracking server, the processor (e.g., 100A, 100B), a Type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), a Type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or another device). Each of the latest CIs (e.g., 111A, 111B) may be the latest in one of one or more TSCIs (e.g., 300). "Latest" may be "current." The latest CI may be the CI with the latest time index. Similarity scores may be calculated for one or more time series. Each similarity score may be calculated based on two CIs (e.g., 111A, 111B) of a particular time series (CI) associated with the similarity score.The two CIs (e.g., 111A, 111B) may be the most recent CI (e.g., 111A, 111B) and a temporally adjacent CI (e.g., 111A, 111B) within a time window associated with the current movement (e.g., 120) of the object (e.g., 120). The set of temporally adjacent CIs used to calculate the similarity score of the time series may include all recent past CIs within the time window. At least one curve may be calculated. Each curve may be determined based on the similarity score of the time series. At least one feature point of the curve may be identified. Spatiotemporal information of the object (e.g., 112) may be determined / calculated based on the at least one feature point. The at least one feature point may include a local maximum, a local minimum, a first local maximum, a second local maximum, another local maximum, a first local minimum, a second local minimum, another local minimum, a zero crossing, a first zero crossing, a second zero crossing, another zero crossing, a point having a predetermined relationship to the second feature point, and / or another feature point.

[0160] At least one similarity score may be calculated (e.g., by the object tracking server, the processor (e.g., 100A, 100B), the type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices). Each similarity score may be calculated based on an initial CI (e.g., 111A, 111B) and a current CI (e.g., 111A, 111B). The initial CI (e.g., 111A, 111B) may be close in time to the start of the current movement (e.g., 120). The current CI (e.g., 111A, 111B) may be close in time to the end of the current movement (e.g., 120) or the present time. A characteristic second similarity score may be determined based on at least one second similarity score. If the characteristic second similarity score is greater than (or “greater than or equal to,” “less than,” or “less than or equal to”) a threshold, the object (e.g., 112) may be determined to be stationary, and the current motion (e.g., 120) may be determined to be zero motion, small motion, and / or negligible motion. Get the initial position of the origin register

[0161] The spatiotemporal information (e.g., initial position 114) of an object (e.g., 112) at a start time (e.g., 116) prior to the current movement (e.g., 120) of the object (e.g., 112) may be determined (e.g., by an object tracking server, a processor (e.g., 100A, 100B), a type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), a type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices) based on other time series (e.g., 300, 111A, 111B) of other CIs (e.g., 111A, 111B) associated with the previous movement of the object (e.g., 112). Another time series of another CI (e.g., 300) may be extracted from a second wireless signal (e.g., 140) transmitted over a wireless multipath channel (e.g., 144) between a second Type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at a third position (e.g., 408) of the location (e.g., 142) and a second Type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at a fourth position (e.g., 410) of the location (e.g., 142). The wireless multipath channel (e.g., 144) may be affected by previous motion of the object (e.g., 112). Initial spatiotemporal information (e.g., initial location 114) of the object (e.g., 112) at an initial time (e.g., 116) may be determined to be a known location associated with the second Type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and / or the second Type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404). A time period may be determined during which essentially all (e.g., all but a few, or all but outliers) of the corresponding CIs (e.g., 111A, 111B) of other time series of other CIs (e.g., 111A, 111B) may be greater than the first adaptive threshold and less than the second adaptive threshold.An initial time (e.g., 116) may be determined as a characteristic point of the period. Spatiotemporal information of an object (e.g., 112) may be determined based on the initial spatiotemporal information (e.g., initial position 114) of the initial (e.g., 112) object (e.g., 112). The characteristic point of the period may be a midpoint, a quartile point, a percentile point, a start point, an end point, a point where other CIs (e.g., 111A, 111B) are local maxima, a point where other CIs (e.g., 111A, 111B) are locally minimal, a point where other CIs (e.g., 111A, 111B) have specific characteristics, and / or other characteristic points.

[0162] At an initial time (e.g., 116) during the previous movement of the object (e.g., 112), one of the second Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the second Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be spatially proximate to and moving with the object (e.g., 112) during the previous movement of the object (e.g., 112). At an initial time (e.g., 116), the second Type 1 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the "other one" of the second Type 2 disparate devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may have a directional antenna. Initial spatiotemporal information (e.g., initial position 114) of an object (e.g., 112) at an initial time (e.g., 116) is acquired from (e.g., object tracking server, processors (e.g., 100A, 100B), type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 3 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 4 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 5 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 6 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), type 7 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 402, 404), and / or by other devices), may be determined to be a known location associated with another one of the second Type 1 disparate devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the second Type 2 disparate devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) having a directional antenna.

[0163] As an example, “one” of the second Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the second Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be Module 1 106A and / or a wireless transmitter (e.g., 108), while “one” of the second Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be Module 1 106A and / or a wireless transmitter (e.g., 108). The "other one" among the first type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the second type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be module 2 106B and / or a wireless receiver (e.g., 109). "One" may be a smartphone and / or other portable device. The object (e.g., 112) may be a person. As the object (e.g., 112) (person) moves around, the object (e.g., 112) (person) may be carrying "one" (a smartphone, spatially close to the person). A previous movement (e.g., walking) may be a person with a smartphone entering a location (e.g., 142) (e.g., an airport, terminal, mall, supermarket, etc.). Thus, "one" may be spatially close to the object (e.g., 112), may move with the object (e.g., 112) during its previous movement, or may be a smartphone or another mobile device spatially close to the object (e.g., 112) (e.g., a person).

[0164] The "other" may be installed at the top, side, or bottom of the entrance, around, or near the entrance. The directional antenna of the "other" may be aimed / pointed at the entrance so as to pick up a special signal carried by the object (e.g., 112) when the "one" passes through the entrance. The initial time (e.g., 116) may be the moment when the person passes through the entrance. As the beam (cone-shaped coverage area) of the directional antenna narrows (e.g., compared to the speed of human movement), the special signal may become stronger for a short period of time (e.g., 0.2 seconds). The initial time (e.g., 116) may be determined as a characteristic point of a period (e.g., the midpoint). The initial spatiotemporal information (e.g., initial location 114) of the person at the initial time (e.g., 116) may be the location of the entrance or a predetermined location near the entrance.

[0165] At a current time during the movement of the object (e.g., 112), "one" of the first Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the first Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be spatially proximate to and moving with the object (e.g., 112) during the object's (e.g., 112) movement. This "one" may be a smartphone. The object (e.g., 112) may be a person. The person may be carrying a smartphone. The movement of the object (e.g., 112) may be the person walking around a location (e.g., 142) (e.g., a mall). "One" (e.g., a smartphone) may be spatially close to and / or move with an object (e.g., 112) (e.g., a person) while the object (e.g., 112) is moving.

[0166] The current time may be real time during the movement of the object (e.g., 112). For example, the initial time (e.g., 116) may be seconds, minutes, hours, days, weeks, months, and / or years before the current time. At the current time, one of the first Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the first Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be communicatively coupled to a network server and / or may be communicatively coupled to a local device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) that may be communicatively coupled to the network server. One and / or the local devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may connect to a cloud server in the Internet using 2.5G / 3G / 4G / LTE / 5G / 6G, WiFi, WiMax, Zigbee, Bluetooth, BLE, wireless networks, etc. One may connect to the cloud server through a local device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), which in turn connects to the cloud server by connecting to the local device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404).

[0167] When "one" (e.g., a smartphone, a handheld / mobile device, a portable gadget communicatively connected to the smartphone via WiFi and / or BLE and / or other communication channels, or a mobile device) moves along with an object (e.g., 112) (e.g., a person) within a location (e.g., 142) (e.g., a mall), the movement of the object (e.g., 112) and / or "one" may affect the wireless signals (e.g., 140) passing through the wireless multipath channels associated with the location (e.g., 142) such that the position (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) of the object (e.g., 112) can be tracked based on the CIs (e.g., 111A, 111B) derived / extracted / obtained / calculated from the wireless signals (e.g., 140). Tracking using antenna matching

[0168] As an example, an object (e.g., 112) may be at a current location L1 (e.g., new location 115B) at a current time T1 (e.g., current time t_N 130). Immediately before the current time T1 (e.g., 130), the object (e.g., 112) may be at a previous location L0 (e.g., previous location 115A) at a previous time T0 (e.g., previous time t_{N-1} 132). One of the Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be in spatial proximity to the object (e.g., 112) and may move with the object (e.g., 112). "One" may include two antennas A1 and A2 that are spatially close to each other. The distance between A1 and A2 may be d. The time difference between T1 and T0 may be t = T1 - T0. The line connecting A1 and A2 is in the direction theta.

[0169] At a time T0 (e.g., 132) immediately before the object (e.g., 112) may be at a previous position L0 (e.g., 115A), two antennas A1 and A2 may be at positions L0A1 and L0A2, respectively, both of which may be close to L0 (e.g., 115A). At time T0, a CI C10A1 associated with antenna A1 is obtained, and a CI C10A2 associated with antenna A2 is obtained.

[0170] At a current time T1 (e.g., 130), when an object (e.g., 112) may be at a previous position L1 (e.g., 115A), two antennas A1 and A2 may be at positions L1A1 and L1A2, respectively, and both of them may be close to L1 (e.g., 115B). At time T1, a CI CI1A1 associated with antenna A1 is obtained, and a CI CI1A associated with antenna A2 is obtained. Spatiotemporal information of the object may be determined based on CI0A1, CI0A2, CI1A1, and CI1A2. The spatiotemporal information may include at least one of a displacement, a distance, a direction, a speed, and an acceleration of a current movement (e.g., current movement 120). CI1A1 and CI1A2 may be compared with CI0A1 and CI0A2.

[0171] If CI1A1 can be matched with CI0A2 (e.g., with a similarity score greater than a certain threshold), it may be an indication that antenna A2 (and "one" and the object itself) may have moved from position L0A2 at time T0 to position L1A1 at time T1. Distance (i.e., spatiotemporal information) may be determined as d, the distance between antennas A1 and A2, i.e., the distance between L0A2 and L1A1. Velocity may be determined as distance d divided by time t = T1 - T0, i.e., velocity = d / (T1 - T0). The axis of movement may be determined as the axis connecting antennas A1 and A2. The direction may be determined as the direction from antenna A2 to A1 (e.g., theta).

[0172] If CI1A2 can be matched to CI0A1 (e.g., with a similarity score greater than a certain threshold), it may be an indication that antenna A1 (and "one" and the object itself) may have moved from position L0A1 at time T0 to position L1A2 at time T1. Distance (i.e., spatiotemporal information) may be determined as d, the distance between antennas A1 and A2, i.e., the distance between L0A1 and L1A2. Velocity may be determined as distance d divided by time t = T1 - T0, i.e., velocity = d / (T1 - T0). The axis of movement may be determined as the axis connecting antennas A1 and A2. The direction may be determined as the direction from antenna A1 to A2.

[0173] The "one" can have three or more antennas. For example, it can be a central antenna surrounded by several other antennas (e.g., N antennas). The central antenna can be antenna A1. One of the other antennas can be antenna A2. Alternatively, two of the other antennas can be antenna A1 and antenna A2. The other antennas can be equally spaced (e.g., evenly spaced on a circle such as angle = 90 degrees for N = 4, angle = 60 degrees for N = 6, angle = 45 degrees for N = 8, angle = 30 degrees for N = 12, etc.) and / or unevenly spaced (e.g., randomly spaced or locally uniformly spaced). The angles maintained by neighboring antennas at the central antenna can be evenly spaced and / or unevenly spaced. For example, the other antennas may be positioned at the vertices of one or more polyhedra (e.g., squares, pentagons, hexagons, heptagons, octagons, non-convex polyhedra, etc.), equidistant from the central antenna (i.e., on a circle) and / or randomly spaced.

[0174] In one example, a "one" may have nine antennas arranged in a tic-tac-toe configuration (on a square or rectangular grid), with a central antenna in the center. A "one" may have nine (or any other number) antennas, one in the center and the rest arranged in a circular, elliptical, or other shape (e.g., concave, convex, non-convex, non-concave) around the center. Shapes around a center, irregular shapes (e.g., shapes around a center, irregular shapes) may be used. In one example, a "one" may have antennas arranged in a three-dimensional arrangement (or constellation). For example, the antennas may be arranged in a three-dimensional square and / or rectangular grid and / or on a sphere (like GPS satellites encircling the Earth). Alternatively, there may be nine antennas, one at the center of a cube and eight at the corners. In one example, a "one" may have a central antenna, one set of antennas arranged in one direction around the central antenna (e.g., N1 antennas), and another set of antennas arranged in another way around the central antenna (e.g., N2 antennas). In one example, the "one" may have multiple center antennas. In one example, the "one" may have movable antennas so that the relative spacing and placement of the antennas can be changed over time. The adaptive spacing and placement of the antennas may depend on current motion, past motion, and / or future predicted motion. In one example, the "multi" may have at least one movable / adjustable barrier around the antennas so that wireless multipath signals can be controlled and / or steered. The barrier can be moved / adjusted so that the multipath signals are focused / steered in a certain direction. In one example, the "one" may have M collinearly arranged antennas A1, A2, ..., M (or A i , i = 1, ..., M). Because they are collinearly arranged, there is a unique "direction," "pointing," and / or "axis" associated with the M antennas. The distance between adjacent antennas is denoted by d 1, d2, …, d M-1 Here, d i is antenna A i and A i + 1 where i = 1, …, M-1. Let d1, d2, …, d M-1may or may not all be equal. If the antennas are evenly spaced, the distances d1, d2, ..., d M - 1 may all be equal.

[0175] N ascending time instances, T1, T2, … , T N (or T i Let t1, t2, …, t N - 1 Here, for i = 1, 2, …, N - 1, t i = T i + 1 -T i The time difference is t1, t2, ..., t N - 1 may or may not be equal over time. Time instances t1, t2, ..., t N - 1 If is a uniform space, the time differences can all be equal. An object (e.g., 112) has positions L1 at T1, L2 at T2, ..., and T N L in N At time T1, the object may be at position L1. i jth antenna A in j Position L i、j Then, for i = 1, …, N, j = 1, …, M, d1, d2, …, d M-1 may vary over time. The orientation of the M antennas may or may not vary over time. Time T i In (i = 1, ..., N, j = 1, ..., M), antenna A j CIs associated with CIs i、j The spatiotemporal information of the object can be obtained by i、j : i = 1, ... N, j = 1, ... , M}. The spatiotemporal information may include at least one of a displacement, a distance, a direction, a velocity, and an acceleration of the current movement (e.g., current movement 120).

[0176] Different time shifts k may be investigated / tested, where k ranges from 1 to N − 1. For time shift k, j 1 = 1, …, M, j 2 = 1, …, M, the CI i、j1 CI i + k、j 2 It can be compared with CI. i、j1 CI i + k、j 2 If it matches (e.g., the similarity score is greater than some threshold T1), then this is i、j 1 is position L i+k、j 2 is identical (or nearly identical) to the antenna A j2 (and the same as "one" and the object itself) at time T i Position L in i、j2 From time T i + k Position L in i + k、j2 (i.e., L i、j1 ) is an indication that the object (together with its antenna) has moved in the "direction" or "point" or "axis" of the antenna. For example, if i = 1, k = 1, j1 = 1, j2 = 2, then CI 1,1 CI 2,2 This corresponds to the fact that antenna A2 is at position L at time T1. 1,2 to position L at T2 1,1 In other words, in one unit of time (since k = 1) (assuming a constant time interval such that 1 unit of time = T2 - T1 = T3 - T2), the object along with its antennas can move in the direction of the antenna axis (1 unit distance = d1 = d2 = d3, assuming the antennas are evenly spaced) per unit of time.

[0177] As another example, when i=1, k=1, j1=3, and j2=4, CI 1,3 CI 3,4 This corresponds to the fact that antenna A4 is at position L at time T1. 1,4 to position L at T3 1,3In other words, in two time units (assuming a constant time interval such that 1 time unit = T2 - T1 = T3 - T2) (since k = 2), the object along with its antennas can move in the direction of the antenna axis (1 unit distance = d1 = d2 = d3, assuming the antennas are evenly spaced) per time unit.

[0178] CI i、j1 and CI i+k、j 2 The similarity score between may include a dot product, a dot product measure, a correlation-based measure, a covariance-based measure, a discrimination score, a distance, and a measure such as Euclidean distance, absolute distance, Lk distance, weighted distance, distance between the first vector and the second vector, and / or another similarity value.

[0179] An aggregate similarity score may be obtained by combining multiple similarity scores, e.g., from multiple time instances, multiple antennas, and / or multiple "others." For example, a combined similarity score may be obtained by combining similarity scores, each of which is a CI for a set of l values. i+l、j 1 and CI i+l+k、j 2 For example, the set may include a subset of {l:l=0, 1, 2, ..., Nlk}

[0180] The combined similarity score may be normalized. It can be normalized by dividing by the cardinality of the set of l values. The combined similarity score may include the mean, median, mode, weighted average, trimmed mean, and / or robust mean of the similarity scores. The combined similarity score may be a combination of the mean, median, mode, weighted average, trimmed mean, and / or robust mean.

[0181] The threshold T1 for similar scores may be adaptively determined. It may depend on the cardinality of the set of l values. If the cardinality of the set of l values ​​is large, the threshold may be smaller. The threshold may depend on a combination of the mean, median, mode, weighted mean, trimmed mean, and / or robust mean. The distance (i.e., possible spatiotemporal information) is determined by L i、j1 and L i、j2 This may be determined as the distance between j2 and j1, which is d j1 + d j1+l +…+ d j2-l If j1 > j2, then d j2 + d j2+l +…+ d j1-l It will be.

[0182] The velocity (i.e., possible spatiotemporal information) is calculated by dividing the distance by T i and T i+k If the time difference is equal, the period can be k time units. i + t i+l + t i+k-l The axis of motion may be determined as the axis connecting the antennas. j2 From A j1 The direction may be determined as the direction to

[0183] Multiple CIs (associated with multiple antennas) can be matched for strong matching. i、j1 CI i+k、j2 (e.g., similarity score is greater than some threshold T2) and CI i、j1+1 CI i+k、j2+1 coincides with, …, and CI i、j1+p CI i+k、j2+P (i.e., CI for r = 0, 1, ..., P simultaneously) i、j1+r CI i+k、j2+r ), which corresponds to position L i、j1+r is the position L for r = 0,1, …,P i+ k、j2+r is the same (or nearly the same) as antenna A. j2+r(and "one" and the object itself) at time T i Position L in i、j2+r From time T i+k Position L in i+k、j2+r (i.e., L i、j1 ), the strong matching threshold T2 can be different from the individual matching threshold T1. Multiple CIs can be matched (e.g., with a similarity score greater than some threshold T3) for a subset of r = 0, 1, ..., P (i.e., CIs i、j1+r is a simultaneous CI for a subset of r = 0, 1, …, P i+k、j2+r , P). This matching is not as strong as matching against the full set of r = 0, 1, ..., P, but is stronger than individual CI matching. Threshold T3 may be different from T1 and / or T2.

[0184] Multiple CIs (associated with multiple instances) can be compared for strong matching. i、j1 CI i+k、j2 (e.g., similarity score is greater than a certain threshold T4) and CI i、j1 CI i+2k、j1+2(j2-j1) coincides with, …, and CI i、j1 CI i+Pk、j1+P(j2-j1) (i.e., for r = 0, 1, ..., P, CI i、j1 CI i+rk、j1+r(j2-j1) ), which corresponds to position L i、j1 is the position L for r = 0,1, …,P i+rk、j1+r(j2-j1) is the same (or nearly the same) as antenna A. j2 (and "one" and the object itself) at time T i Position L in i、j2 From time T i+k Position L in i+k、j2 (i.e., L i、j1 ) is a strong indicator that the antenna A j1+r(j2-j1) (and "one" and the object itself) at time T i Position L i、j1+r(j2-j1) From r=0,1,...,P at time Ti+rk Position L in i+rk、j1+2(j2-j1) (i.e., L i、j1 ) for various antenna movements (r = 0, 1, ..., P). j1+r(J2-j1) ) all point at the same speed and direction, this is a strong match. The threshold T4 for this strong match can be different from the thresholds T1, T2 and / or T3 for the individual matches.

[0185] Multiple CIs can be matched (i.e., CIs) for a subset of r=0, 1, ..., P (e.g., with a similarity score greater than some threshold T5). i、j1 is a subset of r = 0, 1, …, P. i+rk、j1+r(j2ーj1) , P). This matching is not as strong as matching against the full set of r = 0, 1, ..., P, but is stronger than individual CI matching. Threshold T5 may be different from T1, T2, T3, and / or T4. A method / apparatus / system for an object tracking system includes acquiring at least two TSCIs of a wireless multipath channel using a processor, a memory communicatively coupled to the processor, and an instruction set stored in the memory. At least one TSCI is extracted from a wireless signal transmitted over a wireless multipath channel between a Type 1 heterogeneous wireless device at a first position within the location and a Type 2 heterogeneous wireless device at a second position within the location. The wireless multipath channel is affected by the current movement of an object within the location. The Type 1 device is near the object and moves with the object in its current movement. The Type 1 device includes at least two antennas.

[0186] At least two first CIs may be acquired at a first time. Each of the first CIs may be associated with a respective antenna of the Type 1 device. At least two second CIs may be acquired at a second time. Each of the second CIs may be associated with a respective antenna of the Type 1 device. The first time may be close to the second time. At least one similarity score may be calculated. Each similarity score may be between one of the at least two first CIs and one of the at least two second CIs. A match may be identified between a specific first CI associated with the first antenna and a specific second CI associated with the second antenna. The first antenna and the second antenna may be different. Spatiotemporal information of the object may be determined based on at least one of the specific first CI and the specific second CI.

[0187] Distance, direction, and / or other quantities may be determined based on the position of the first antenna and the position of the second antenna. Spatiotemporal information of the object may be determined based on at least one of distance, direction, a first time, and a second time. The spatiotemporal information of the object may be calculated by dividing the distance by the time difference between the first time and the second time. The spatiotemporal information may be the velocity of the object. At least one integrated similarity score may be calculated. The integrated similarity score may be based on at least two similarity scores. The spatiotemporal information of the object may be determined based on the at least one integrated similarity score. The similarity score may include time reversal resonating strength (TRRS), correlation, cross-correlation, autocorrelation, covariance, cross-covariance, autocovariance, dot product of two vectors, distance score, discrimination score, metric, neural network output, deep learning network output, and / or another score.

[0188] The spatial distribution of the at least two antennas may include a centrally located central antenna and the remaining antennas arranged in a two-dimensional and / or three-dimensional configuration around the central antenna. The spatial distribution of the at least two antennas may include a centrally located central antenna and the remaining antennas arranged uniformly around the central antenna. The spatial distribution of the at least two antennas may include some antennas forming a rectangular grid. The spatial distribution of the at least two antennas may include some antennas forming a line. The spatial distribution of the at least two antennas may include some antennas forming a circular grid, an elliptical grid, a convex shape, a non-convex shape, a concave shape, a non-concave shape, and / or an irregular shape.

[0189] At least two first CIs may be obtained for each of a set of time instances, each of the first CIs being associated with a respective antenna of the Type 1 device. At least one similarity score may be calculated between two CIs, each of which is a first CI associated with the first antenna at a first time and a second CI associated with the second antenna at a second time. A match may be identified between a specific first CI associated with a specific first antenna and a specific first time, and a specific second CI associated with a specific second antenna and a specific second time. Spatiotemporal information of the object may be determined based on at least one of the specific first CI, the specific first antenna, the specific first time, the associated first information, the specific second CI, the specific second antenna, the specific second time, and the associated second information.

[0190] At least one combined similarity score may be calculated, each combined similarity score based on at least two similarity scores. Spatiotemporal information of the object may be determined based on the at least one combined similarity score. The similarity score may include time reversal resonating strength (TRRS), correlation, cross-correlation, autocorrelation, covariance, cross-covariance, autocovariance, dot product of two vectors, distance score, discrimination score, metric, neural network output, deep learning network output, and / or another score.

[0191] One of the at least two similarity scores may be a similarity score between a first CI associated with a first time and a second CI associated with a second time. The remaining of the at least two similarity scores may be a similarity score between a CI associated with the first time and a CI associated with the second time. One of the at least two similarity scores may be a similarity score between the first CI associated with the first time and a second CI associated with the second time. The at least two similarity scores may each include at least one similarity score between a third CI and a fourth CI. A difference between a time instance associated with the third CI and a time instance associated with the fourth CI may be equal to a difference between the first time and the second time. One or more time series of power information (PI) may be calculated based on one or more TSCIs (e.g., 300) (e.g., by the object tracking server, the processor (e.g., 100A, 100B), the type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices). Each PI may be associated with a CI (e.g., 111A, 111B). The PI may be a complex number. The real part of the power information (PI) may be calculated based on the magnitude, magnitude squared, phase, real part, imaginary part, and other functions of the CI (e.g., 111A, 111B). A first function of the one or more time series of PI may be calculated. The first function may include computing a function such as an autocorrelation function, a squared autocorrelation function, an autocovariance function, a squared autocovariance function, a dot product, an autocorrelation-like function, and / or a covariance.

[0192] The characteristics of the first function may be determined by (e.g., an object tracking server, a processor (e.g., 100A, 100B), a Type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), a Type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and / or other devices). Type 1 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) are (e.g., object tracking servers, processors (e.g., 100A, 100B), Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), a Type 2 disparate wireless device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and / or other devices), and operates to switch a second Type 2 disparate wireless device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at another location within the location (e.g., 142). A Type 1 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be capable of wireless connection with at least two Type 2 disparate wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404). A second wireless signal (e.g., 140) may be transmitted between the Type 1 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and the second Type 2 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) over a wireless multipath channel (e.g., 144). The wireless multipath channel (e.g., 144) may be affected by the current movement (e.g., 120) of an object within the location (e.g., 142). One or more second TSCIs (e.g., 300, 111A, 111B) (CIs) of the wireless multipath channel (e.g., 144) may be obtained.One or more second TSCIs (e.g., 300, 111A, 111B) may be extracted from the second radio signal (e.g., 140). Spatiotemporal information of the object (e.g., 112) may be determined based on the one or more second TSCIs (e.g., 300, 111A, 111B).

[0193] In another embodiment, an object tracking server includes a processor (e.g., 100A, 100B), a memory (e.g., 102A, 102B) communicatively coupled to the processor (e.g., 100A, 100B), and an instruction set (e.g., 104A, 104B) stored in the memory (e.g., 102A, 102B) and executed by the processor (e.g., 100A, 100B). When the instruction set (e.g., 104A, 104B) is executed by the processor (e.g., 100A, 100B) using the memory (e.g., 102A, 102B), the object (e.g., 112) tracking server is configured to acquire one or more TSCIs (e.g., 300, 111A, 111B) (CIs) of a wireless multipath channel (e.g., 144). At least one TSCI (e.g., 300, 111A, 111B) is connected to a Type 1 heterogeneous wireless device (e.g., 106A, 106B, 108, 109, 113A) at a first location (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) within a location (e.g., 142). , 113B, 400, 402, 404) and a type 2 heterogeneous wireless device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404 in FIG. 1 ) at a second location (114, 115B, 115A, 117, 119, 406, 408, 410, 414) within the location (e.g., 142) via a wireless multipath channel (e.g., 144). The wireless multipath channel (e.g., 144) is affected by the current movement (e.g., 120) of the object (e.g., 112) within the location (e.g., 142). The object (e.g., 112) tracking system determines spatiotemporal information of the object (e.g., 112) based on at least one TSCI (e.g., 300, 111A, 111B), a time parameter associated with the current movement (e.g., 120), and / or past spatiotemporal information of the object (e.g., 112). One TSCI (e.g., 300, 111A, 111B) is pre-processed.The processors (e.g., 100A, 100B) share the computational load with Type 1 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and Type 2 heterogeneous wireless devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404).

[0194] In yet another embodiment, the object tracking server system detects a type 1 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at a first location (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) within a location (e.g., 142), and the type 1 disparate device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) over a wireless multipath channel (e.g., 144) associated with the location (e.g., 142). For example, the location (e.g., 142) may include a Type 2 heterogeneous wireless device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at a second location (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) within the location (e.g., 142) that is wirelessly connected to the object (e.g., 112) and / or an object tracking server (e.g., 112). The object (e.g., 112) tracking server has a processor (e.g., 100A, 100B), a memory (e.g., 102A, 102B) communicatively coupled to the processor (e.g., 100A, 100B), and an instruction set (e.g., 104A, 104B) stored in the memory (e.g., 102A, 102B) and executed by the processor (e.g., 100A, 100B).

[0195] An object (e.g., 112) tracking server, Type 1 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), and Type 2 heterogeneous devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) are configured to acquire at least one TSCI (e.g., 300, 111A, 111B) (CI) of a wireless multipath channel (e.g., 144). At least one TSCI (e.g., 300, 111A, 111B) is extracted from a wireless signal (e.g., 140) transmitted between a Type 1 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and a Type 2 heterogeneous device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404). A wireless multipath channel (e.g., 144) is affected by the current movement (e.g., 120) of an object within a location (e.g., 142).

[0196] In one example, a system, apparatus, and / or method is disclosed for tracking spatiotemporal information (e.g., current movement 120, previous movement 118, position 115A, initial position 114, velocity, speed, acceleration, etc. in FIG. 1 ) of an object (e.g., object 112) in real time (e.g., times t_N 130, t_(N − 1) 132, t_(Nk) 134, t_(Nk − 1) 136 in FIG. 1 ). The system, apparatus, and / or method may include a processor (e.g., 100A, 100B) and a memory (e.g., 102A, 102B) communicatively coupled to the processor (e.g., 100A, 100B). The system, device, and / or method includes acquiring initial spatiotemporal information (e.g., initial position 114, previous position 115A) of an object (e.g., 112) prior to a movement (e.g., 120, 118) of the object (e.g., 112), and acquiring at least one wireless signal (e.g., 140) from a multipath channel (e.g., 144) affected by the movement (e.g., 120, 118) of the object (e.g., 112). The system, device, and / or method further includes extracting TSCI (e.g., 111A, 111B, or channel state information (CSI) 110A / 110B) for the multipath channel (e.g., 144) from the at least one wireless signal (e.g., 140), and determining a distance (e.g., 122) of the movement (e.g., 120) of the object (e.g., 112) based on the time series of CSI (e.g., 300, 110A, 110B). The system, device, and / or method includes estimating a direction (e.g., 123) of a movement (e.g., 120) of an object (e.g., 112) and determining new spatiotemporal information (e.g., a position 115B of the object (e.g., 112) after the movement (e.g., current movement 120) associated with time t_N 130) based on the distance (e.g., 122), direction (e.g., 123), and initial spatiotemporal information (e.g., initial position 114).

[0197] In another example, a system, apparatus, and / or method is disclosed for determining initial spatiotemporal information (e.g., initial position 114) of an object (e.g., 112) prior to a current movement (e.g., 120) of the object (e.g., 112). The system, apparatus, and / or method acquires at least one wireless signal (e.g., 140) from a wireless multipath channel (e.g., 144) affected by a previous movement (e.g., 118) of the object (e.g., 112) prior to the current movement (112) using a processor (e.g., 100A and / or 100B), memory (e.g., 102A and / or 102B) communicatively coupled to the processor (e.g., 100A, 100B), and a set of instructions (e.g., 104A, 104B) stored in the memory (e.g., 102A, 102B). The system, device, and / or method extracts at least one TSCI (e.g., 300, 111A, 111B) of a wireless multipath channel (e.g., 144) from a wireless signal (e.g., 140) and determines initial spatiotemporal information (e.g., initial position 114) of an object (e.g., 112) prior to a current movement (e.g., 120) of the object (e.g., 112) based on the CI (e.g., 300, 111A, 111B) associated with a previous movement (e.g., 118) of the object (e.g., 112). The spatiotemporal information (e.g., current movement 120) of the object (e.g., 112) may be tracked. The system, device, and / or method may make the spatiotemporal information (e.g., 114, 115A) available for tracking another spatiotemporal information (e.g., current movement 120) of the object (e.g., 112).

[0198] A wireless multipath channel (e.g., 144) may be associated with at least one wireless transmitter (e.g., 108) and at least one wireless receiver (e.g., 109). A processor (e.g., 100A, 100B), a memory (e.g., 102A, 102B), and an instruction set (e.g., 104A, 104B) may be associated with one of the at least one wireless transmitter (e.g., 108, 113B) and / or one of the at least one wireless receiver (e.g., 109, 113A). An object (e.g., 112) may be associated with one of the at least one wireless transmitter (e.g., 108, 113B) and / or one of the at least one wireless receiver (e.g., 109, 113A).

[0199] A wireless transmitter (e.g., 108, 113B), a wireless receiver (e.g., 109, 113A), a first device (e.g., 106A / B, 108, 109, 113A, 113B, 400, 402, 404) including a wireless transmitter / receiver, and / or a second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) including a wireless transmitter / receiver and connected to the first device (e.g., via Bluetooth, ZigBee, BLE, WiFi, LTE, 3G / 4G / 5G / 6G / 7G / 8G, etc.) may each be associated with an identifier (ID).

[0200] One of the at least one wireless transmitters (e.g., 118) may have at least one transmitting antenna (e.g., antennas 200A, 200B, 200C, etc. in FIG. 2). One of the at least one wireless receivers (e.g., 109) may have at least one receiving antenna (e.g., antennas 202A, 202B, 202C, etc. in FIG. 2). The TSCI (e.g., 300, 111A and / or 111B) of the wireless multipath channel (e.g., 144) may be extracted from the wireless signal (e.g., 140) of each link (e.g., links 202A, 202B, 202C, 202D, 202E, 202F, 202G, 202J) connecting between one of the at least one transmit antenna (e.g., 200A, 200B, 200C) and one of the at least one receive antenna (e.g., 202A, 202B, 202C). Each CI (e.g., 111A, 111B) (CI) in the time series (e.g., 300) may be associated with a time instance (e.g., times 130, 132, 134, 136) and / or a link (e.g., links 202A, 202B, 202C, 202D, 202E, 202F, 202G, 202J) between a corresponding transmit antenna (e.g., 200A, 200B, 200C) and a corresponding receive antenna (e.g., 202A, 202B, 202C).

[0201] The initial spatiotemporal information (e.g., initial location 114) may be determined based on at least one local characteristic (e.g., temporally local or time-local, spatially local or location-local, frequency-local, and / or some resolution-local, etc.) of at least one TSCI (e.g., 300, 111A, 111B). A time set of at least one time series including a set of time instances (e.g., times 130, 132, 134, 136, etc.) may be identified. At (or around) each time instance of the time set, a CI (e.g., 111A, 111B) (of at least one time series) may have certain local characteristics (e.g., time, space, frequency, and / or some resolution, etc.). The local characteristic may be that each CI (e.g., 111A, 111B) (e.g., a CI can be a quantity related to time, space, frequency, and / or some decomposition) is greater than a first threshold and / or less than a second threshold. The local characteristic may be that each CI (e.g., 111A, 111B) belongs to a set of acceptable information. The local characteristic may also be that a function of one and / or more CIs (e.g., 111A, 111B) is greater than a first threshold and / or less than a second threshold. The first threshold and / or the second threshold may be locally adaptive, with different values ​​at different times. The first threshold and / or the second threshold may be locally adaptive, with different values ​​at different times. The time set may be a continuous period of time. The time set may also be a combination of disjoint continuous periods of time and / or individual time instances. The acceptable information may be continuous and / or discrete. The local properties may include other factors.

[0202] Each CI (e.g., 111A, 111B) of a time series (e.g., 300) may be associated with a time instance (e.g., 301). Initial spatiotemporal information (e.g., initial position 114) may be associated with an initial time (e.g., 116) and / or a time period (e.g., 310 in FIG. 3). The initial time (e.g., 116) may be a function (e.g., 306 in FIG. 3) of the time instances (e.g., 301 in FIG. 3) in that time period. A time period may include the entire time from A to B, such that the initial time (e.g., 116) is between time A and time B. Within a time period (e.g., 310), there may be many time instances (e.g., 301) of at least one TSCI (e.g., 300). The function (e.g., 306) may be a mean, median, mode, quartile, percentile, and / or other function. The initial spatiotemporal information (eg, initial location 114) may be determined based on the CIs (eg, 111A, 111B) within the period (eg, 314) associated with the corresponding time instance within the period (eg, 312).

[0203] The initial spatiotemporal information (e.g., initial position 114) may be determined based on at least one known spatiotemporal information (e.g., positions 117, 119) associated with at least one of the wireless transmitter (e.g., 108) and the wireless receiver (e.g., 109). The initial position (e.g., 114) may be determined based on a location map (e.g., 142). The initial time (e.g., 116) may be significantly before (i.e., a large difference between time t_N 132 and time t_(Nk) 134) and / or immediately before (i.e., a small difference between time t_N 132 and time t_(Nk) 134) the current movement (e.g., 120). The difference between time t_N 132 and time t_(N-k) 134 may be less than one second, between one second and one minute, between one minute and one hour, and / or more than one hour.

[0204] The initial spatiotemporal information (e.g., initial location 114) may be associated with multiple initial time (e.g., 116) instances at corresponding multiple locations (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414). The initial spatiotemporal information (e.g., initial location 114) associated with the initial time (e.g., 116) may be determined based on the multiple initial time (e.g., 116) instances (e.g., 301) and / or the corresponding multiple locations (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414). The multiple initial time (e.g., 116) instances (e.g., 301) and the multiple corresponding spatiotemporal information (e.g., locations 114, 115B, 115A, 117, 119, 406, 408, 410, 414) may be processed using at least one of linear regression, several regressions, optimization, joint optimization, cost minimization, and / or curve fitting, etc. The initial time (e.g., 116) may be determined based on the multiple initial time (e.g., 116) instances and / or the multiple corresponding spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414).

[0205] A wireless multipath channel (e.g., 144) may be associated with at least one wireless transmitter (e.g., 108) and at least one wireless receiver (e.g., 109). The at least one wireless transmitter (e.g., 108) may transmit a wireless signal (e.g., 140) to the at least one wireless receiver (e.g., 109). A processor (e.g., 100AB), memory (e.g., 102A, 102B), and instruction set (e.g., 104A, 104B) may be associated with a particular wireless transmitter (e.g., 108) and / or a particular wireless receiver (e.g., 109). A particular wireless transmitter (e.g., 108) and / or a particular wireless receiver (e.g., 109) may move with an object (e.g., 112) in a previous movement (e.g., 118) of the object (e.g., 112), a current movement (e.g., 120) of the object (e.g., 112), and / or a future movement of the object (e.g., 112).

[0206] A particular wireless transmitter (e.g., 108) and / or a particular wireless receiver (e.g., 109) may be associated with at least one known spatiotemporal information (e.g., location 117, 119). Initial spatiotemporal information (e.g., initial location) may be determined based on the at least one known spatiotemporal information (e.g., location) associated with at least one of the particular wireless transmitter (e.g., 108) and the particular wireless receiver (e.g., 109). The initial spatiotemporal information (e.g., initial location 114) may be one of the at least one known spatiotemporal information (e.g., location).

[0207] The initial spatiotemporal information (e.g., initial position 114) and / or the initial time (e.g., 116) may further be determined based on another specific radio receiver (e.g., 113A) and / or another specific radio transmitter (e.g., 113B) associated with other known spatiotemporal information (e.g., other known positions). The initial spatiotemporal information (e.g., initial position 114, initial time 116) may be jointly determined based on both the specific radio receiver (e.g., 109), the specific radio transmitter (e.g., 108), the other specific radio receiver (e.g., 113A), and / or the other specific radio transmitter (e.g., 113B). The specific radio receiver (e.g., 109), the specific radio transmitter (e.g., 108), the other specific radio receiver (e.g., 113A), and / or the other specific radio transmitter (e.g., 113B) may be associated with at least one time series selected from at least one TSCI (e.g., 300) selected based on at least one criterion. For example, one criterion may be that some CIs (eg, 111A, 111B) of the time series are greater than a third threshold and / or less than a fourth threshold.

[0208] The initial spatiotemporal information (e.g., initial location) may also be determined based on at least one representative CI (e.g., 111A, 111B) in the time series associated with a specific wireless receiver (e.g., 109), a specific wireless transmitter (e.g., 108), another specific wireless receiver (e.g., 113A), and / or another specific wireless transmitter (e.g., 113B). The wireless receiver (e.g., 109) and / or the wireless transmitter (e.g., 108) may each have at least one directional antenna (e.g., 200A, 200B, 200C, 200D, 200E, 200F). The initial spatiotemporal information (e.g., initial location 114) of the object (e.g., 112) may be determined based on at least one characteristic of the at least one directional antenna. The characteristic may be directivity, angle, beamwidth, reference signal strength, directional antenna attenuation characteristics, directional antenna gain characteristics, and / or other characteristics.

[0209] Initial spatiotemporal information (e.g., initial location 114) of the object (e.g., 112) may be determined based on a comparison of a particular TSCI (e.g., 300, 111A, 111B) with stored reference TSCIs (e.g., 300, 111A, 111B) of non-uniform duration. The stored reference time series may have been acquired and / or collected in the past (before the initial time 116). The stored reference time series may or may not be associated with the object (e.g., 112) and / or another past movement (e.g., 118) of the object (e.g., 112). The stored reference time series may or may not be associated with a wireless receiver (e.g., 109) and / or a wireless transmitter (e.g., 108). The comparison may include aligning the particular time series with the stored reference time series and identifying at least one particular moment in the particular time series that is aligned with at least one particular particular moment in the stored reference time series. The matching may be based on dynamic programming, a variant of dynamic programming, dynamic time warping (DTW), and / or a variant of DTW with respect to at least one mismatch measure. Initial spatiotemporal information (e.g., initial location 114) of an object (e.g., 112) may be determined based on CIs at or around a particular special instant in a particular time series. The initial spatiotemporal information (e.g., initial location 114) of an object (e.g., 112) may be determined based on a function of CIs (e.g., 308 in the figure) at the particular special instant in a particular time series. The initial spatiotemporal information (e.g., initial location 114) of an object (e.g., 112) may be determined based on a comparison of at least one function of CIs (e.g., 111A, 111B) at at least one of the particular special instants in a particular time series with a particular threshold. The comparing may include a hard threshold and / or a soft threshold. At least one other special instant in at least one time series may be identified. The initial spatiotemporal information (eg, 114, 115B, 115A, 117, 119, 406, 408, 410, 414) of an object (eg, 112) may be determined based on at least one other particular instant in time.

[0210] The initial spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) may be near one of at least one predetermined location (e.g., 117 and / or 119). In one example, at least one wireless receiver (e.g., 109) and / or at least one wireless transmitter (e.g., 108) may move with the object (e.g., 112) in a previous movement (e.g., 118) of the object (e.g., 112), a current movement (e.g., 120) of the object (e.g., 112), and / or a future movement of the object (e.g., 112). At least one wireless receiver (e.g., 109) and / or at least one wireless transmitter (e.g., 108) may be located at at least one predetermined location (e.g., 117 and / or 119). The initial spatio-temporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) may be associated with a plurality of predetermined spatio-temporal information (e.g., locations). The initial spatio-temporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) may be determined based on the plurality of predetermined spatio-temporal information (e.g., locations). The initial spatio-temporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) may be determined by triangulation and / or other fusion techniques based on the plurality of predetermined spatio-temporal information (e.g., locations). At least one wireless receiver (e.g., 109) and / or at least one wireless transmitter (e.g., 108) may be associated with at least one predetermined spatio-temporal information (e.g., location).

[0211] In another example, a first device (e.g., 106A, 106B, 108, 109, 113A, 113B in FIG. 1 , 400 in FIG. 4 ) having one of a wireless receiver (e.g., 109) and a wireless transmitter (e.g., 108) may move with the object (e.g., 112) in a previous motion (e.g., 118), a current motion (e.g., 120), and / or a future motion. The first device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400) may be associated with first spatiotemporal information (e.g., position) that changes with the position (e.g., 114, 115B, 115A) of the object (e.g., 112) as the object (e.g., 112) moves. A second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), which is the other of the wireless receiver (e.g., 109) and wireless transmitter (e.g., 108), may be located on and / or around a second location (e.g., 406 in FIG. 4). When an object (e.g., 112) moves near the second location (e.g., 406), the first device (e.g., 400) and the second device (e.g., 402) become so close that the CI (e.g., 111A, 111B), such as RSSI, is maximized at the second location (e.g., 117, 119), with the maximum value of the CI (e.g., 111A, 111B) being greater than a threshold. The initial position 114 (e.g., 112) of the object may be determined to be the second position (e.g., 406) at a corresponding initial time (e.g., 116), which is the time instance (e.g., one of 301) associated with the maximum value.

[0212] In another example, a first device (e.g., 400), which is one of a wireless receiver (e.g., 109) and a wireless transmitter (e.g., 108), may move with an object (e.g., 112) in a previous movement (e.g., 118), a current movement (e.g., 120), and / or a future movement. The object (e.g., 112) may move between two devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 402, 404), i.e., a second device (e.g., 402) and a third device (e.g., 404). The first device (e.g., 400) may be wirelessly communicatively coupled to the second device (e.g., 402) and the third device (e.g., 404). In other words, a wireless signal (e.g., 140) may be transmitted between the first device (e.g., 400) and the second device (e.g., 402). Similarly, a wireless signal (e.g., 140) may be transmitted between a second device (e.g., 400) and a third device (e.g., 404). The second device (e.g., 402), which is the other of the wireless receiver (e.g., 109) and wireless transmitter (e.g., 108), may be located on and / or around a second location (e.g., 406). The third device (e.g., 404), which is the other instance of the wireless receiver (e.g., 109) and wireless transmitter (e.g., 108), may be located on and / or around a third location (e.g., 408).

[0213] When an object (e.g., 112) along with a first device (e.g., 400) moves between a second location (e.g., 406) and a third location (e.g., 408), the first device (e.g., 400) and the second device (e.g., 402) may be so close that a CI (e.g., 111A and / or 111B) such as an RSSI corresponding to the second device (e.g., 402) shows a maximum value at a fourth time instance (e.g., 412) at a fourth location (e.g., 410) that is somewhere between the second location (e.g., 406) and the third location (e.g., 408). On the other hand, a first device (e.g., 400) and a third device (e.g., 404) may be so close that a CI (e.g., 111A and / or 111B) such as an RSSI corresponding to the third device (e.g., 404) exhibits a maximum value at a fifth time instance (e.g., 416) at a fifth location (e.g., 416) somewhere between the second location (e.g., 406) and the third location (e.g., 408). If the movement (e.g., 418) of the object (e.g., 112) is perpendicular to a line (e.g., 420) connecting the second location (e.g., 406) and the third location (e.g., 408) (i.e., the angle (e.g., 422) is 90 degrees, as shown in FIG. 4B ), then the fourth time instance (e.g., 412) may be equal to the fifth time instance (e.g., 416) and / or the fourth location (e.g., 410) may be identical to the fifth location (e.g., 414). If the movement (e.g., 418) of the object (e.g., 112) is at an angle (e.g., 422) with respect to a line (e.g., 420) connecting the second position (e.g., 406) and the third position (e.g., 408), such that the angle (e.g., 422) is less than 90 degrees, as shown in FIG. 4A , the fourth time instance (e.g., 412) may precede the fifth time instance (e.g., 416). If the angle (e.g., 422) is greater than 90 degrees, as shown in FIG. 4C , the fourth time instance (e.g., 412) may follow the fifth time instance (e.g., 416). Thus, the direction (e.g., 123) of the movement 418 of the first device (e.g., 400) may be estimated based generally on whether the fourth time instance (e.g., 412) is equal to, before, or after the fifth time instance (e.g., 416).The direction of movement 418 (e.g., 400) of the first device can be further estimated by estimating an angle (e.g., 422) based on the formula: cos θ=(vΔt / L), where v is the instantaneous velocity of the first device (e.g., 400), Δt is the fifth time instance (e.g., 416) minus the fourth time instance (e.g., 412), and L is the distance between the second position (e.g., 406) of the second device (e.g., 402) and the third position (e.g., 408) of the third device (e.g., 404).

[0214] The CIs (e.g., 111A, 111B) may include channel state information (CSIs, e.g., 110A, 110B). The instantaneous velocity v of the first device (e.g., 400) is obtained by comparing the instantaneous time-reversed resonance strength (TRRS) between two adjacent CSIs with a model function, as described in our previous patent application. Similarly, the two adjacent CSIs may be the CSIs between the first device (e.g., 400) and the second device (e.g., 402). The two adjacent CSIs may be the CSIs between the first device (e.g., 400) and the third device (e.g., 404).

[0215] An initial position (e.g., 114) and an initial time (e.g., 116) of the object (e.g., 112) may be determined based on the fourth position (e.g., 410), the fourth time instance (e.g., 412), the fifth position (e.g., 414), and / or the fifth time instance (e.g., 416). The initial position 114 may also be determined based on CIs (e.g., 111A, 111B) associated with the first device (e.g., 400), the second device (e.g., 402), and the third device (e.g., 404). For example, the CIs (e.g., 111A, 111B) may include RSSIs. A reference function (e.g., 111A, 111B) of the CIs may be obtained. The instantaneous CIs (e.g., 111A, 111B) between the first device (e.g., 400) and the second device (e.g., 402) are compared with the RSSI to estimate the instantaneous distance between the first device (e.g., 400) and the second device (e.g., 402). The instantaneous CI (e.g., 111A, 111B) between the first device (e.g., 400) and the third device (e.g., 404) is compared with the RSSI to estimate the instantaneous distance between the first device (e.g., 400) and the third device (e.g., 404). The instantaneous position of the object (e.g., 112) may be calculated based on the two instantaneous distances. The instantaneous position of the object (e.g., 112) may be calculated based on triangulation, triangulation, and other methods.

[0216] An initial time (e.g., 116) may be selected to be the fourth time instance (e.g., 412) such that the instantaneous position of the object (e.g., 112) at the fourth time instance (e.g., 412) is calculated and selected as the initial position 114. An initial time (e.g., 116) may be selected to be the fifth time instance (e.g., 412) such that the instantaneous position of the object (e.g., 112) at the fifth time instance (e.g., 416) is calculated and selected as the initial position 114. An initial time (e.g., 116) may be selected to be the fourth time instance (e.g., 412) and / or the fifth time instance (e.g., 412) such that the instantaneous position of the object (e.g., 112) at the initial time (e.g., 116) is calculated and selected as the initial position 114. Interpolation and / or extrapolation may be used to calculate the instantaneous position of the object (e.g., 112) at the initial time (e.g., 116). If the initial time (eg, 116) is between the fourth time instance (eg, 412) and the fifth time instance (eg, 416), interpolation may be used; otherwise, extrapolation may be used.

[0217] Apart from the second device (e.g., 402) and the third device (e.g., 404), additional similar devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) at nearby locations may be connected to the first device and used to determine the initial location 114 and initial time (e.g., 116) of the first device (e.g., 400). The second device (e.g., 402), the third device (e.g., 404) and / or additional similar devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may also be communicatively connected.

[0218] A second device (e.g., 402) can have two or more antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F) at different locations around a second location (e.g., 406). A third device (e.g., 404) can have two or more antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F) at different locations around a third location (e.g., 408). Multiple lines can be drawn between the antennas of the second device (e.g., 402) and the antennas of the third device (e.g., 404). Similar to lines (e.g., 420), each of these lines can provide a corresponding estimated distance, estimated instantaneous position, and / or estimated angle (e.g., 422) of movement of the first device (e.g., 418), which can be combined to provide greater accuracy to the estimates.

[0219] A wireless receiver (e.g., 109) can receive a wireless signal (e.g., 140) and / or other wireless signals (e.g., 140) from a wireless transmitter (e.g., 108). A wireless receiver (e.g., 109) can receive other wireless signals (e.g., 140) from other wireless transmitters (e.g., 113B). A wireless transmitter (e.g., 108) can transmit a wireless signal (e.g., 140) and / or other wireless signals (e.g., 140) to other wireless receivers (e.g., 113A). A wireless transmitter (e.g., 108), a wireless receiver (e.g., 109), other wireless receivers (e.g., 113A), and / or other wireless transmitters (e.g., 113B) may move with an object (e.g., 112) and / or other objects. The other objects may be tracked. A wireless receiver (e.g., 109) can simultaneously switch from a wireless transmitter (e.g., 108) to other wireless transmitters (e.g., 113B). The wireless transmitter (e.g., 108) may switch from the wireless receiver (e.g., 109) to another wireless receiver (e.g., 113A) at another time. The wireless transmitter (e.g., 108) and the other wireless transmitter (e.g., 113B) may use the same and / or different radios. The wireless receiver (e.g., 109) and the other wireless receiver (e.g., 113A) may use the same and / or different radios. The wireless signal (e.g., 140) and / or the other wireless signal (e.g., 140) may have data embedded therein.The wireless receiver (e.g., 109), the wireless transmitter (e.g., 108), the other wireless receiver (e.g., 113A), and / or the other wireless transmitter (e.g., 113B) may include at least one processor (e.g., 100A and / or 100B), a memory (e.g., 102A, 102B) communicatively coupled to the respective processor (e.g., 100A and / or 100B), and / or respective instructions stored in the memory (e.g., 102A, 102B). and at least one instruction set (e.g., 104A, 104B) that, when executed, causes a processor (e.g., 100A and / or 100B) to perform any and / or all steps for determining spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414), initial spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414), initial time (e.g., 116), direction (e.g., 123), instantaneous position (e.g., 115B, 406, 408, 410, 414), instantaneous angle (e.g., 422), and / or velocity (e.g., 112) of the object.

[0220] A wireless transmitter (e.g., 108), a wireless receiver (e.g., 109), other wireless transmitters (e.g., 113B), and / or other wireless receivers (e.g., 113A) may move with the object (e.g., 112) and / or other objects in a previous motion (e.g., 118), a current motion (e.g., 120), and / or a future motion.

[0221] A wireless receiver (e.g., 109), a wireless transmitter (e.g., 108), another wireless receiver (e.g., 113A), and / or another wireless transmitter (e.g., 113B) may be communicatively connected to one or more nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404). The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may or may not move with the object (e.g., 112). A wireless receiver (e.g., 109), a wireless transmitter (e.g., 108), another wireless receiver (e.g., 113A), and / or another wireless transmitter (e.g., 113B) transmits information associated with at least one TSCI (e.g., 111A, 111B) and / or at least one TSCI (e.g., 300), a wireless transmitter (e.g., 108), a wireless receiver (e.g., 108), another wireless transmitter (e.g., 113B), and / or another wireless receiver (e.g., 113A) to at least one TSCI (e.g., 111A, 111B) and / or nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404). Other wireless transmitters (e.g., 113B) and / or other wireless receivers (e.g., 113A) may be present with nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404).

[0222] The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) can be portable / non-portable / mobile / immobile. The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) can use battery power, solar power, AC power, and / or other power sources. The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) can have replaceable / non-replaceable batteries and / or rechargeable / non-rechargeable batteries. The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be similar to the object (e.g., 112). The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may have the same (and / or similar) hardware and / or software as the object (e.g., 112). The nearby devices (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be devices that connect to WiFi / 3G / 4G / 5G / 6G / Zigbee / Bluetooth / ad hoc networks / other networks, smart speakers, smart watches, smart clocks, smart appliances, smart machines, smart devices, smart tools, smart vehicles, Internet of Things (IoT) devices, internet-enabled devices, computers, portable computers, tablets, and other smart devices and network-enabled devices that have connectivity with other devices.

[0223] At least one processor (e.g., 100A, 100B) associated with a nearby device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) and / or a wireless receiver (e.g., 109), a wireless transmitter (e.g., 108), another wireless receiver (e.g., 113A), another wireless transmitter 113B, and / or a cloud server (in the cloud) may determine initial spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414) of the object (e.g., 112). Two or more of them may jointly determine the initial spatiotemporal information (e.g., 114, 115B, 115A, 117, 119, 406, 408, 410, 414). Two or more of them may share intermediate information in determining the initial spatiotemporal information (eg, initial position 114).

[0224] In one example, a wireless transmitter (e.g., 108) may travel with an object (e.g., 112) in a previous movement (e.g., 118) and / or a current movement (e.g., 120). The wireless transmitter (e.g., 108) may transmit a wireless signal (e.g., 140) to a wireless receiver (e.g., 109) to determine initial spatiotemporal information (e.g., initial position 114) of the object (e.g., 112). The wireless transmitter (e.g., 108) may also transmit a wireless signal (e.g., 140) and / or another wireless signal (e.g., 140) to another wireless receiver (e.g., 113A) to track the current movement (e.g., 120) of the object (e.g., 120). The wireless receiver (e.g., 109) may also transmit a wireless signal (e.g., 140) and / or another wireless signal (e.g., 140) to the wireless transmitter (e.g., 108) and / or another wireless transmitter (e.g., 113B) to track the current movement (e.g., 120) of the object (e.g., 120).

[0225] In another example, a wireless receiver (e.g., 109) may travel with the object (e.g., 112) in a previous movement (e.g., 118) and / or a current movement (e.g., 120). The wireless receiver (e.g., 109) may receive wireless signals (e.g., 140) from a wireless transmitter (e.g., 108) to determine initial spatiotemporal information (e.g., initial position 114) of the object (e.g., 112). The wireless receiver (e.g., 109) may also receive wireless signals (e.g., 140) and / or other wireless signals (e.g., 140) from other wireless transmitters (e.g., 113A) to track the current movement (e.g., 120) of the object (e.g., 120). Coin-sized Bot

[0226] The present disclosure also relates to object tracking using a small (e.g., coin-sized, cigarette-pack-sized, or smaller) bot / origin. In this disclosure, a wireless transmitter (e.g., 108, 113B in FIG. 1 ) and / or a wireless receiver (e.g., 109, 113A) may be part of a small, lightweight first portable device (e.g., 106A, 106B, 108, 109A, 113A, 113B, 400, 402, 404). The first portable device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be wirelessly connected to a second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404).

[0227] The second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be a smartphone, iPhone, Android phone, smart device, smart appliance, smart vehicle, smart gadget, smart TV, smart refrigerator, smart speaker, smart watch, smart glasses, smart pad, iPad, computer, wearable computer, notebook computer, or gateway. The second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404) may be connected to a cloud server, a local server, and / or other servers via the Internet, a wired Internet connection, and / or a wireless Internet connection. The second device may be portable.

[0228] The first portable device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the second device (e.g., 106A, 106B, 108, 109, 113A, 113B, 400, 402, 404), the local server, and / or the cloud server may perform any of the tasks (e.g., obtaining a TSCI (e.g., 300, 111A, 111B), determining features of an object (e.g., 112) associated with the motion of the object (e.g., 112), calculating time series of power information, determining a specific function / functions, etc. Computation and / or storage may be shared for computation, searching for local extrema, classification, identifying specific values ​​of time offset, noise removal, processing, simplification, cleaning, wireless smart sensing tasks, CI (e.g., 111A, 111B) extraction from wireless signals (e.g., 140), switching, segmentation, trajectory estimation, map processing, correction, correction adjustment, adjustment, map correction, error detection, boundary hit check, thresholding, etc.) and information associated with the present disclosure (e.g., TSCI (e.g., 300, 111A, 111B)). CSI-compatible chip

[0229] The present disclosure also relates to communications hardware capable of providing CI. The communications hardware may be a WiFi-enabled chip (e.g., an integrated circuit or IC), a next-generation WiFi-enabled chip, an LTE-enabled chip, a 5G-enabled chip, a 6G / 7G / 8G-enabled chip, etc. The communications hardware calculates the CI and stores the CI (e.g., 111A, 111B) in a buffer memory (e.g., 102A, 102B) to make the CI (e.g., 111A, 111B) available for extraction. The CI (e.g., 111A, 111B) may include data related to channel state information (e.g., CSI, 110A, 110B in FIG. 1) and / or at least one matrix. The at least one matrix can be used for channel equalization, beamforming, etc. Passive Speed ​​Estimation

[0230] Another aspect of the present disclosure relates to estimating one or more characteristics and / or spatiotemporal information (e.g., 144) of an object (e.g., 112), such as speed, acceleration, walk cycle, gait, object motion, and events. More specifically, the present teachings relate to determining at least one characteristic and / or spatiotemporal information (e.g., 144) of an object (e.g., 112) based on time reversal techniques in a rich scattering environment.

[0231] In one embodiment, a system, apparatus, and / or method for determining at least one feature and / or spatiotemporal information (e.g., characteristic 146) of an object (e.g., object (e.g., 112)) associated with the object's (e.g., current motion 120 and / or previous motion (e.g., 118)) includes obtaining one or more TSCIs (e.g., CIs 111A and / or 111B) of a wireless multipath channel (e.g., wireless multipath channel 144) using a processor (e.g., processor 100A and / or 100B), a memory (e.g., memory 102A and / or 102B) communicatively coupled to the processor (e.g., processor 100A and / or 100B), and an instruction set (e.g., instruction set 104A and / or 104B) stored in the memory (e.g., processor 102A and / or 102B). At least one TSCI (e.g., CI 111A and / or 111B), or channel state information 110A and / or 110B, is obtained from a wireless signal (e.g., wireless signal 140) transmitted over a wireless multipath channel (e.g., 144). The wireless multipath channel (e.g., 144) is affected by an object's motion (e.g., 120). The system, apparatus, and / or method further includes determining one or more characteristics and / or spatiotemporal information (e.g., 146) of the object (e.g., 112) associated with the object's motion (e.g., 120) based on the one or more TSCIs (e.g., 110A, 110B, 111A, and / or 111B). The CI (e.g., 111A and / or 111B) may include channel state information (CSI) (e.g., 110A and / or 110B) of the wireless multipath channel (e.g., 144). Each CSI (e.g., 144) of the wireless multipath channel may be associated with time. Each CSI may include a time-domain element, a frequency-domain element, and / or a time-frequency-domain element. Each CSI may include one or more elements. Each element of CSI may be a real number, an imaginary number, a complex number, a magnitude, a phase, a flag, and / or a set.Each CI (eg, 111A, 111B) (CI) can include a vector of complex numbers, a matrix of complex numbers, a set of mixed quantities, and / or at least one multidimensional collection of complex numbers.

[0232] One or more time-series power information may be calculated based on one or more TSCIs (e.g., 300, 111A, 111B) of the wireless multipath channel (e.g., 144). Each power information may be associated with a CI (e.g., 111A, 111B) of the wireless multipath channel (e.g., 144). Both the power information and the CI (e.g., 111A, 111B) may be associated with time. The vector of CIs (e.g., 111A, 111B) and the vector of power information may have the same number of elements. Each real element of the power information vector may be calculated based on the respective magnitude, phase, real part, imaginary part, and / or another function of the corresponding complex element of the CI (e.g., 111A, 111B). In particular, the real element of the power information vector may be based on the square of the magnitude of the corresponding complex element of the channel information (e.g., 111A / B).

[0233] The particular function may be determined based on a first time, a time offset, a first window of power information, and a second window of power information. Although power information is referred to herein, the particular function may be applied to CIs (e.g., 111A, 111B) and / or other information. The first window of power information may be a first subset of a particular time series of power information associated with the first time. The second window of power information may be a second subset of a particular time series of power information associated with a second time at a time offset from the first time. The first window and the second window may be the same size. The second window is offset in time from the first window. The first window and the second window may both be windows of the same particular time series of power information (and / or CIs (e.g., 111A, 111B), and / or other information). The cardinality of the first subset may be equal to the cardinality of the second subset. A particular function may be calculated for at least one value of the first time and for at least one value of the time offset for each of the first times. One or more features and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) may be determined based on an articular function that calculates at least one value of the time offset. The particular value of the first time may be associated with the motion of the object (e.g., 112) and at least one feature and / or spatiotemporal information of the object (e.g., 112). The particular function may be a second function of at least one of a covariance function, a covariance-like function, an autocovariance function, an auto-covariance-like function, a correlation function, a correlation-like function, an autocorrelation function, an auto-correlation-like function, a dot product, another function, a combination of any of these functions with a pre-processing function, a combination of any of these functions with a post-processing function, and a combination of any of these functions with at least one of a frequency decomposition, a time decomposition, a time-frequency decomposition, and another decomposition.

[0234] The second function is noise removal, smoothing, adjustment, enhancement, restoration, feature extraction, weighted averaging, low-pass filtering, bandpass filtering, high-pass filtering, median filtering, rank filtering, quartile filtering, percentile filtering, mode filtering, linear filtering, nonlinear filtering, finite impulse response (FIR) filtering, infinite impulse response (IIR) filtering, moving average (MA) filtering, autoregressive (AR) filtering, and autoregressive moving average (ARMA) filtering. The function may be at least one of: filtering, thresholding, soft thresholding, hard thresholding, soft clipping, maximization, minimization, cost function optimization, neural network, machine learning, supervised learning, unsupervised learning, semi-supervised learning, transform, Fourier transform, Laplace, Hadamard transform, transform, decomposition, selective filtering, adaptive filtering, derivative, first derivative, second derivative, higher order derivative, integral, zero crossing, indicator function, absolute transform, convolution, multiplication, division, pre-processing, post-processing, another transform, another processing, another filter, and a third function. A particular function may be an estimate of a weighted sum of squares of the autocorrelation function (ACF) of the electric field component of the received electromagnetic (EM) wave of the wireless signal (e.g., 140).

[0235] The power information may include a frequency decomposition, a time decomposition, a time-frequency decomposition, and / or another decomposition. The other function may be determined based on a frequency of the frequency decomposition (or a variable / index of the other decomposition), a first time, a time offset, a first window of the power information, and a second window of the power information. The other function may be calculated for at least one value of the frequency of the frequency decomposition. The particular function may be calculated by averaging another function that calculates at least one value of the frequency of the frequency decomposition.

[0236] In another example, the power information may include a frequency subband decomposition. The other function may be determined based on a frequency subband of the frequency decomposition, a first time, a time offset, a first window of power information, and a second window of power information. The other function may be calculated for at least one instance of a frequency subband of the frequency decomposition. The particular function may be calculated by averaging the other function calculated for at least one instance of a frequency subband of the frequency decomposition. In another example, the power information may include a time-frequency decomposition. The other function may be determined based on a time-frequency division of the time-frequency decomposition, a first time, a time offset, a first window of power information, and a second window of power information. The other function may be calculated for at least one instance of a time-frequency division of the time-frequency decomposition. The particular function may be calculated by averaging the other function calculated for at least one instance of a time-frequency division of the time-frequency decomposition.

[0237] Calculating the specific function may include determining a preliminary function based on the first time, the third time, the time offset, the third window of power information, and the fourth window of power information. The third window of power information may be a third subset of the specific time series of power information associated with the third time. The fourth window of power information may be a fourth subset of the specific time series of power information associated with a fourth time offset from the third time. The preliminary function may be calculated for a specific value of the first time and a specific value of the time offset for at least one value of the third time close to the first time. The specific function for a specific value of the first time and a specific value of the time offset may be calculated by averaging the preliminary functions for the specific values ​​of the time offset calculated for the specific value of the first time and at least one value of the third time close to the first time.

[0238] At least one special feature and / or special spatiotemporal information (e.g., local maximum, local minimum, zero crossing, etc.) may be determined / calculated for a particular value of the first time. At least one particular value of the time offset may be identified, each associated with a respective special feature and / or spatiotemporal information (e.g., local maximum, local minimum, zero crossing, etc.) of the particular function for a particular value of the first time. At least one special feature and / or special spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) may be determined based on the at least one particular value of the time offset.

[0239] A search may be performed for a local extremum of a particular function for a particular value of a first time. The search may include the following steps: (1) a current time offset value may be initialized, followed by a recursion; (2) at each iteration of the recursion, a regression window around the current time offset value may be determined; (3) the particular function within the regression window around the current time offset value may be approximated with / by a regression function using regression analysis; (4) a regression error of the regression function with respect to the particular function may be determined; (5) the current time offset value may be updated based on the convexity of the regression function, other characteristics and / or spatiotemporal information of the regression function, and / or the regression error of the regression function with respect to the particular function within the regression window; and (6) performing the recursion iterations until at least one stopping criterion is met.

[0240] A local extremum of the particular function may be determined for a particular value of the first time based on a corresponding extremum of the regression function within the regression window. A particular value of the time offset associated with the local extremum of the particular function may be identified within the regression window. At least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) may be calculated based on the particular value of the time offset.

[0241] The initial current time offset value may be positive, negative, and / or zero. During at least one iteration, the regression window length and / or regression window location may vary. The regression function may be a linear function, a quadratic function, a cubic function, a polynomial function, and / or another function.

[0242] The regression analysis may be performed to obtain an absolute error, a squared error, a higher order error (e.g., third order, fourth order, etc.), a robust error (e.g., squared error if the error magnitude is small, absolute error if the error magnitude is large, or type I error if the error magnitude is small, type II error if the error magnitude is large), another error, a weighted sum of absolute errors (e.g., for a wireless transmitter (e.g., 108) having multiple antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F) and a wireless receiver (e.g., 108) having multiple antennas (e.g., 200A, 200B, 200C, 200D, 200E, 200F), 200B, 200C) and receiver antennas (e.g., 200D, 200E, 200F) form a link. Errors associated with different links may have different weights. One possibility is that some links and / or some components with significant noise may have smaller or larger weights. ), weighted sum-of-squares error, weighted sum of higher-order errors, weighted sum of robust errors, weighted sum of other errors, absolute cost, squared cost, higher-order cost, robust cost, other cost, weighted sum of absolute costs, weighted sum-of-squares cost, weighted sum of higher-order costs, weighted sum of robust costs, and / or weighted sum of other costs. The time offset associated with the maximum regression error of the regression function for a particular function within the regression window may be the current time offset updated during the iteration.

[0243] In one embodiment, if the regression function is locally convex within the regression window, the time offset associated with the maximum regression error of the regression function for a particular function within the regression window can be the updated current time offset. In another embodiment, if the regression function is locally concave within the regression window, the time offset associated with the maximum regression error of the regression function for a particular function within the regression window can be the updated current time offset.

[0244] If the regression function is locally convex within the regression window, at least one of the time offsets associated with both ends of the regression window can be the updated current time offset.If the regression function is locally concave within the regression window, at least one of the time offsets associated with both ends of the regression window can be the updated current time offset.

[0245] At least one of the time offsets associated with both ends of the regression window may be the updated current time offset based on the local convexity of the regression function within the regression window. If the regression is locally convex in one portion of the window and locally concave in another portion of the window, the window size may be reduced. The time offset closest to the time offset associated with the maximum regression error of the regression function for a particular function within the regression window may be the updated current time offset based on the condition.

[0246] One of the at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) may be the speed of the object (e.g., 112). The speed of the object (e.g., 112) may be calculated based on a particular value of a time offset associated with a local extremum of a particular function for a particular value of a first time within the regression window.

[0247] The speed of an object (e.g., 112) at a certain time may be calculated based on a particular value of the time offset associated with a local extremum of the particular function for a particular value of the first time within the regression window, which may be equal to that time.

[0248] The acceleration of an object (e.g., 112) at a particular time may be determined based on at least one value of the speed of the object (e.g., 112) at each time near the particular time. The acceleration of an object (e.g., 112) at a particular time may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112). The acceleration of an object (e.g., 112) at a particular time may be determined based on at least one value of the speed of the object (e.g., 112) at one or more times near the particular time. Determining the acceleration of an object (e.g., 112) at a particular time may include approximating the speed of the object (e.g., 112) at one or more times by a time function, determining one or more slopes of the time function near the particular time, and calculating the acceleration based on the one or more slopes of the time function. The acceleration may be calculated as a function of the one or more slopes of the time function. The function may be a mean, a weighted mean, a median, a mode, a quartile, a percentile, a robust function, a linear function, a non-linear function, and / or another function.

[0249] One of the one or more slopes of the time function may be determined as the difference between two values ​​of the velocity of the object (e.g., 112) at two adjacent times divided by the difference between two adjacent time values. The time function may be a piecewise linear function. The acceleration may be calculated as the slope of the piecewise linear function at a particular time. The one or more times may be within a time window around the particular time. The one or more times may include a time window around the particular time.

[0250] The acceleration of the object (e.g., 112) may be the acceleration of the object (e.g., 112) associated with the motion of the object (e.g., 112) at a particular time. The acceleration of the object (e.g., 112) at a particular time may be determined based on the speed of the object (e.g., 112) calculated at multiple times proximate to the particular time. The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on at least one TSCI (e.g., 300, 111A, 111B).

[0251] The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on a specific function calculated for a specific value of the first time (the specific value of the first time is one of multiple values ​​of time close to the specific time) and at least one value of a time offset. The speed of the object (e.g., 112) may also be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on a time offset associated with a local extremum of the specific function for a specific value of the first time within a regression window (the specific value of the first time is equal to time). The speed of the object (e.g., 112) at a specific value of the first time may be calculated as a quantity divided by a specific value of the time offset associated with a local extremum of the specific function within the regression window.

[0252] In another embodiment, one of the at least one feature and / or spatiotemporal information may be the speed of the object (e.g., 112). The speed of the object (e.g., 112) at a first time may be calculated based on a specific value of a time offset associated with a local extremum of a specific function within the regression window. The local extremum may be a first local maximum with a positive time offset and / or a first local maximum with a negative time offset. The specific function may be a derivative of a specific autocorrelation function with respect to the time offset. The CIs (e.g., 111A, 111B) may be CSIs associated with the frequency decomposition. Each CSI may be a vector of complex components, each corresponding to a frequency in the frequency decomposition. An autocorrelation function for each component may be calculated for each frequency in the frequency decomposition based on a specific time sequence of the CSI. The feature autocorrelation may be calculated as an average of the autocorrelation functions for each component across the frequencies in the frequency decomposition.

[0253] The iteration may be applied at least once to obtain at least one extremum. Each extremum may be associated with a respective current time offset value, a respective regression window, a respective regression function, and / or a respective specific time offset value in each iteration. At least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) may be calculated based on at least one of the specific time offset values ​​of the specific function within the regression window. The iteration may be applied twice to obtain a first local extremum and a second local extremum. The first local extremum may be associated with a first current time offset value, a first regression window, a first regression function, and / or a first specific time offset value. The second local extremum may be associated with a second current time offset value, a second regression window, a second regression function, and / or a second specific time offset value. At least one feature and / or spatiotemporal information of an object (e.g., 112) associated with the motion of the object (e.g., 112) may be calculated based on a first specific time offset value and / or a second specific time offset value of a specific function within a regression window. The iteration may be applied twice to obtain a first local extremum, which is a first local maximum with a positive time offset, and a second local extremum, which is a first local maximum with a negative time offset. The first local extremum may be associated with a first current time offset value, a first regression window, a first regression function, and / or a first specific time offset value. The second local extremum may be associated with a second current time offset value, a second regression window, a second regression function, and / or a second specific time offset value. At least one feature and / or spatiotemporal information of an object (e.g., 112) associated with the motion of the object (e.g., 112) may be calculated based on a first specific time offset value and / or a second specific time offset value of a specific function within a regression window.

[0254] A walking cycle of an object (e.g., 112) may be determined based on at least one value of the object's (e.g., 112) speed and / or at least one value of the object's (e.g., 112) acceleration. At least one walking cycle of an object (e.g., 112) may also be determined based on at least one value of the object's (e.g., 112) speed and / or at least one value of the object's (e.g., 112) acceleration. At least one walking cycle may be determined by detecting a sequence of periods of positive acceleration and periods of negative acceleration. At least one walking cycle may be determined by detecting a sequence of periods of increasing speed and periods of decreasing speed. At least one walking cycle may be determined by detecting a periodic behavior underlying the object's (e.g., 112) speed, a change in the object's (e.g., 112) speed, and / or an acceleration of the object (e.g., 112) during a relevant period. At least one walking cycle may be determined by detecting a periodic behavior underlying a maximum speed of the object (e.g., 112), a minimum speed of the object (e.g., 112), a maximum acceleration of the object (e.g., 112), a minimum acceleration of the object (e.g., 112), a maximum change in speed of the object (e.g., 112), and / or a minimum change in speed of the object (e.g., 112) during an associated time period. At least one walking cycle may be determined to be associated with the time period. A function of the speed of the object (e.g., 112) and / or the acceleration of the object (e.g., 112) may exhibit a local maximum at a location associated with the time period. At least one walking cycle may be determined to be associated with the time period. A function of the speed of the object (e.g., 112) and / or the acceleration of the object (e.g., 112) may be determined to exhibit peak localization having multiple pairs of local maximum and local minimum values ​​at associated multiple pairs of peak locations associated with the time period.

[0255] The period may be determined based on the distance between at least one of two adjacent local maxima, two adjacent local minima, a pair of adjacent local maxima and local minima, two local maxima separated by a predetermined number of maxima, two local minima separated by a predetermined number of minima, and a pair of local maxima and local minima separated by a predetermined number of extrema of the function. The function may include an autocorrelation function, an autocorrelation like function, an autocovariance function, an autocovariance like function, a product of speed and a shifted version of acceleration, a product of speed and a shifted version of a periodic function of known period, a product of acceleration and a shifted version of a periodic function of known period, an optimization of a cost function, a constrained optimization of a cost function, an unconstrained optimization of a cost function, a Fourier transform, a Laplace transform, another transform, and / or another function.

[0256] At least one walking cycle of the object (e.g., 112) may be determined based on at least one value of the speed of the object (e.g., 112) and at least one value of the acceleration of the object (e.g., 112). The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on at least one TSCI (e.g., 300, 111A, 111B). The acceleration of the object (e.g., 112) may be determined based on the speed of the object (e.g., 112). At least one walking cycle of the object (e.g., 112) may be determined based on at least one value of the speed of the object (e.g., 112) and at least one value of the acceleration of the object (e.g., 112). The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on a specific function calculated for a specific value of a first time (the specific value of the first time is one of multiple values ​​of time close to the specific time) and at least one value of a time offset. The acceleration of the object (e.g., 112) may be determined based on the speed of the object (e.g., 112). At least one walking cycle of the object (e.g., 112) may be determined based on at least one value of the speed of the object (e.g., 112) and at least one value of the acceleration of the object (e.g., 112). The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on a time offset associated with a local extremum of a specific function for a specific value of a first time within a regression window (the specific value of the first time is equal to time). The acceleration of an object (eg, 112) may be determined based on the speed of the object (eg, 112).The object's (e.g., 112) walking and / or events may be detected based on at least one of at least one segment of a TSCI (e.g., 300, 111A, 111B), at least one segment of a time-series channel state information (CSI), at least one value of a specific function, at least one value of the object's (e.g., 112) speed, and / or at least one value of the object's (e.g., 112) acceleration, each based on at least one value of the object's (e.g., 112) speed. The object's (e.g., 112) walking and / or events may be detected based on a time trend of at least one of a CI (e.g., 111A, 111B), a channel state information (CSI, e.g., 110A, 110B in FIG. 1), a specific function, the object's (e.g., 112) speed, and / or the object's (e.g., 112) acceleration. Detecting the gait and / or events of the object (e.g., 112) may include using an engine to determine the gait and / or events of the object (e.g., 112) based on a time trend of the object's (e.g., 112) speed. Input to the engine may include at least one of at least one segment of TSCI (e.g., 300, 111A, 111B), at least one segment of time-series channel state information (CSI, e.g., 110A, 110B in FIG. 1), at least one value of a specific function, at least one value of the object's speed, and at least one value of the object's acceleration. The engine may be a neural network and / or a classification engine. The engine may be trained with training inputs in a training phase to generate at least one reference time trend associated with the object's (e.g., 112) known gait and known events. The engine's training inputs may be associated with the object's (e.g., 112) known gait and known events. The training may be at least one of discriminative training, decision feedback training, machine learning, supervised learning, unsupervised learning, shallow learning, deep learning, and / or another training.

[0257] Determining the gait and / or event of the object (e.g., 112) using the engine may include obtaining a time trend, comparing the time trend with each of the reference time trends, and aligning the time trend of the first duration with each of the at least one reference time trend of a different duration. Calculating at least one similarity score between the time trend and each of the aligned reference time trends and / or determining the gait and / or event of the object (e.g., 112) as one of the known gaits and / or one of the known events of the object (e.g., 112) based on the at least one similarity score may further include: obtaining a time trend; comparing the time trend with each of the reference time trends; aligning the time trend of the first duration with each of the at least one reference time trend of a different duration; and determining the gait and / or event of the object (e.g., 112) as one of the known gaits and / or one of the known events of the object (e.g., 112).

[0258] The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on at least one TSCI (e.g., 300, 111A, 111B). The speed of the object (e.g., 112) may be one of at least one feature and / or spatiotemporal information of the object (e.g., 112) associated with the motion of the object (e.g., 112) determined based on a specific function calculated for a specific value of a first time (the specific value of the first time is one of multiple values ​​of time close to the specific time) and at least one value of a time offset.

[0259] At least one threshold may be determined. At least one feature and / or spatiotemporal information of the object (e.g., 112) may be determined based on the at least one threshold. The at least one threshold may be determined based on at least one time series of training CIs (e.g., 111A, 111B) (CIs) of the wireless multipath channel (e.g., 144) using a second processor (e.g., 100A, 100B), a second memory (e.g., 102A, 102B) communicatively coupled to the second processor (e.g., 100A, 100B), and a second instruction set (e.g., 104A, 104B) stored in the second memory (e.g., 102A, 102B). The second processor (e.g., 100A, 100B) may be the first processor (e.g., 100A, 100B), a part of the first processor (e.g., 100A, 100B), and / or connected to the first processor (e.g., 100A, 100B). Some resources (hardware, network, communication, user interface, and / or software) of the second processor (e.g., 100A, 100B) may be shared with the first processor (e.g., 100A, 100B). The second memory (e.g., 102A, 102B) may be the first memory (e.g., 102A, 102B), a part of the first memory (e.g., 102A, 102B), connected to the first memory (e.g., 102A, 102B), and / or shared with the first memory (e.g., 102A, 102B). The second instruction set (e.g., 104A, 104B) may be a part of the first instruction set (e.g., 104A, 104B) and / or the first instruction set (e.g., 104A, 104B). The second instruction set (e.g., 104A, 104B) and the first instruction set (e.g., 104A, 104B) may share some software libraries, some software routines, some hardware resources, some communication resources, and / or other resources. The second instruction set (e.g., 104A, 104B) and the first instruction set (e.g., 104A, 104B) can operate independently, cooperatively, jointly, sequentially, in parallel, alternately, and / or interactively.

[0260] At least one time series of training CIs (e.g., 111A, 111B) (CIs) of a wireless multipath channel (e.g., 144) can be obtained from a second wireless signal (e.g., 140) transmitted through the wireless multipath channel (e.g., 144) during a training phase. The wireless multipath channel (e.g., 144) can be affected by training motion of a second object in the training phase. The training phase can be a training session, which can be performed once, occasionally, periodically, and / or on demand. At least one time series of first training CIs (e.g., 111A, 111B) of the wireless multipath channel (e.g., 144) can be obtained that are associated with target positive training motion of the second object during the training phase. The positive training motion can be target motion that is to be recognized, monitored, measured, studied, processed, detected, estimated, verified, and / or captured. A second training CI (e.g., 111A, 111B) of at least one time series of the wireless multipath channel (e.g., 144) associated with target negative training motion of a second object during the training phase can be obtained. The negative training motion can be target motion that is to be ignored, missed, unmonitored, undetected, unestimated, unrecognized, unverified, uncaptured, unmeasured, and / or unstudy.

[0261] At least one first quantity from at least one time series of first training CI (e.g., 111A, 111B) and / or at least one second quantity from at least one time series of second training CI (e.g., 111A, 111B) may be calculated. The at least one first quantity and / or the at least one second quantity may include movement statistics, position statistics, map coordinate statistics, height statistics, speed statistics, acceleration statistics, movement angle statistics, rotation statistics, size statistics, volume statistics, and time. Trend, time trend statistics, time profile statistics, periodic movement statistics, frequency statistics, transient statistics, breathing statistics, gait statistics, behavior statistics, event statistics, suspicious event statistics, danger event statistics, alarm event statistics, warning statistics, belief statistics, proximity statistics, collision statistics, power statistics, signal statistics, signal power statistics, signal strength statistics, received signal strength indicator (RSSI), signal amplitude, signal phase, signal frequency component, signal frequency band component, channel state information (CSI, e.g., 110A, 110B in Figure 1) (CSI), CSI statistics, map statistics, tim-electronic statistics, frequency statistics, time - Including frequency statistics, decomposition statistics, orthogonal decomposition statistics, non-orthogonal decomposition statistics, tracking statistics, respiration statistics, heart rate statistics, biometric statistics, baby statistics, patient statistics, machine statistics, equipment statistics, temperature statistics, vehicle statistics, parking statistics, venue statistics, lift statistics, elevator statistics, space statistics, road statistics, fluid statistics, residential statistics, room statistics, office statistics, house statistics, building statistics, warehouse statistics, storage statistics, system statistics, ventilation statistics, fan statistics, pipe statistics, duct statistics, people statistics, human statistics, car statistics, boat statistics, truck statistics, plane statistics, drone statistics, downtown statistics, crowd statistics, impulse event statistics, cyclostationary statistics, environmental statistics, vibration statistics, material statistics, surface statistics, 3D statistics, 2D statistics, local statistics, global statistics, presence statistics, and / or other statistics.

[0262] At least one threshold value may be determined based on at least one first amount and / or at least one second amount. The at least one threshold value may be determined such that a first percentage of the first amount is greater than, equal to, and / or less than the first threshold value (but not the at least one threshold value). The at least one threshold value may be determined such that a second percentage of the second amount is greater than, equal to, and / or less than the second threshold value (but not the at least one threshold value). The first threshold value may be greater than, equal to, and / or less than the second threshold value. The first threshold value may be the second threshold value. The first percentage may be greater than, equal to, and / or less than the second percentage.

[0263] A first training CI (e.g., 111A, 111B) of at least one time series of the wireless multipath channel (e.g., 114) may be associated with a training movement of a second object within the monitored area during a training phase. The target positive training movement of the second object may be a training movement of the second object within the monitored area. A second training CI (e.g., 111A, 111B) of at least one time series of the wireless multipath channel (e.g., 144) may be associated with a training movement of a second object outside the monitored area during a training phase. The target negative training movement of the second object may be a training movement of the second object outside the monitored area.

[0264] The second object may be the first object. The second object may be a copy, replacement, backup, and / or replica of the first object. The second object may be another object similar to the first object. The second object may be similar to the first object in terms of structure, size, shape, functionality, periodicity, deformation characteristics, movement characteristics, speed, acceleration, gait, tendencies, habits, radio characteristics, and other characteristics. Handover Technology

[0265] The present disclosure also relates to a wireless sensing handover system, apparatus, and / or method including a Type 1 disparate wireless transceiver wirelessly connected with at least one Type 2 disparate wireless transceiver via a wireless multipath channel (e.g., 144) associated with a location (e.g., 142), and / or at least one Type 2 disparate wireless transceiver at each location of the location (e.g., 142). The system, apparatus, and / or method further includes a processor (e.g., 100A, 100B), a memory (e.g., 102A, 102B) communicatively coupled to the processor (e.g., 100A, 100B), and / or a set of instructions (e.g., 104A, 104B) stored in the memory (e.g., 102A, 102B) that, when executed, causes the processor (e.g., 100A, 100B) to: The method causes the type 1 transceiver to establish a wireless connection with a first type 2 transceiver at a first position in the location (e.g., 142) through a wireless multipath channel (e.g., 144) associated with the location (e.g., 142). The wireless multipath channel (e.g., 144) is affected by an object in the location (e.g., 142). It (i.e., the system, apparatus, and / or method) further causes the type 1 transceiver to transmit a first wireless signal (e.g., 140) to the first type 2 transceiver and / or receive a second wireless signal (e.g., 140) from the first type 2 transceiver. It extracts at least one first CI (e.g., 111A, 111B) from the first wireless signal (e.g., 140) and / or the second wireless signal (e.g., 140). It performs a wireless smart sensing task associated with the object in the location based on the at least one first CI (e.g., 111A, 111B). It further causes the Type 1 transceiver to switch the wireless connection from the first Type 2 transceiver to a second Type 2 transceiver at a second location (e.g., 142) of the location based on at least one first CI (e.g., 111A, 111B).It further causes the Type 1 transceiver to transmit a third wireless signal (e.g., 140) to the second Type 2 transceiver and / or receive a fourth wireless signal (e.g., 140) from the second Type 2 transceiver, extract at least one second CI (e.g., 111A, 111B) from the third wireless signal and / or the fourth wireless signal (e.g., 140), and / or perform a wireless smart sensing task associated with the object (e.g., 112) in the location (e.g., 142) based on the at least one second CI (e.g., 111A, 111B).

[0266] A Type 1 transceiver may wirelessly communicate with multiple, potentially heterogeneous, Type 2 transceivers. A Type 2 transceiver may wirelessly communicate with multiple, potentially heterogeneous, Type 1 transceivers. A wireless multipa...

Claims

1. 1. A system for monitoring a user's sleep movements at a location, comprising: at least one sensor at the location comprising a wireless, non-contact sensor that does not make physical contact with the user; a processor communicatively coupled to the at least one sensor; A memory communicatively coupled to the processor; a set of instructions stored in the memory, the set of instructions, when executed by the processor, causing the processor to: Obtaining a plurality of time series sensing features (TSSFs) associated with sleep movements of the user at the location based on the at least one sensor, At least one TSSF of the plurality of TSSFs, communicating a wireless signal in a wireless multipath channel at the location based on the wireless non-contact sensor; extracting from the wireless signals time series channel information (TSCI) of the wireless multipath channel at the location, each channel information (CI) comprising at least one of channel state information (CSI), a channel impulse response (CIR) or a channel frequency response (CFR); and obtaining the at least one TSSF based on the TSCI; Obtaining the plurality of TSSFs includes: Calculating a first TSSF based on the TSCI, wherein each sensing feature (SF) of the first TSSF includes one of a distance score, a Euclidean distance, a similarity score, a correlation, a covariance, an inner product, a cross product, or a transform calculated based on two adjacent groups of channel information (CI) of the TSCI; calculating a second TSSF comprising a periodicity characteristic associated with a periodicity of the user's sleep movements, the second TSSF being associated with a respiration rate of the user, each SF of the second TSSF including an instantaneous respiration rate calculated based on the TSCI; Calculating a trend function of a time series of the second TSSF by low-pass filtering the second TSSF; and calculating a detrended function of the time series of the second TSSF by subtracting the trend function of the time series from the second TSSF; and monitoring the user's sleep movements based jointly on the plurality of TSSFs, a detrended function of the time series of the second TSSF, and a variance function of the detrended function of the time series of the second TSSF.

2. 2. The system for monitoring a user's sleep movement at a location according to claim 1, wherein monitoring the user's sleep movement comprises: training a sleep classifier based on the data relating to respiration rate variance and respiration rate deviation using machine learning including at least one of supervised learning, unsupervised learning, semi-supervised learning, active learning, reinforcement learning, support vector machine, deep learning, feature learning, clustering, regression, or dimensionality reduction; Calculating a first time function of respiration rate variance by calculating the variance of the detrended respiration rate function within a first sliding time window; Calculating a second time function of respiration rate deviation by calculating a distance between the mean non-rapid eye movement (NREM) respiration rate and a percentile of the respiration rate within a second sliding time window; and jointly recognizing a sleep stage of the user as one of rapid eye movement (REM) or NREM based on the first time function of respiration rate variance and the second time function of respiration rate deviation and based on the sleep classifier.

3. 3. The system of claim 2, The average NREM respiration rate is calculated by identifying peaks in a histogram of the time function of respiration rate during sleep stages in the period; The sleep classifier is further trained based on sleep training data.

4. 2. The system of claim 1, wherein the set of instructions, when executed by the processor, further causes the processor to: calculating at least one of a time series of sleep stages or a time series of sleep sub-stages based on the plurality of TSSFs, wherein at least one of the sleep stages is associated with rapid eye movement (REM) or non-REM (NREM); The system, wherein monitoring the sleep behavior includes monitoring at least one of the user's sleep timing, sleep duration, sleep stages, sleep states, sleep quality, sleep apnea, sleep problems or sleep disorders.

5. 5. The system of claim 4, wherein the set of instructions, when executed by the processor, further causes the processor to: and recognizing the sleep stages or the sleep sub-stages together, wherein the sleep stages or the sleep sub-stages are recognized as at least one of REM, NREM, light sleep, deep sleep, sleep apnea, insomnia, hypersomnia, parasomnia, sleep interruptions, nightmares, sleep walking, tossing and turning, sleep problems, sleep states, or sleep behaviors when the user's sleep stage is recognized as sleep.

6. 6. The system of claim 5, wherein the set of instructions, when executed by the processor, further causes the processor to: calculating at least one detrended statistic including at least one of a mean, a weighted mean, a variance, or a deviation based on a detrended function of the time series of the second TSSF within each sliding time window; and recognizing the sleep stage or the sleep sub-stage together based on the at least one detrended statistic, wherein the sleep stage or the sleep sub-stage is recognized as at least one of REM, NREM, light sleep, deep sleep, sleep apnea, insomnia, hypersomnia, parasomnia, sleep interruptions, nightmares, sleep walking, tossing and turning, sleep problems, sleep states, or sleep behaviors.

7. 5. The system of claim 4, wherein the set of instructions, when executed by the processor, further causes the processor to: calculating each periodicity feature of the second TSSF based on at least one of an autocorrelation function (ACF), a frequency spectrum, or a frequency transform of channel information (CI) of the TSCI in a sliding time window; each periodicity feature is set to a default value if the periodicity is not detected in the sleep movements of the user; The system, wherein each periodicity feature includes at least one of a frequency, a phase, a rate, a frequency index, a period, or a time index.

8. 8. The system of claim 7, wherein the set of instructions, when executed by the processor, further causes the processor to: monitoring the sleep movements of the user by jointly recognizing a sleep stage of the user based on the first TSSF and the second TSSF; Calculating motion statistics based on the sensing features of the first TSSF; and calculating periodicity statistics based on periodicity characteristics of the second TSSF; and jointly recognizing the sleep stage based on the motion statistics and the periodicity statistics as either asleep or awake; The motion statistics may include: Sensing feature ratio, motion ratio, Percentage of time non-sleep motion is detected; Percentage of time non-sleep motion is detected in a sliding time window; at least one of: a percentage of time that the function of the sensing feature exceeds a threshold value; or a percentage of time that the function of the sensing feature exceeds the threshold value over a sliding time window; The periodicity statistics may be Periodicity feature ratio, breathing ratio, heart rate ratio, the percentage of time that the periodicity is detected; the percentage of time in a sliding time window that the periodicity is detected; the percentage of time that the periodicity feature is not the default value or the percentage of time that the periodicity feature is not the default value over a sliding time window.

9. 9. The system of claim 8, if the motion statistics or periodicity statistics satisfy a first integration criterion, the sleep stages are jointly recognized as sleep; If the motion statistics or periodicity statistics meet a second integration criterion, the sleep stage is jointly recognized as an awakening.

10. 10. The system of claim 9, (a) the first integration criterion is: (a1) the motion statistics are smaller than a first threshold and the periodicity statistics are larger than a second threshold; and (a2) the second integration criterion is: the motion statistics are larger than the first threshold or the periodicity statistics are smaller than the second threshold; (b) the first integration criterion is: (b1) the motion statistics are less than a first threshold or the periodicity statistics are greater than a second threshold; and (b2) the second integration criterion is: the motion statistics are greater than the first threshold and the periodicity statistics are less than the second threshold. A system, either (a) or (b).

11. 11. The system of claim 10, wherein the set of instructions, when executed by the processor, further causes the processor to: calculating a sleep analysis based on at least one of any sleep stage, any sleep sub-stage, the first TSSF, the second TSSF, the plurality of TSSFs, or any associated timing; The system, wherein the sleep analysis includes at least one of a sleep quality score, a sleep quantity score, timing, duration, or a linear combination of periods of sleep with a positive coefficient and periods of awake with a negative coefficient.

12. A method for a system for monitoring a user's sleep movements at a location, wherein each step of the method is performed by the system, comprising: communicating a wireless signal in a wireless multipath channel of the location in a contactless manner based on a wireless contactless sensor, the wireless contactless sensor not being in physical contact with the user; obtaining time series channel information (TSCI) of the wireless multipath channel at the location extracted from the wireless signal, each channel information (CI) comprising at least one of a channel state information (CSI), a channel impulse response (CIR) or a channel frequency response (CFR); acquiring a plurality of time series sensing features (TSSFs) associated with the user's sleep movements at the location; Calculating a first TSSF of the plurality of TSSFs based on the TSCI, the first TSSF comprising motion characteristics associated with the sleep motion of the user; calculating a second TSSF of the plurality of TSSFs comprising a periodicity characteristic associated with a periodicity of the sleep movement of the user based on the TSCI, the second TSSF being associated with a respiration rate of the user; and monitoring the user's sleep movements based jointly on the plurality of TSSFs; Monitoring the sleep movements of the user includes: using machine learning to train a sleep classifier based on data related to respiration rate variance and respiration rate deviation; calculating a first time function of respiration rate variance by calculating the variance of a detrended function of the second TSSF within a first sliding time window; Calculating a second time function of respiration rate deviation by calculating a distance between the mean non-rapid eye movement (NREM) respiration rate and a percentile of the respiration rate within a second sliding time window; and jointly recognizing a sleep stage of the user based on the first time function of respiration rate variance and the second time function of respiration rate deviation and based on the sleep classifier.

13. 13. The method of claim 12 further comprising: Calculating a trend function of a time series of the second TSSF by filtering the second TSSF, the filtering comprising one of low-pass filtering, band-pass filtering, or high-pass filtering; calculating the detrended function of the time series of the second TSSF by subtracting the trend function of the time series from the second TSSF; monitoring the sleep movements of the user based on the plurality of TSSFs together with a detrended function of the time series of the second TSSFs and time functions of variance and deviation of the detrended function of the second TSSFs; A method in which each sensing feature (SF) of the first TSSF includes one of a distance score, a Euclidean distance, a similarity score, a correlation, a covariance, an inner product, a cross product or a transformation calculated based on two adjacent groups of channel information (CI) of the TSCI.

14. 14. The method of claim 13, The average NREM respiration rate is calculated by identifying peaks in a histogram of the time function of respiration rate during sleep stages in the period; The method, wherein the sleep classifier is further trained based on sleep training data.

15. 15. The method of claim 14, further comprising: calculating at least one of a time series of sleep states or a time series of sleep sub-states based on the plurality of TSSFs, wherein at least one of the sleep states is associated with rapid eye movement (REM) or non-REM (NREM); The method, wherein monitoring the sleep behavior includes monitoring at least one of the user's sleep timing, sleep duration, sleep stages, sleep states, sleep quality, sleep apnea, sleep problems or sleep disorders.

16. 16. The method of claim 15 further comprising: and recognizing the sleep state or the sleep sub-state together, wherein the sleep state or the sleep sub-state is recognized as at least one of REM, NREM, light sleep, deep sleep, sleep apnea, insomnia, hypersomnia, parasomnia, sleep interruptions, nightmares, sleep walking, tossing and turning, sleep problems, sleep states, or sleep behaviors when the user's sleep state is recognized as sleep.

17. 17. The method of claim 16 further comprising: calculating at least one detrended statistic including at least one of a mean, a weighted mean, a variance, or a deviation based on a detrended function of the time series of the second TSSF within each sliding time window; and recognizing the sleep state or the sleep sub-state together based on the at least one detrended statistic, wherein the sleep state or the sleep sub-state is recognized as at least one of REM, NREM, light sleep, deep sleep, sleep apnea, insomnia, hypersomnia, parasomnia, sleep interruptions, nightmares, sleep walking, tossing and turning, sleep problems, sleep states, or sleep behaviors.

18. 18. The method of claim 17, further comprising: Calculating each periodicity feature of the second TSSF based on at least one of an autocorrelation function (ACF), a frequency spectrum, or a frequency transform of channel information (CI) of the TSCI in a sliding time window; each periodicity feature is set to a default value if the periodicity is not detected in the sleep movements of the user; A method, wherein each periodicity feature includes at least one of a frequency, a phase, a rate, a frequency index, a period, or a time index.

19. 20. The method of claim 18, further comprising: Monitoring the user's sleep movements by jointly recognizing the user's sleep state based on the first TSSF and the second TSSF; Calculating motion statistics based on the sensing features of the first TSSF; and calculating periodicity statistics based on periodicity characteristics of the second TSSF; and and jointly recognizing the sleep state based on the motion statistics and the periodicity statistics as either asleep or awake; The motion statistics may include: Sensing feature ratio, motion ratio, Percentage of time non-sleep motion is detected; Percentage of time non-sleep motion is detected in a sliding time window; at least one of: a percentage of time that the function of the sensing feature exceeds a threshold value; or a percentage of time that the function of the sensing feature exceeds the threshold value over a sliding time window; The periodicity statistics may be Periodicity feature ratio, breathing ratio, heart rate ratio, the percentage of time that the periodicity is detected; the percentage of time in a sliding time window that the periodicity is detected; the percentage of time that the periodicity feature is not the default value or the percentage of time that the periodicity feature is not the default value over a sliding time window.

20. 20. The method of claim 19 further comprising: calculating a sleep analysis based on at least one of any sleep state, any sleep sub-state, the first TSSF, the second TSSF, the plurality of TSSFs, or any associated timing; The method, wherein the sleep analysis includes at least one of a sleep quality score, a sleep quantity score, timing, duration, or a linear combination of periods of sleep with a positive coefficient and periods of awake with a negative coefficient.

Images