Method and apparatus for respiratory rate estimation

Abstract

A method for estimating a respiratory rate by an electronic device is provided. The method includes: receiving a channel state information (CSI) frame of P channel state information (CSI) frames transmitted from another electronic device; estimating the CSI as an available CSI estimate according to each of the received CSI frames; identifying a model for taking into account the inclusion of one or more errors in the amplitude or phase of an estimated CSI having an index p and corresponding to the pth CSI frame of the P CSI frames; compensating the one or more errors in the amplitude or phase of the P CSI estimates by the identified model; and estimating a respiration rate of the subject based on a plurality of spatial dimensions of the P compensated CSI estimates.

Description

Technical Field

[0001] The present disclosure relates generally to respiration rate estimation in wireless communication systems. More specifically, the present disclosure relates to respiration rate estimation based on multi-antenna WiFi. Background Art

[0002] A person's breathing rate is a vital sign. A person's breathing rate can be a parameter for implementing many technical use cases. In addition to the use cases of many health applications such as sleep quality detection, sleep apnea, asthma diagnosis, etc., the parameters corresponding to a person's breathing rate can also be used to detect human presence and monitoring in home security. The increase in demand for such health monitoring has also led to an expansion market for wearable devices ("wearable devices"). Although such wearable devices can accurately measure breathing rate, they are generally expensive and can be invasive and uncomfortable. In addition, wearable devices cannot be used for some use cases such as detection of human presence or home security monitoring. Correspondingly, interest in wireless breathing rate monitoring has increased recently. In such a solution, there is a transmitter (TX) that periodically sends a known wireless signal to a channel including an object performing breathing. The receiver (RX) estimates the channel response from the received signal and then observes its time variation to predict the breathing rate of the object, a task known as wireless sensing. In some methods, a single-base radar is used as both TX and RX to estimate the channel. However, the use of such a dedicated radar device requires additional deployment of a radar system and may bring additional cost burdens, especially in home scenarios. Summary of the invention

[0003] Technical Solution

[0004] The present disclosure provides respiration rate estimation based on multi-antenna WiFi.

[0005] In one embodiment, a method for estimating a respiratory rate by an electronic device is provided. The method may include receiving a CSI frame in P channel state information (CSI) frames sent from another electronic device. The method may include estimating CSI as an available CSI estimate based on each CSI frame in the received CSI frames. The method may include identifying a model for considering one or more errors in the amplitude or phase of the estimated CSI, the estimated CSI being indexed as p and corresponding to the pth CSI frame in the P CSI frames. The method may include compensating for one or more errors in the amplitude or phase of the P CSI estimates by the identified model. Compensating for the one or more errors in the amplitude or phase of the P CSI estimates may include at least one of the following operations: approximating a lost CSI estimate corresponding to a lost CSI frame in the P CSI frames based on the available CSI estimate; determining at least one automatic gain control (AGC) gain value; or calculating one or more errors in the phase based on a weighted least squares solution. The method may include estimating the respiratory rate of the object based on multiple spatial dimensions of the P compensated CSI estimates.

[0006] In one embodiment, an electronic device for estimating a breathing rate is provided. The electronic device may include: a transceiver including N receiving (Rx) antennas; and a processor operably connected to the transceiver. The processor may be configured to receive a CSI frame in P channel state information (CSI) frames sent from another electronic device. The processor may be configured to estimate CSI as an available CSI estimate based on each CSI frame in the received CSI frame. The processor may be configured to identify a model for considering one or more errors in the amplitude or phase of the estimated CSI, the index of the estimated CSI being p and corresponding to the pth CSI frame in the P CSI frames. The processor may be configured to compensate for one or more errors in the amplitude or phase of the P CSI estimates by the identified model. In order to compensate for the one or more errors in the amplitude or phase of the P CSI estimates, the processor may be configured to perform at least one of the following operations: approximate the lost CSI estimate corresponding to the lost CSI frame in the P CSI frames based on the available CSI estimate; determine at least one automatic gain control (AGC) gain value; or calculate one or more errors in the phase based on a weighted least squares solution. The processor may be configured to estimate a respiration rate of the subject based on a plurality of spatial dimensions of the P compensated CSI estimates.

[0007] In one embodiment, a non-transitory computer-readable storage medium storing instructions is provided, and when the instructions are executed by a processor of an electronic device, the electronic device performs operations. The operations may include: receiving a CSI frame in P channel state information (CSI) frames sent from another electronic device. The operations may include: estimating CSI as an available CSI estimate based on each CSI frame in the received CSI frames. The operations may include: identifying a model for considering one or more errors in the amplitude or phase of the estimated CSI, the estimated CSI having an index of p and corresponding to the pth CSI frame in the p CSI frames. The operations may include: compensating for one or more errors in the amplitude or phase of the P CSI estimates by the identified model. Compensating for the one or more errors in the amplitude or phase of the P CSI estimates may include at least one of the following operations: approximating a lost CSI estimate corresponding to a lost CSI frame in the P CSI frames based on an available CSI estimate; determining at least one automatic gain control (AGC) gain value; or calculating one or more errors in the phase based on a weighted least squares solution. The operations may include: estimating the respiration rate of the object based on multiple spatial dimensions of the P compensated CSI estimates.

[0008] Other technical features may be apparent to those skilled in the art from the following drawings, descriptions and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] For a more complete understanding of the present disclosure and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, wherein like reference numerals represent like parts:

[0010] Figure 1 An example wireless network according to the present disclosure is shown;

[0011] Figure 2A An example access point (AP) according to the present disclosure is shown;

[0012] Figure 2B An example station (STA) according to the present disclosure is shown;

[0013] Figure 3 An example wireless communication system that performs respiration rate estimation to determine a respiration rate vital sign of a subject in an environment in accordance with an embodiment of the present disclosure is shown;

[0014] Figure 4 An example of channel state information (CSI) frame transmission according to an embodiment of the present disclosure is shown;

[0015] Figure 5An example method 500 for multi-antenna WiFi CSI-based respiration rate estimation according to an embodiment of the present disclosure is shown;

[0016] Figure 6 shows an example CSI amplitude with discrete jumps over time due to automatic gain control (AGC) gain according to an embodiment of the present disclosure;

[0017] Figure 7 shows an example channel impulse response (CIR) magnitude with random jitter over time due to symbol start error according to an embodiment of the present disclosure;

[0018] Figure 8 shows an example clustered histogram as part of AGC gain compensation using a clustering algorithm of CSI power based on density-based spatial clustering of band noise applications ("DB-scan") for a particular antenna pair for a particular CSI frame in accordance with an embodiment of the present disclosure;

[0019] Fig. 9 shows an example CSI phase before compensating for timing and phase errors according to an embodiment of the present disclosure;

[0020] Fig.10 shows the compensation Fig. 9 Example CSI phase after timing and phase error of ;

[0021] Fig.11A and Fig. 11B exemplifying CSI amplitudes before and after compensating for impairments according to an embodiment of the present disclosure, respectively;

[0022] Fig. 12A and Fig. 12B illustrating example CSI phases before and after compensating for impairments, respectively, according to an embodiment of the present disclosure;

[0023] Fig.13A and Fig. 13B Shown as Fig.11A and Fig. 11B An example CIR magnitude of the Fourier transform of the CSI magnitude;

[0024] Fig.14 shows compensated CSI according to an embodiment of the present disclosure, where there is no residual error and no corrupted samples are identified;

[0025] Fig.15 The embodiment according to the present disclosure is shown Fig.14 Inverse Fast Fourier Transform of the compensated CSI;

[0026] Fig.16The embodiment according to the present disclosure is shown as Fig.15 An example Doppler spectrum of a Fourier transform in the time domain of the compensated CSI, which contains a peak corresponding to the breathing frequency of the subject;

[0027] Fig.17A and Fig. 17B The real and imaginary parts of compensated CSI before and after rotation of the compensated CSI according to an embodiment of the present disclosure are respectively shown to demonstrate the impact of residual impairments; and

[0028] Fig.18 A process for estimating respiration rate based on multi-antenna WiFi signals according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0029] Before proceeding to the following detailed description, it may be advantageous to set forth definitions of certain words and phrases used throughout this patent document. The terms "send," "receive," and "communicate," and their derivatives, encompass both direct and indirect communications. The terms "include," "comprise," and their derivatives, mean including, but not limited to. The term "or" is inclusive, meaning and / or. The phrase "associated with," and its derivatives, means including, included within, interconnected with, including, contained within, connected to or connected with, coupled to or coupled with, communicable with, collaborate with, interlaced, juxtaposed, close to, bound to or bound with, having, having the nature of, having a relationship to or with, etc.

[0030] In addition, the various functions described below can be implemented or supported by one or more computer programs, each of which is formed by a computer-readable program code and embodied in a computer-readable medium. The terms "application" and "program" refer to one or more computer programs, software components, instruction sets, processes, functions, objects, classes, instances, related data or parts thereof suitable for implementation in a suitable computer-readable program code. The phrase "computer-readable program code" includes any type of computer code, including source code, object code and executable code. The phrase "computer-readable medium" includes any type of medium that can be accessed by a computer, such as a read-only memory (ROM), a random access memory (RAM), a hard drive, a compact disk (CD), a digital video disc (DVD) or any other type of memory. "Non-transitory" computer-readable media excludes wired, wireless, optical or other communication links that transmit temporary electrical signals or other signals. Non-transitory computer-readable media include media that can permanently store data and media that can store data and rewrite it later, such as rewritable optical disks or erasable memory devices.

[0031] As used herein, terms and phrases such as "having", "may have", "include" or "may include" a feature (such as a number, function, operation or component such as a part) indicate the presence of the feature and do not exclude the presence of other features. In addition, as used herein, the phrases "A or B", "at least one of A and / or B" or "one or more of A and / or B" may include all possible combinations of A and B. For example, "A or B", "at least one of A and B" and "at least one of A or B" may indicate all of the following: (1) including at least one A, (2) including at least one B, or (3) including at least one A and at least one B. In addition, as used herein, the terms "first" and "second" may modify various components regardless of importance and do not limit the components. These terms are only used to distinguish one component from another. For example, a first user device and a second user device may indicate user devices that are different from each other, regardless of the order or importance of the devices. Without departing from the scope of the present disclosure, a first component may be represented as a second component, and vice versa.

[0032] It will be understood that when an element (such as a first element) is referred to as being (operably or communicatively) "coupled" / "coupled to" another element (such as a second element) or "connected" / "connected to" another element (such as a second element), it may be coupled or connected / coupled or connected to the other element directly or via a third element. Conversely, it will be understood that when an element (such as a first element) is referred to as being "directly coupled" / "directly coupled to" another element (such as a second element) or "directly connected" / "directly connected to" another element (such as a second element), no other elements (such as a third element) are interposed between the element and the other element.

[0033] As used herein, the phrase "configured (or set) to" may be used interchangeably with the phrases "suitable for", "capable of", "designed to", "suitable for", "manufactured to", or "capable of", depending on the circumstances. The phrase "configured (or set) to" does not substantially mean "specially designed in hardware to". On the contrary, the phrase "configured to" may mean that a device may perform an operation together with another device or component. For example, the phrase "a processor configured (or set) to perform A, B, and C" may mean a general-purpose processor (such as a CPU or an application processor) that can perform operations by executing one or more software programs stored in a memory device, or a dedicated processor (such as an embedded processor) for performing operations.

[0034] The terms and phrases used herein are provided only for describing some embodiments of the present disclosure, rather than limiting the scope of other embodiments of the present disclosure. It will be understood that, unless the context clearly indicates otherwise, the singular forms "one", "an" and "the" include plural references. All terms and phrases used herein, including technical and scientific terms and phrases, have the same meaning as the meanings commonly understood by those of ordinary skill in the art to which the embodiments of the present disclosure belong. It will be further understood that terms and phrases (such as those defined in common dictionaries) should be interpreted as having meanings consistent with their meanings in the context of the relevant art, and will not be interpreted in an idealized or overly formalized sense, unless explicitly defined as such herein. In some cases, the terms and phrases defined herein may be interpreted as excluding embodiments of the present disclosure.

[0035] Definitions for certain other words and phrases may be provided throughout this patent document. Those of ordinary skill in the art should understand that in many, if not most instances, such definitions apply to prior, as well as future uses of such defined words and phrases.

[0036] Discussed below Figures 1 to 18 The various embodiments used to describe the principles of the present disclosure in this patent document are for illustration only and should not be interpreted in any way as limiting the scope of the present disclosure. Those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any appropriately arranged wireless communication system.

[0037] As described above, a radar system for wirelessly monitoring a person's breathing rate creates a cost burden for households that need to acquire radar equipment. In an alternative approach, an existing wireless infrastructure (such as commercial WiFi) is used to perform breathing rate detection, where there may be no time / frequency synchronization between TX and RX. Typically, households are already equipped with WiFi equipment (such as access points (APs)), and households equipped with WiFi do not need to acquire additional equipment (e.g., hardware for a radar system).

[0038] Due to the low cost and large-scale deployment of commercial WiFi, such commercial WiFi systems are attractive solutions for estimating the breathing rate of an object. By way of example only, in the examples described in the present disclosure, the object is a person, however, it is understood that in other examples the object may be another breathing body. WiFi systems are not error-free. Due to the lack of synchronization between the WiFi transmitter and the WiFi receiver, and due to the influence of automatic gain control (AGC), etc., the channel state information (CSI) obtained from such a WiFi system suffers from errors in amplitude and phase. In addition, the WiFi receiver may miss some packets, and as a result, the CSI may not be available to the WiFi receiver. These errors may reduce the accuracy of the estimation of the breathing rate of the object detected and / or measured at the WiFi receiver. In order to overcome these problems, embodiments of the present disclosure provide a CSI preprocessing method to remove such errors at the WiFi receiver. In addition, some modern WiFi devices are equipped with multiple antenna elements, and as a result, CSI from multiple antennas may be available at the WiFi receiver. Embodiments of the present disclosure provide a method for combining CSI from multiple antennas at RX to accurately estimate the breathing rate of an object.

[0039] As described above, in order to determine the respiratory rate vital sign of an object, an embodiment of the present disclosure includes a method for performing respiratory rate estimation using CSI obtained from a multi-antenna WiFi infrastructure. An embodiment of the present disclosure develops a detailed system model (e.g., a mathematical system model) that takes into account errors in CSI amplitude and errors in CSI phase. An embodiment of the present disclosure provides a method for compensating for these amplitude and phase errors to "clean" CSI (e.g., reduce / remove errors from CSI, thereby reconstructing CSI). An embodiment of the present disclosure provides several different methods for estimating the respiratory rate of an object from clean CSI (e.g., clean multi-antenna CSI). In particular, these methods utilize different spatial dimensions of clean CSI to estimate the respiratory rate of an object. Then, the algorithm provided according to an embodiment of the present disclosure is evaluated by comparing with data obtained from a real WiFi test environment (e.g., a WiFi test bed) in several different scenarios. Test results show that both the estimated root mean square error and the probability of false detection are reduced by 50% compared to existing conventional solutions.

[0040] The present disclosure provides a method for compensating (e.g., "cleaning") a multi-antenna WiFi CSI and then using the compensated multi-antenna WiFi CSI to perform respiration rate estimation of an object (e.g., a person breathing near a WiFi device). The respiration rate estimation can have a variety of applications, such as monitoring a subject's biosignals for medical purposes, detecting occupancy of a room, detecting the number of objects in a room, for generating an intruder alarm, etc.

[0041] Figure 1 An example wireless network 100 is shown in accordance with various embodiments of the present disclosure. Figure 1 The embodiment of the wireless network 100 shown in FIG. 1 is for illustration only. Other embodiments of the wireless network 100 may be used without departing from the scope of the present disclosure.

[0042] The wireless network 100 includes access points (APs) 101 and 103. The APs 101 and 103 communicate with at least one network 130, such as the Internet, a proprietary Internet Protocol (IP) network, or other data network. The AP 101 provides wireless access to the network 130 for a plurality of stations (STAs) 111-114 within a coverage area 120 of the AP 101. The APs 101-103 may communicate with each other and with the STAs 111-114 using WI-FI or other WLAN communication technologies.

[0043] Depending on the network type, other well-known terms may be used instead of "access point" or "AP", such as "router" or "gateway". For convenience, the term "AP" is used in the present disclosure to refer to a network infrastructure component that provides wireless access to a remote terminal. In a WLAN, assuming that the AP also competes for a wireless channel, the AP may also be referred to as a STA. In addition, depending on the network type, other well-known terms may be used instead of "station" or "STA", such as "mobile station", "subscriber station", "remote terminal", "user equipment", "wireless terminal" or "user device". For convenience, the terms "station" and "STA" are used in the present disclosure to refer to a remote wireless device that wirelessly accesses an AP or competes for a wireless channel in a WLAN, whether the STA is a mobile device (e.g., a mobile phone or smart phone) or is generally considered to be a fixed device (e.g., a desktop computer, AP, media player, fixed sensor, television, etc.).

[0044] Dashed lines illustrate the approximate extents of coverage areas 120 and 125, which are shown as approximately circular for purposes of illustration and explanation only. It should be clearly understood that coverage areas associated with APs, such as coverage areas 120 and 125, may have other shapes, including irregular shapes, depending on the configuration of the AP and changes in the radio environment associated with natural and man-made obstacles.

[0045] As described in more detail below, one or more of the APs may include circuitry and / or programming for estimating respiration rate based on multi-antenna WiFi signals in the WLAN. Figure 1 One example of a wireless network 100 is shown, but may be Figure 1Various changes may be made. For example, wireless network 100 may include any number of APs and any number of STAs in any suitable arrangement. In addition, AP 101 may communicate directly with any number of STAs and provide those STAs with wireless broadband access to network 130. Similarly, each AP 101-103 may communicate directly with network 130 and provide STAs with direct wireless broadband access to network 130. In addition, AP 101 and / or 103 may provide access to other or additional external networks (such as an external telephone network or other type of data network).

[0046] Figure 2A An exemplary AP 101 is shown in accordance with various embodiments of the present disclosure. Figure 2A The embodiment of AP 101 shown in FIG. 1 is for illustration only, and Figure 1 The APs 103 may have the same or similar configurations. However, APs have a variety of configurations, and Figure 2A The scope of the present disclosure is not limited to any particular implementation of an AP.

[0047] The AP 101 includes multiple antennas 204a-204n, multiple RF transceivers 209a-209n, transmit (TX) processing circuitry 214, and receive (RX) processing circuitry 219. The AP 101 also includes a controller / processor 224, a memory 229, and a backhaul or network interface 234. The RF transceivers 209a-209n receive incoming RF signals from the antennas 204a-204n, such as signals transmitted by STAs in the network 100. The RF transceivers 209a-209n down-convert the incoming RF signals to generate IF or baseband signals. The IF or baseband signals are sent to the RX processing circuitry 219, which generates processed baseband signals by filtering, decoding, and / or digitizing the baseband or IF signals. The RX processing circuitry 219 sends the processed baseband signals to the controller / processor 224 for further processing.

[0048] The TX processing circuit 214 receives analog or digital data (such as voice data, web data, email, or interactive video game data) from the controller / processor 224. The TX processing circuit 214 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. The RF transceivers 209a-209n receive the outgoing processed baseband or IF signals from the TX processing circuit 214 and up-convert the baseband or IF signals to RF signals that are transmitted via the antennas 204a-204n.

[0049] The controller / processor 224 may include one or more processors or other processing devices that control the overall operation of the AP 101. For example, the controller / processor 224 may control the reception of the forward channel signal and the transmission of the reverse channel signal by the RF transceiver 209a-209n, the RX processing circuit 219, and the TX processing circuit 214 according to well-known principles. The controller / processor 224 may also support additional functions, such as more advanced wireless communication functions. For example, the controller / processor 224 may support beamforming or directional routing operations, in which the outgoing signals from the multiple antennas 204a-204n are weighted differently to effectively direct the outgoing signals to the desired direction. The controller / processor 224 may also support OFDMA operations, in which the outgoing signals are assigned to different subcarrier subsets for different recipients (e.g., different STAs 111-114). The controller / processor 224 may support any of a variety of other functions in the AP 101, including estimating the breathing rate based on the multi-antenna WiFi signal. In some embodiments, the controller / processor 224 includes at least one microprocessor or microcontroller. The controller / processor 224 is also capable of executing programs and other processes, such as an OS, that reside in the memory 229. The controller / processor 224 may move data into or out of the memory 229 as required by the executing process.

[0050] The controller / processor 224 is also coupled to a backhaul or network interface 234. The backhaul or network interface 234 allows the AP 101 to communicate with other devices or systems over a backhaul connection or over a network. The interface 234 can support communication over any suitable (one or more) wired or wireless connection. For example, the interface 234 can allow the AP 101 to communicate over a wired or wireless local area network or over a wired or wireless connection to a larger network (such as the Internet). The interface 234 includes any suitable structure that supports communication over a wired or wireless connection, such as an Ethernet or RF transceiver. The memory 229 is coupled to the controller / processor 224. A portion of the memory 229 may include RAM, and another portion of the memory 229 may include flash memory or other ROM.

[0051] As described in more detail below, AP 101 may include circuitry and / or programming for estimating respiration rate based on multi-antenna WiFi signals. Figure 2A An example of AP 101 is shown, but the Figure 2A For example, AP101 may include any number of Figure 2AAs a specific example, the access point may include multiple interfaces 234, and the controller / processor 224 may support routing functions to route data between different network addresses. As another specific example, although shown as including a single instance of TX processing circuit 214 and a single instance of RX processing circuit 219, AP 101 may include multiple instances of each (such as one for each RF transceiver). Alternatively, only one antenna and RF transceiver path may be included, such as in a traditional AP. In addition, Figure 2A The various components in may be combined, further subdivided, or omitted, and additional components may be added according to specific needs.

[0052] Figure 2B An example STA 111 is shown according to various embodiments of the present disclosure. Figure 2B The embodiment of STA 111 shown in FIG. is for illustration only, and Figure 1 STAs 111-115 may have the same or similar configurations. However, STAs have a variety of configurations, and Figure 2B The scope of this disclosure is not limited to any particular implementation of STA.

[0053] STA 111 includes antenna(s) 205, radio frequency (RF) transceiver 210, TX processing circuit 215, microphone 220, and receive (RX) processing circuit 225. STA 111 also includes speaker 230, controller / processor 240, input / output (I / O) interface (IF) 245, touch screen 250, display 255, and memory 260. Memory 260 includes operating system (OS) 261 and one or more applications 262.

[0054] The RF transceiver 210 receives incoming RF signals sent by the APs of the network 100 from the antenna(s) 205. The RF transceiver 210 down-converts the incoming RF signals to generate an intermediate frequency (IF) or baseband signal. The IF or baseband signal is sent to the RX processing circuit 225, which generates a processed baseband signal by filtering, decoding, and / or digitizing the baseband or IF signal. The RX processing circuit 225 sends the processed baseband signal to the speaker 230 (such as for voice data) or to the controller / processor 240 for further processing (such as for web browsing data).

[0055] The TX processing circuit 215 receives analog or digital voice data from the microphone 220, or other outgoing baseband data (such as web data, email, or interactive video game data) from the controller / processor 240. The TX processing circuit 215 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. The RF transceiver 210 receives the outgoing processed baseband or IF signal from the TX processing circuit 215 and up-converts the baseband or IF signal to an RF signal that is transmitted via the antenna(s) 205.

[0056] The controller / processor 240 may include one or more processors and execute a basic OS program 261 stored in the memory 260 to control the overall operation of the STA 111. In one such operation, the main controller / processor 240 controls the reception of forward channel signals and the transmission of reverse channel signals by the RF transceiver 210, the RX processing circuit 225, and the TX processing circuit 215 according to well-known principles. The main controller / processor 240 may also include a processing circuit configured to estimate the respiration rate based on the multi-antenna WiFi signal. In some embodiments, the controller / processor 240 includes at least one microprocessor or microcontroller.

[0057] The controller / processor 240 is also capable of executing other processes and programs resident in the memory 260, such as operations for estimating breathing rate based on multi-antenna WiFi signals. The controller / processor 240 can move data into or out of the memory 260 as needed to execute the process. In some embodiments, the controller / processor 240 is configured to execute multiple applications 262, such as an application for jointly estimating breathing rate based on multi-antenna WiFi signals. The controller / processor 240 can operate multiple applications 262 based on the OS program 261 or in response to a signal received from an AP. The main controller / processor 240 is also coupled to an I / O interface 245, which provides the STA 111 with the ability to connect to other devices such as laptops and handheld computers. The I / O interface 245 is a communication path between these accessories and the main controller 240.

[0058] The controller / processor 240 is also coupled to a touch screen 250 and a display 255. An operator of the STA 111 can use the touch screen 250 to enter data into the STA 111. The display 255 can be a liquid crystal display, a light emitting diode display, or other display capable of rendering text and / or at least limited graphics (such as from a website). A memory 260 is coupled to the controller / processor 240. A portion of the memory 260 may include a random access memory (RAM), and another portion of the memory 260 may include flash memory or other read-only memory (ROM). The memory 260 may store instructions that, when executed by the controller / processor 240, cause the STA to perform the operations described herein. The memory 260 may be referred to as a non-transitory computer-readable storage medium.

[0059] although Figure 2B An example of STA 111 is shown, but the Figure 2B Make various changes. For example, Figure 2B The various components in the embodiment may be combined, further subdivided, or omitted, and additional components may be added as required. In a specific example, the STA 111 may include any number of (one or more) antennas 205 for MIMO communication with the AP 101. In another example, the STA 111 may not include voice communications, or the controller / processor 240 may be divided into multiple processors, such as one or more central processing units (CPUs) and one or more graphics processing units (GPUs). In addition, although Figure 2B STA 111 is shown configured as a mobile phone or smart phone, but the STA may be configured to operate as other types of mobile or fixed devices.

[0060] Figure 3 An example wireless communication system 300 is shown that performs respiratory rate estimation to determine a respiratory rate vital sign of an object in an environment according to an embodiment of the present disclosure. In particular, the system 300 is composed of an AP 301 and a STA 311 located in an environment 370, and the system 300 performs respiratory rate estimation to determine a respiratory rate vital sign of an object 380 located in the environment 370. Figure 3 The embodiment of the wireless communication system 300 shown in FIG. is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure. For example, Figure 3 The embodiment of the wireless communication system 300 shown in FIG. 3 may have Figure 1 The network 100 has the same or similar configuration. Figure 3 The AP 301 and STA 311 may each have Figure 2A AP 201 and Figure 2BThe AP 301 has the same or similar configuration as the STA 111. 1 -304 M Can be expressed Figure 2A The STA 311 includes N receiving (Rx) antennas, including the first antenna 305 1 To the Nth antenna 305 N , which can be expressed as Figure 2B The system 300 includes MN antenna pairs consisting of N Rx antennas paired with each of the M Tx antennas, and each of the MN antenna pairs has an index of m.

[0061] For simplicity, the commercial Wi-Fi infrastructure may refer to AP 301, STA 311, or a system 300 including both AP 301 and STA 311. The commercial Wi-Fi infrastructure may be a modern WiFi device equipped with multiple antenna elements. Three main features of the commercial Wi-Fi infrastructure according to an embodiment of the present disclosure will be described, including the following features: (1) obtaining CSI from a multi-antenna commercial Wi-Fi infrastructure to estimate the respiration rate of the object 380 based on a system model configured to consider one or more errors in CSI amplitude and one or more errors in CSI phase; (2) compensating for one or more errors in CSI amplitude and one or more errors in CSI phase to reconstruct CSI; and (3) estimating the respiration rate of the object 380 from the reconstructed multi-antenna CSI based on different spatial dimensions of the reconstructed CSI.

[0062] As an example, its breathing rate f will be estimated br (in Hz) of object 380 near STA 311. Object 380 is relatively stationary or moving slowly, for example, the object may be lying on floor 371 facing ceiling 372.

[0063] In the illustrated scenario, AP 301 transmits signals 310_1 to 310_M, such as CSI frame p, from each of the M Tx antennas to STA 311 (as a receiver). Some of the transmitted signals are reflected from stationary surfaces of objects in environment 370, such as ceilings, walls, floors, or furniture. For example, the Tx antenna 304 1 The transmitted signal 310_1 is reflected from the ceiling 372; 2The transmitted signal 310_2 is reflected from the wall 373. As a result, the STA 311 receives the reflected signals 320_1 and 320_2, respectively. Some of the transmitted signals are reflected from the moving surface of the object 380 in the environment 370, such as the moving torso (i.e., chest) of a breathing person. For example, the signals 310_4 and 10_M transmitted by the AP 311 are incident on the moving surface of the torso of the object 380 at an angle θ relative to the surface plane 382 of the torso. As a result, the STA 311 receives the reflected signals 320_4 and 320_M, respectively. When the object lies flat on the floor 371, the surface plane 382 of the torso of the object 380 may be parallel to the surface of the floor 371. Some of the transmitted signals 310_3 are received by the STA 311 without being reflected at any surface.

[0064] although Figure 3 An example environment 370 is shown in which the wireless communication system 300 estimates the breathing rate of a subject 380, but may be used for Figure 3 Various changes may be made. In particular, the presence of only one object 380 is for simplicity and ease of explanation and is only an example, not a limitation of the embodiments presented in the present disclosure. Multiple objects 380 may be present in the environment, and the system 300 may estimate their breathing rates separately. As another specific example, in one embodiment, the transmitter may be an AP and the receiver may be a WiFi STA, or in another embodiment, vice versa. For simplicity, the examples used in the present disclosure refer to embodiments in which AP 301 is also referred to as transmitter 301 and STA 311 is also referred to as receiver 311.

[0065] Figure 4 An example channel state information (CSI) frame transmission 400 is shown in accordance with an embodiment of the present disclosure. Figure 4 The embodiment of CSI frame transmission shown in is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0066] In order to be able to estimate the breathing rate f of the subject 380 br , the transmitter may send a sequence of WiFi channel state information (CSI) acquisition frames 402, each acquisition frame having a T rep The sequence of WiFi CSI acquisition frames includes P CSI frames 402, which have a frame index p ranging from 0 to P-1. For ease of explanation, CSI frame p is the pth CSI frame.

[0067] In one embodiment, each CSI frame 402 may include a short training field (STF) 404 and a plurality of pilot symbols 406, one pilot symbol being sent from each of the M TX antennas of the AP 301 to help the STA 311 obtain CSI from each of the M TX antennas. 1 The first pilot symbol 4061 is transmitted, which is shown as long training field (LTF) 1. Similarly, the Mth TX antenna 304 M Send the Mth pilot symbol 406 M (shown as LTF M). In one embodiment, these CSI frames 402 may be beacon frames or null data packets, and transmission may be performed at indexes Orthogonal frequency division multiplexing (OFDM) is used on the K subcarriers.

[0068] Using the received signal of each CSI frame p, STA311 obtains the corresponding MN×1 noise CSI on subcarrier k (from each TX antenna to each RX antenna) as the vectored channel To avoid loss of generality, this disclosure describes vectorization of CSI, where the zth component of the MN×1 vector (ie, ) indicates TX antenna: and CSI between RX antenna: mod{z,N}+1.

[0069] Figure 5 An example method 500 for multi-antenna WiFi CSI based respiration rate estimation according to an embodiment of the present disclosure is shown. Figure 5 The embodiment of the method 500 for multi-antenna WiFi CSI-based respiration rate estimation shown in FIG. 5 is for illustration only, and other embodiments (eg, Fig.18 Method 1800).

[0070] At block 510, a trigger from a timer, an external source, or another CSI-based algorithm is used to trigger the respiration rate estimation. To obtain an estimate of the respiration rate of subject 380, receiver 311 may use the CSI estimate obtained from the previous p CSI frames. However, firstly, due to the uncertainty of frame reception, the CSI estimate obtained for a certain frame p is At block 520, interpolation for the lost CSI frame is performed. Secondly, due to synchronization errors and hardware impairments, the obtained CSI may be noisy and / or erroneous. Before estimating the breathing signal (at block 580), these errors are compensated from the obtained CSI (at blocks 520-570), specifically by performing a CSI preprocessing method according to an embodiment of the present disclosure to remove these errors at the receiver.

[0071] Now return to reference box 520. Sometimes, when AP 301 transmits P CSI frames, STA 311 fails to receive one or more of the P CSI frames transmitted, which are referred to as one or more lost CSI frames. From the P CSI frames transmitted, STA 311 successfully receives some of the CSI frames, which are referred to as received CSI frames. As part of performing channel estimation, STA 311 estimates CSI based on each of the received CSI frames, and such CSI estimates are referred to as available CSI estimates. That is, the received CSI frames are available for processing by a channel estimation algorithm that produces available CSI estimates.

[0072] Embodiments of the present disclosure identify (e.g., generate) an impairment model to account for (e.g., identify; measure; quantize) errors contained in the amplitude or phase of the estimated CSI indexed p and corresponding to the pth CSI frame among P CSI frames. Lost CSI frames result in some errors that are accounted for in the impairment model.

[0073] For any lost CSI frame with index p, the previous CSI frame and the subsequent CSI frame for which CSI is available and closest to the lost CSI frame are determined as p=q, respectively. 1 and p = q 2 Previously available CSI estimates Compared with the previous CSI frame (p = q 1 ) corresponds to. The CSI estimation that can be used later and the subsequent CSI frame (p = q 2 ) corresponds to. Based on these previous and subsequent available CSI estimates and STA 311 approximates the lost CSI estimate corresponding to the lost CSI frame among the P CSI frames. In one embodiment, the lost CSI estimate may be approximated to be equivalent to the previously available CSI estimate corresponding to the previous CSI frame closest to the lost CSI frame. As shown in Equation 1. In another embodiment, the lost CSI estimate can be approximated as being equivalent to the subsequent available CSI estimate corresponding to the subsequent CSI frame closest to the lost CSI frame As shown in Equation 2. More generally, the lost CSI frame can be approximated as a function of multiple available CSI estimates including previous and subsequent available CSI frames, as shown in Equation 3. For example, the function can be an interpolation of at least one of the amplitude or phase. For example, the function can be based on two available CSI estimates and Linear interpolation of .

[0074]

[0075] As an example scenario, if frames 49 and 52 are successfully received, but the lost CSI frames are frames 50 and 51, STA 311 has available CSI estimates corresponding to each of frames 49 and 52, but STA 311 does not have available data from frames 50 and 51 to perform channel estimation. To approximate the lost CSI estimates corresponding to the lost frames 49 and 50, STA 311 may calculate As the first function and calculate as a second function that is the same as or different from the first function.

[0076] At box 530, compensation for AGC gain error, timing offset error, and phase offset error of the CSI is performed to obtain clean CSI. At box 540, outliers in the clean CSI are identified, removed, and interpolated. At box 550, the clean CSI can be used to further determine whether breathing rate estimation is needed. If the determination at box 550 is no, the method 500 ends. If the determination at box 550 is yes, the method 500 proceeds to box 560. At box 560, different spatial components of the clean CSI are identified, and their Doppler delays or Doppler frequency responses are obtained. At box 570, based on the clean CSI, it is determined which of the multiple proposed algorithms is used for breathing rate estimation. At box 580, the determined algorithm for breathing rate estimation is run to generate a breathing estimate. Finally, at block 590, additional steps may be performed on the respiration estimation, such as feeding into a tracking algorithm, using as prior information for the next iteration of the estimation, determining whether the respiration estimation process needs to be terminated, and the like.

[0077] Figure 6 and Figure 7 Graphs 600 and 700 are shown plotting the unprocessed CSI received at STA 311 as a function of time, where the frame index p represents time. Due to synchronization errors and hardware impairments of STA 311, the obtained noisy CSI can be modeled as the impairment model shown in Equation 4.

[0078]

[0079] In this injury model, represents the uncompensated (e.g., including the effects of impairments, original, unprocessed) CSI estimated from the pth CSI frame. This uncompensated CSI It can be represented mathematically in the form of a channel matrix. represents the true (i.e., without the effects of impairments) MN×1 CSI on the kth subcarrier of the pth CSI frame between TX and RX. The term HPK can be expressed according to Equation 5, where the term represents the static component of CSI; and the term D p，k Represents the dynamic component of CSI.

[0080]

[0081] Item p is a MN x 1 real vector representing the gain from the automatic gain control circuit at the receiver 311. The operator ⊙ represents the element-wise product (also called the Hadamard product) of two vectors. The term Δτ p represents the timing synchronization offset at the receiver 311 of the p-th CSI frame p. k represents the frequency offset of the kth subcarrier from the carrier frequency. p represents the carrier phase offset for CSI frame p of the receiver 311. In some embodiments, to correct errors, the impairment model is based on the assumption that the true CSI is quasi-static over time, as shown in Equation 6.

[0082] For any

[0083] In particular, Figure 6 A graph 600 is shown of an example CSI magnitude with discrete jumps over time due to AGC gain according to an embodiment of the present disclosure. The CSI is a function of time and frequency, shown as |CSI(t,f)|. The x-axis represents subcarrier indices, e.g., in a range from 0 to 60. The y-axis represents time (e.g., in units of 100 milliseconds), e.g., in a range from 0 to 800. Figure 6 The embodiment of the CSI magnitude over time graph 600 shown in FIG. 6 is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0084] Figure 7 An example graph 700 of a channel impulse response (CIR) magnitude with random jitter over time due to a symbol start error is shown according to an embodiment of the present disclosure. The CIR is the inverse fast Fourier transform (IFFT) of the time-dependent frequency-dependent CSI, shown as |IFFTf ([CSI(t,f)]|. The x-axis represents a delay index (center), for example, in a range from -150 to 150. The y-axis represents time (for example, in units of 100 milliseconds), for example, in a range from 0 to 800. Figure 7 The embodiment of the CIR magnitude over time graph 700 shown in FIG. 7 is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0085] Figure 8 An example clustering histogram 800 is shown as part of an AGC gain compensation using a density-based spatial clustering of noisy applications ("DB-scan") based on Γ according to an embodiment of the present disclosure. p，m The term Γ p，m represents the CSI power of the mth antenna pair in the pth CSI frame. Figure 8 The embodiment of the clustered histogram 800 shown in FIG. 8 is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure. In order to remove or reduce the amplitude error caused by the AGC gain at the receiver, and to generate AGC gain compensated CSI, a CSI may be generated. Figure 8 The cluster histogram 800 shown in FIG. Figure 5 Part of the compensation process performed at box 530 of .

[0086] In many environments, most surfaces are stationary, such as Figure 3 An indoor environment 370 is shown in which a few signals transmitted from AP 301 are reflected from moving surfaces, such as the torso of subject 380.

[0087] Due to the quasi-static nature of the channel and / or CSI, it can be expected that for any antenna pair m, the CSI power for the real channel as expressed according to Equation 7 is constant during the time period 0≤p≤P-1 when P CSI frames 402 are transmitted.

[0088]

[0089] Correspondingly, in one embodiment, for each antenna pair 1≤m≤MN, the CSI power {Γ p，m |0≤p≤P-1} to identify clusters corresponding to different AGC gain values. For a particular antenna pair m, the number of identified clusters is denoted as W, where the index of each cluster is w. Examples of clustering algorithms include DB-can clustering or K-means clustering. For each identified cluster, the square root of each measured CSI power value within the cluster is calculated. The average value (e.g., median or mean) of For each antenna pair m, the CSI with AGC gain compensation can be obtained as expressed in Equation 8:

[0090]

[0091] The cluster histogram 800 shows three clusters (W=3), including a first cluster 802 (Cluster 1) having a first median 804 (Median-1), a second cluster 806 having a second median 808, and a third cluster 810 (Cluster 3) having a third median 812.

[0092] In another embodiment, it is determined whether to compensate for one or more errors in the amplitude of the corresponding CSI frame indexed p in the P CSI frames. To make this determination, a sequential method can be performed, wherein if |Γ p，m -Γ p-1，m |≥threshold, then we obtain (eg, generate) ACG gain compensated CSI, as shown in Equation 9. That is, the determination is based on whether the difference between CSI power values ​​measured for consecutive CSI frames is exceeded by a threshold.

[0093]

[0094] Fig. 9 An example CSI phase 900 is shown before compensating for timing and phase errors according to an embodiment of the present disclosure. Fig.10 shows the compensation of timing and phase errors. Fig. 9 Example CSI phase 1000. Fig. 9 and Fig.10 In the example, the CSI phase is the AGC gain compensation CSI, including the The angle dimension 902, the time dimension 904 labeled "CSI frame index (p)", and the frequency dimension 906 labeled "subcarrier index (k)".

[0095] Due to the quasi-static nature of the channel and / or CSI, it can be expected that the true CSI is approximately equivalent to the static component of the CSI (i.e., ). Correspondingly, in one embodiment, for a given static component of CSI The timing error Δτ for any CSI frame p can be calculated as shown in Equations 10 and 11, respectively. p and phase error ψ p The term T represents The scope of the search for values ​​of . Represents the static component of CSI This method can be used to estimate A priori estimate of

[0096]

[0097]

[0098] In another embodiment, for a given estimate of the static component The timing error Δτ of any CSI frame p can be p and phase error ψ p is calculated as the solution to the weighted least squares problem shown in Eq. 12.

[0099] This embodiment can utilize A priori estimate of

[0100]

[0101] function is the phase unwrapping function, which ensures that the phase difference between adjacent subcarriers is less than π, as shown in Equation 13:

[0102]

[0103] Correspondingly, the CSI after phase and timing error compensation is called compensated ("clean") CSI, which can be obtained as expressed in Equation 14:

[0104]

[0105] In some embodiments, the estimation of the static component of the CSI can be represented as an MN×1 all-ones vector: In another embodiment, the estimation of the static component of CSI can be expressed according to Equation 15:

[0106]

[0107] In yet another embodiment, the estimation of the static component of CSI can be expressed according to Equation 16: That is, the estimate of the static component from the previous estimation round

[0108]

[0109] In yet another embodiment, the following steps may be performed in time sequence from CSI frame p=0 to CSI frame p=P-1. To estimate for any p The value to be used can be obtained as shown in Equation 17 Values:

[0110]

[0111] In some embodiments, compensated (“clean”) CSI is obtained sequentially for all p Afterwards, the estimation of the static component is fixed as expressed in Equation 18, and is performed using This updated value of is re-estimated for all p=1,...,P-1 The second iteration of .

[0112]

[0113] The above process for compensating CSI for timing and phase errors involves a joint timing and phase correction method, in which all MN antenna dimensions are jointly compensated ("cleaned") for CSI timing and phase errors. In another embodiment, the same process for compensating CSI for timing and phase errors can be used, but the timing and phase errors can be estimated and corrected independently for each antenna dimension.

[0114] Outlier detection and removal

[0115] After completing the following three compensations, some CSI samples (e.g., compensated CSI estimates) may still be too corrupted: (1) approximation of the missing CSI estimate; (2) AGC gain error compensation; (3) (one or more) phase error and timing error compensation. Using such CSI samples will degrade the respiration rate performance. The fact that the CSI is quasi-static can be used to identify such CSI samples that are still too corrupted. For example, in one embodiment, after completing the three compensations listed above, the metric expressed in Equation 19 can be used as an identifier of a corrupted CSI sample. A corrupted CSI frame p is identified based on the determination that the comparison in Equation 19 is "true".

[0116]

[0117] in,

[0118] Such a damaged CSI frame p may be replaced with the previous CSI frame as represented in Equation 20. In another embodiment, interpolation of the previous CSI frame and the subsequent CSI frame may also be used to replace the damaged CSI frame, which is a similar process to block 520.

[0119]

[0120] In some embodiments, the amplitude of the CSI estimate based on the gain compensation and expressed as The comparison includes determining the average value of the magnitude of the P gain-compensated CSI estimates to detect the corrupted CSI estimate. and Whether the sum of the squared errors between (across subcarrier k) is greater than an outlier threshold. In such an embodiment, the receiver may be configured to: identify the magnitude of the AGC gain compensated CSI estimate corresponding to each of the P CSI estimates Identification is represented by The average of the magnitudes of the P AGC gain compensated CSI estimates of the P subcarriers; detecting a corrupted CSI estimate based on determining that the comparison is "true", wherein the comparison includes determining for each of the K subcarriers indexed by k and whether the sum of the square errors between is greater than an outlier threshold; and replacing the damaged CSI frame within the P compensated CSI estimates.

[0121] Fig.11A and Fig. 11B shows example CSI amplitudes before and after compensating for impairments, respectively, according to an embodiment of the present disclosure 1100 and 1150. These impairments include at least one of the following: a lost CSI estimate for any CSI frame p, an AGC gain error or other error in CSI amplitude, a timing error Δτ p Or phase error ψ p The uncompensated CSI amplitude 1100 exhibits abrupt changes in amplitude over time caused by different AGC gain levels, similar to Figure 6 As an example of a sudden change in amplitude within a single subcarrier index, CSI amplitude 1100 switches from a first AGC gain level 1102 to a second AGC gain level 1104 at a first time, and then the amplitude switches to a third AGC gain level 1106 at a second time. The first AGC gain level 1102 and the third AGC gain level 1106 may be the same or approximately equal to each other, but both 1102 and 1106 are significantly different from the second AGC gain level 1104. Fig. 11B It is shown that the CSI amplitude 1150 as a function of time and carrier index exhibits a smooth pattern over time.

[0122] Fig. 12A and Fig. 12B FIG. 4 shows example CSI phases before and after compensating for impairments, respectively, according to an embodiment of the present disclosure. 1200 and 1250. Fig. 12A and Fig. 12B The three dimensions of phase angle 1202, time 1204 in units of 100 milliseconds (ms), and frequency 1206 shown in FIG. Fig. 9 and Fig.10 The three corresponding dimensions 902, 904 and 906 are the same or similar. Fig. 12B In addition to several peaks 1208 ( Fig. 11B 1152), the value of CSI phase 1250 is almost all zero or very close to zero. When the amplitude is too small, the peak is generated by a small amount of noise.

[0123] Fig.13A and Fig. 13B Shown as Fig.11A and Fig. 11B CSI amplitude An example of the Fourier transform of the channel impulse response (CIR) magnitude 1300 and 1350. The temporal random jitter shown in the CIR amplitude 1300 before compensating for impairments is similar to Figure 7 The jitter shown in the CIR amplitude diagram 700 is shown in FIG. However, such jitter is removed, as shown by Fig. 13B The stable CIR amplitude of 1350 is shown in the clean pattern.

[0124] Sometimes, after completing the following three compensations, the compensated CSI estimate Under ideal conditions: (1) Approximate the missing CSI estimate; (2) AGC gain error compensation; (3) (one or more) phase error and timing error compensation. Fig.14 FIG. 4 shows an example compensated CSI under ideal conditions where there is no residual error and no corrupted samples are identified according to an embodiment of the present disclosure. 1400. That is, the compensated CSI 1400 has a sinusoidal variation with the breathing signal of the subject 380. The ideal compensated CSI 1400 may be based on the assumption that all hardware and synchronization impairments are removed by the three compensations listed above, and may be expressed according to Equation 21. Using Taylor expansion, the ideal compensated CSI 1400 may be expressed according to Equation 22.

[0125]

[0126] The breathing signal of the subject 380 is distributed over all subcarriers k and time samples p, and the power of the MPC We can be focused and isolated by converting from the (time, frequency) domain to the (Doppler delay) domain. More specifically, the ideal compensated CSI 1400 is based on the assumption that the dynamic component includes the breathing rate of the subject 380, and the sinusoidal breathing signal can be expressed according to Equation 23.

[0127] sin(2πf br T rep +φ)

[0128] Fig.15 The embodiment according to the present disclosure is shown Fig.14 The compensated CSI is obtained by inverse fast Fourier transform (IFFT k )1500. Fig.15 In IFFT k 1500 includes the number marked "|IFFT f [CSI compensated (t, f)]|" a first dimension 1502, a time dimension 1504 in units of 100 ms, and a third dimension 1506 labeled "center delay index".

[0129] Fig.16 The embodiment according to the present disclosure is shown as Fig.15 The Fourier transform (FFT) of the compensated CSI in the time domain p ) includes an example Doppler spectrum 1600 of the subject 380 with respect to the respiratory frequency (ie, breathing rate) f br The corresponding peak 1602. Doppler spectrum 1600 includes the FFT t (IFFT f (CSI compensated (t,f)))|" a first dimension 1604 in Hertz, a Doppler dimension 1606 in Hertz, and a third dimension 1608 labeled "Center Delay Index".

[0130] To show the effect of residual damage, Fig.17A and Fig. 17B The real and imaginary parts of compensated CSI 1700 and 1750 are shown, respectively, before and after rotating the compensated CSI, according to an embodiment of the present disclosure.

[0131] Even after completing the following three compensations, in some CSI samples (e.g., compensated CSI estimates) Small residual impairments will still remain in: (1) approximation of the missing CSI estimate; (2) AGC gain error compensation; (3) (one or more) phase error and timing error compensation. An example of these small residual impairments is represented by Another example of these small residual impairments is the residual phase error for CSI frame p CSI-based static components Assuming that the compensation CSI does not change with the breathing of the object 380, the compensated CSI can be calculated according to Equation 24. Modeling.

[0132]

[0133] In the case of multiple Tx antennas and multiple Rx antennas, the compensated CSI estimation Rotation ensures minimal residual damage and Only exists along some dimensions but not all dimensions. In the example scenario, the CSI is distributed in a four-dimensional space defined by four antenna pairs corresponding to 2 Tx antennas and 2 Rx antennas, but all noise is only along one axis or two axes. That is, the residual error is concentrated along that one or two axes. The embodiments of the present disclosure ignore those dimensions or directions along which there are large residual errors, and focus on (e.g., continue to process) those dimensions where the errors are small, thereby enabling the receiver 311 to calculate a compensated estimate of very clean CSI. Fig.17A and Fig. 17B The advantages of this technology are demonstrated. Fig.17A The subject's breathing signal 1702 and the static component are shown. The residual phase error 1704 and the non-zero real and imaginary values ​​of the residual gain error 1706. However, as a result of rotating the CSI, Fig. 17B The non-zero real and imaginary values, static components of the subject's respiration signal 1702 are shown. The residual phase error The constant value of the real component of 1704, and the static component The imaginary component of the residual gain error 1756 has a zero value.

[0134] Estimation of breathing rate:

[0135] Without loss of generality, the MN×1 “true” CSI (i.e., without the effects of impairments) on the kth subcarrier of the pth CSI frame between TX and RX can be expressed according to Equation 25. The channel has L MPCs. Each MPC consists of a time-varying amplitude α l (p) and MN×1 array response vector a l (p) and delay τ l (p). Item f k represents the frequency offset of the kth subcarrier from the carrier frequency, and the term f c Indicates the carrier frequency.

[0136]

[0137] Furthermore, without loss of generality, it can be assumed that MPC 0 represents the signal path from AP 301 to STA 311 reflected from the torso of object 308. For this MPC 0, the model of the temporal variation of the MPC amplitude due to the motion of the torso of the object can be expressed as: 0 (p) = α 0 , and the array response can be expressed as: α 0 (p) = α 0 The MPC delay can be expressed according to Equation 26, where τ 0 represents the average path delay, denotes an additional delay due to trunk motion, and φ denotes a parameter representing the respiratory phase of the subject 380 .

[0138]

[0139] Implicitly, it is assumed that the MPC amplitude and array response are not affected by the respiration of subject 380, and the delay is assumed to have a frequency of the respiration rate f br The static component based on CSI Assuming no change with breathing, compensated CSI can be modeled as shown in Equation 27, where represents the MN×1 residual AGC gain error after completing AGC gain compensation, and represents the residual phase error of CSI frame p after completing the phase error(s) and timing error compensation.

[0140]

[0141] By multiplying the static component The phase angle estimate of , can be further expressed as the compensated CSI according to Equation 28:

[0142]

[0143] In one embodiment, the Doppler delay CSI can be obtained by using Fourier transform along p dimension and inverse Fourier transform along k dimension, as shown in Equation 29 and as Figure 14-16 The term v represents the Doppler frequency bin index, and u represents the delay bin index.

[0144]

[0145] Using the Doppler delay CSI of Equation 29 (above) One or more different algorithms may be used to estimate the respiratory rate. For convenience, the following definitions are provided in Equations 30 and 31:

[0146]

[0147] Without loss of generality, we can also consider feasible delay bins as candidates for MPC-0 that include respiratory signals and Doppler binning In one embodiment, these feasible bins can be obtained from side information and As an example, the side information may be previously obtained estimates of breathing rate and delay binning.

[0148] In an embodiment of Algorithm 1, an estimate of the breathing rate may be obtained as: in:

[0149]

[0150] In an embodiment of Algorithm 2, an estimate of the breathing rate may be obtained as: in:

[0151]

[0152] In an embodiment of Algorithm 3, an estimate of the breathing rate may be obtained as: in:

[0153]

[0154] More generally, The different spatial dimensions of can be optimally used to obtain the respiratory rate. Note that Divide into 3 components:

[0155]

[0156] According to the analysis, the weight Affected by the residual AGC gain error However, the residual phase error is elastic. Similarly, Affected by the residual phase error However, the AGC gain error Finally, the weight Residual AGC gain error and residual phase error Correspondingly, in one embodiment, when estimating the respiratory rate, different weights can be assigned to the three components:

[0157] In an embodiment of Algorithm 4, an estimate of the breathing rate may be obtained as: in:

[0158]

[0159] Item 1 、w 3 、w 4 are the weights assigned to the three components. Similar to the embodiment of Algorithm 2, this embodiment of Algorithm 4 enables the use of replace

[0160] In an embodiment of Algorithm 5, an estimate of the breathing rate may be obtained as: in:

[0161]

[0162] Item 1 、w 3 、w 4 are the weights assigned to the three components.

[0163] In an embodiment of Algorithm 6, an estimate of the breathing rate may be obtained as: in:

[0164]

[0165] And w 3 、w 4 are the weights assigned to the two components.

[0166] In an embodiment of Algorithm 7, an estimate of the breathing rate may be obtained as: in:

[0167]

[0168] In the embodiment of Algorithm 8, the components of CSI are: can be extracted and fed as input to a neural network to generate f br or v 0 As output, the neural network can then be trained to perform the estimation of the true respiratory rate f br or v 0 To train the network, a large “labeled” dataset can be collected for the CSI of objects, where the ground truth labels f can be obtained, for example, by attaching a breathing force band to the object in each data sample. br The regularization of CSI can be expressed according to Equation 59, which can be used to calculate the three components is executed before to be fed as input to the neural network.

[0169]

[0170] In some embodiments, the receiver 311 may have algorithmic logic to determine (e.g., select) which of the above-mentioned multiple algorithm options to use for respiration rate estimation. For example, such selection may be based on the level of residual AGC gain error and residual phase error in the compensated CSI, which may be pre-configured values, or may be derived from the statistics of the compensated CSI. The calculated value.

[0171] In some embodiments, the estimated respiration rate Doppler binning v 0 and / or delayed binning u 0 can be input into another tracking algorithm (such as a Kalman filter) to further reduce the error. In one embodiment, v 0 、u 0 The estimates of may also be saved and used as a priori side information for future respiration rate estimation iterations. In one embodiment, the confidence of these estimates may also be used to trigger the stopping of future iterations of the respiration rate estimation process.

[0172] Fig.18 A method 1800 for estimating respiration rate based on multi-antenna WiFi signals according to an embodiment of the present disclosure is shown. Fig.18 The embodiment of method 1800 shown in FIG. 1 is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure. Method 1800 is performed by, for example, Figure 2B or Figure 3 The electronic device of STA 111 or 311 in the embodiment is respectively implemented. More specifically, the method 1800 that can be executed by the processor 240 of the electronic device executing the computer program includes a computer-readable program code, which, when executed, causes at least one processor to execute the method 1800. For ease of explanation, the method 1800 is described as being executed by the processor 240 executing one or more applications 262.

[0173] At block 1802, processor 240 triggers a CSI-based respiration rate estimation. The process performed at block 1802 may be similar to Figure 5 In one embodiment, the CSI-based respiration rate estimation may be triggered by an external source or a timer. In another embodiment, another algorithm may use the received CSI (e.g., Figure 4 As an example, the CSI frame 402 may be used to trigger the respiration rate estimation. A sudden and / or significant change compared to its past values ​​may be used to trigger the breathing rate estimation process. As another example, the results of a previous round of breathing rate estimation may be used as a trigger to start or stop the breathing rate estimation process. In yet another embodiment, the transmitter 301 may trigger the breathing rate estimation process by sending a frame (e.g., CSI frame 402) to the receiver 311. In yet another embodiment, the receiver 311 continuously and periodically estimates the breathing rate of the subject without a trigger.

[0174] At block 1804, the processor 240 receives a CSI frame among P channel state information (CSI) frames transmitted from M transmit (Tx) antennas of another electronic device. For example, STA 311 receives unprocessed CSI transmitted from AP 301. The processor 240 is operably connected to a transceiver of the electronic device. In some embodiments, the multi-antenna STA 311 receives a sequence of P CSI frames 402 transmitted from the multi-antenna AP 301, such that M>1 and N>1. In other embodiments, the multi-antenna STA 311 receives a sequence of P CSI frames 402 transmitted from a transmitter having a single antenna, such that M=1 and N>1. In yet another embodiment, the single antenna STA 311 receives a sequence of P CSI frames 402 transmitted from a transmitter having a single antenna, such that M=N=1.

[0175] At block 1806, the processor identifies an impairment model for accounting for including one or more errors in the amplitude or phase of the estimated CSI indexed by p and corresponding to the pth CSI frame among the P CSI frames. In some embodiments, to identify the impairment model, the processor 240 identifies an impairment model including a combination of a static component and a dynamic component based on an assumption that the dynamic component includes a breathing rate of the subject 380.

[0176] At block 1810, processor 240 compensates for one or more errors in amplitude and phase of the P CSI estimates via an impairment model. To compensate for one or more errors in amplitude and phase of the P CSI estimates, processor 240 performs the process of at least one of blocks 1812, 1814, or 1816.

[0177] At block 1812, the processor 240 approximates a lost CSI estimate corresponding to a lost CSI frame of the P CSI frames based on the available CSI estimates. The approximation compensates for the lost CSI frame. Embodiments of the present disclosure determine a lost CSI estimate corresponding to a lost CSI frame based on a function of multiple available CSI estimates including a previously available CSI frame and a subsequently available CSI frame. In some embodiments, the processor determines the lost CSI estimate corresponding to the lost CSI frame to be equivalent to a previously available CSI estimate corresponding to a previous CSI frame that is closest to the lost CSI frame. In some embodiments, the processor determines the lost CSI estimate corresponding to the lost CSI frame to be equivalent to a subsequently available CSI estimate corresponding to a subsequent CSI frame that is closest to the lost CSI frame.

[0178] At block 1814, the processor 240 determines at least one automatic gain control (AGC) gain value. To determine the at least one AGC gain value, for each of the P CSI estimates, the processor 240 measures a CSI power value for each of the MN antenna pairs. The MN antenna pairs include N Rx antennas paired with each of the M Tx antennas, respectively, and each of the MN antenna pairs is indexed as m. For each antenna pair m among the MN antenna pairs, the processor 240 applies a clustering algorithm to the measured CSI power values ​​to identify a cluster, thereby identifying W clusters for antenna pair m. For each of the W clusters identified for antenna pair m, the processor 240 determines the square root of each measured CSI power value within the cluster. The average value of the determined square roots is determined as the AGC gain value corresponding to the cluster. To generate the CSI for AGC gain compensation Processor 240 estimates the CSI corresponding to the mth antenna pair Divide by the AGC gain value corresponding to the cluster

[0179] In some embodiments, for each of the P CSI estimates, an AGC gain compensated CSI is generated. Processor 240 measures the CSI power value (Γ p，m ). For each antenna pair m among the MN antenna pairs, based on the difference between the CSI power values ​​measured for consecutive CSI estimates not being exceeded by a threshold (ie, |Γ p，m -Γ p-1，m |≥threshold is true), determines to compensate for one or more errors in the magnitude of the corresponding CSI estimate indexed by p among the P CSI estimates. Processor 240 compensates the channel model by generating an AGC gain as expressed in Equation 9 to compensate for one or more errors in the amplitude of the corresponding CSI frame caused by the AGC gain.

[0180] In certain embodiments, the processor detects corrupted CSI estimates based on a function of an outlier threshold and the P compensated CSI estimates.The processor 240 removes and / or replaces corrupted CSI frames (eg, corrupted CSI samples) within the P compensated CSI estimates.

[0181] At block 1816, the processor 240 calculates one or more errors in phase based on the weighted least squares solution. More specifically, the processor 240 calculates the timing error and one or more errors in phase of the pth CSI estimate as a weighted least squares solution based on an estimate of a static component of the pth CSI estimate.

[0182] At block 1820 , processor 240 determines whether to perform respiration estimation. In response to determining to perform respiration estimation, the method proceeds to block 1830 .

[0183] At block 1830, the processor estimates the respiration rate (f br ). Box 1830 includes boxes 1832, 1834 and 1836.

[0184] At block 1832, the processor 240 defines different spatial dimensions of the Doppler delay transform of the compensated CSI representing different spatial dimensions of the P compensated CSI estimates. For example, as shown and provided in Equations 30 and 31, in some embodiments, the processor 240 splits the Doppler delay CSI into different components corresponding to different spatial sub-dimensions, respectively (e.g., and and ). The different spatial sub-dimensions include: a first spatial sub-dimension, in which a first component of the Doppler delay CSI is affected by the residual AGC gain error and is resilient to the residual phase error; a second spatial sub-dimension, in which a second component of the Doppler delay CSI is affected by the residual phase error and is resilient to the residual AGC gain error; and a third spatial sub-dimension, in which a third component of the Doppler delay CSI is resilient to both the residual AGC gain error and the residual phase error.

[0185] At block 1834, processor 240 generates a The Doppler delay transform of the compensated CSI is used to estimate the respiratory rate by selecting a respiratory rate algorithm from a plurality of algorithms (eg, Algorithms 1-8). Here, v represents the Doppler frequency bin index, and u represents the delay bin index.

[0186] At block 1836, processor 240 generates a respiration rate estimate f using the spatial components of the CSI and the selected algorithm for estimating respiration rate. br In certain embodiments, to perform the selected respiration rate algorithm, processor 240 combines signals in each of the first, second, and third spatial sub-dimensions to reduce the effects of residual AGC gain error and residual phase error on the estimated respiration rate. In certain embodiments, to perform the selected respiration rate algorithm, processor 240 identifies different components of the Doppler delay transform of the compensated CSI and assigns weights to each of the different components of the Doppler delay transform of the compensated CSI.

[0187] although Fig.18 An example method 1800 for estimating respiration rate based on multi-antenna WiFi signals is shown, but may be used for Fig.14 For example, although shown as a series of steps, Fig.18 The various steps in can overlap, occur in parallel, occur in a different order, or occur any number of times. For example, processor 240 can perform additional steps or functions based on the estimated breathing rate.

[0188] The above flow charts illustrate example methods that can be implemented according to the principles of the present disclosure, and various changes can be made to the methods shown in the flow charts herein. For example, although shown as a series of steps, the various steps in each figure can overlap, occur in parallel, occur in different orders, or occur multiple times. In another example, steps can be omitted or replaced by other steps.

[0189] Although the drawings show different examples of user equipment, various changes may be made to the drawings. For example, the user equipment may include any number of each component in any suitable arrangement. Generally, the drawings do not limit the scope of the present disclosure to any particular (one or more) configurations. In addition, although the drawings show operating environments in which various user equipment features disclosed in this patent document may be used, these features may be used in any other suitable system.

[0190] Although the present disclosure has been described using exemplary embodiments, various changes and modifications may be suggested to those skilled in the art. The present disclosure is intended to cover these changes and modifications that fall within the scope of the appended claims. The description in this application should not be interpreted as implying that any particular element, step or function is an essential element that must be included in the scope of the claims. The scope of the patent subject matter is defined by the claims.

Claims

1. A method for estimating a respiratory rate by an electronic device, comprising: receiving (1804) a CSI frame among P channel state information CSI frames transmitted from another electronic device; estimating the CSI as an available CSI estimate based on each of the received CSI frames; identifying (1806) a model for accounting for including one or more errors in an amplitude or phase of an estimated CSI indexed p and corresponding to a pth CSI frame of the P CSI frames; The one or more errors in the amplitude or phase of the P CSI estimates are compensated (1810) by the identified model, wherein compensating for the one or more errors in the amplitude or phase of the P CSI estimates comprises at least one of performing the following operations: approximating a lost CSI estimate corresponding to a lost CSI frame of the P CSI frames based on the available CSI estimate (1812), determining (1814) at least one automatic gain control (AGC) gain value, or calculating (1816) the one or more errors in phase based on a weighted least squares solution; and A respiration rate of the subject is estimated (1830) based on multiple spatial dimensions of the P compensated CSI estimates.

2. The method according to claim 1, wherein: The lost CSI estimate corresponding to the lost CSI frame is approximated as one of: is equal to a previously available CSI estimate corresponding to a previous CSI frame closest to the lost CSI frame; is equal to a subsequently available CSI estimate corresponding to a subsequent CSI frame that is closest to the lost CSI frame; or Based on a function of a plurality of available CSI estimates, the plurality of available CSI estimates comprising the previously available CSI frame and the subsequently available CSI frame.

3. The method according to claim 1, wherein: Determining the at least one AGC gain value comprises: For each of the P CSI estimates, measuring a CSI power value of each of MN antenna pairs, the MN antenna pairs comprising N Rx antennas of the electronic device respectively paired with each of the M Tx antennas of the other electronic device, the index of each of the MN antenna pairs being m; For each antenna pair m among the MN antenna pairs, applying a clustering algorithm to the measured CSI power values ​​to identify W clusters; and For each of the W identified clusters: Determine the square root of each measured CSI power value within the cluster The average value of the determined square roots is determined as the AGC gain value corresponding to the cluster as well as The CSI estimate corresponding to the mth antenna pair is Divided by the AGC gain value corresponding to the cluster Thus generating the AGC gain compensated CSI 4. The method according to claim 1, further comprising: For each of the P CSI estimates, measure the CSI power value Γ of each of the MN antenna pairs p，m , the MN antenna pairs include N Rx antennas of the electronic device respectively paired with each of the M Tx antennas of the other electronic device, and the index of each of the MN antenna pairs is m; as well as For each antenna pair m of the MN antenna pairs, based on determining that a difference between CSI power values ​​measured for consecutive CSI estimates is not exceeded by a threshold, determine to compensate for the one or more errors in the magnitude of a corresponding CSI estimate indexed by p in the P CSI estimates.

5. The method according to claim 1, further comprising: detecting a corrupted CSI estimate based on a function of an outlier threshold and P compensated CSI estimates; as well as The corrupted CSI frames within the P compensated CSI estimates are replaced.

6. The method according to claim 1, further comprising: The one or more errors in timing error and phase for the pth CSI estimate are calculated as a weighted least squares solution based on an estimate of a static component of the pth CSI estimate.

7. The method according to claim 1, further comprising: Based on the expression , selecting a respiration rate algorithm from a plurality of algorithms for estimating respiration rate based on a Doppler delay transform of the compensated CSI of , wherein v represents a Doppler frequency bin index and u represents a delay bin index; and The different spatial dimensions of the Doppler delay transform of the compensated CSI represent the different spatial dimensions of the P compensated CSI estimates.

8. The method according to claim 7 further comprises splitting the Doppler delay CSI into different components corresponding to different spatial sub-dimensions, wherein the different spatial sub-dimensions include: a first spatial sub-dimension, wherein a first component of the Doppler delay CSI is affected by residual AGC gain error and is resilient to residual phase error; a second spatial sub-dimension, wherein a second component of the Doppler delay CSI is affected by residual phase error and is resilient to residual AGC gain error; and a third spatial sub-dimension, wherein a third component of the Doppler delay CSI is resilient to both residual AGC gain error and residual phase error; and Wherein executing the selected respiration rate algorithm comprises combining signals in each of the first spatial sub-dimension, the second spatial sub-dimension, and the third spatial sub-dimension to reduce the effect of residual AGC gain error and residual phase error on the estimated respiration rate.

9. The method according to claim 7, wherein: Execution of the selected respiration rate algorithm also includes: identifying different components of the Doppler delay shift of the compensated CSI; A weight is assigned to each of the different components of the Doppler delay transform of the compensated CSI.

10. The method according to claim 1, wherein: Identifying the model further comprises identifying the model comprising a combination of a static component and a dynamic component based on an assumption that the dynamic component comprises a breathing rate of the subject.

11. An electronic device (111) for estimating a respiratory rate, the electronic device comprising: A transceiver (210), comprising N receiving Rx antennas; as well as A processor (240) operatively connected to the transceiver and configured to: receiving a CSI frame among P channel state information CSI frames sent from another electronic device; estimating the CSI as an available CSI estimate based on each of the received CSI frames; identifying a model for accounting for inclusion of one or more errors in an amplitude or phase of an estimated CSI indexed p and corresponding to a p-th CSI frame of the P CSI frames; Compensating the one or more errors in the amplitude or phase of the P CSI estimates by the identified model, wherein, to compensate for the one or more errors in the amplitude or phase of the P CSI estimates, the processor is further configured to perform at least one of the following operations: approximating a lost CSI estimate corresponding to a lost CSI frame of the P CSI frames based on the available CSI estimate, determining at least one automatic gain control (AGC) gain value, or calculating the one or more errors in the phase based on a weighted least squares solution; as well as A respiration rate of the subject is estimated based on a plurality of spatial dimensions of the P compensated CSI estimates.

12. The electronic device according to claim 11, wherein: The processor is further configured to operate according to one of claims 2 to 10.

13. A non-transitory computer-readable storage medium storing instructions, the instructions, when executed by a processor (240) of an electronic device (111), causing the electronic device (111) to perform operations, the operations comprising: receiving a CSI frame among P channel state information CSI frames sent from another electronic device; estimating the CSI as an available CSI estimate based on each of the received CSI frames; identifying a model for accounting for inclusion of one or more errors in an amplitude or phase of an estimated CSI indexed p and corresponding to a p-th CSI frame of the P CSI frames; Compensating the one or more errors in the amplitude or phase of the P CSI estimates by the identified model, wherein compensating the one or more errors in the amplitude or phase of the P CSI estimates comprises at least one of performing the following operations: approximating a lost CSI estimate corresponding to a lost CSI frame of the P CSI frames based on the available CSI estimate, determining at least one automatic gain control (AGC) gain value, or calculating the one or more errors in the phase based on a weighted least squares solution; as well as A respiration rate of the subject is estimated based on a plurality of spatial dimensions of the P compensated CSI estimates.

14. The non-transitory computer-readable storage medium of claim 13, wherein: The operations also include at least one operation of the method according to one of claims 2 to 10.

Images