ToF estimation method based on RF signal and related product

Abstract

The invention discloses a ToF estimation method based on an RF signal, and the method comprises the steps: carrying out the difference of a CSI reading of a CSI signal at a time point, and obtaining a time domain difference vector from which static multipath is removed; then selecting a reference time domain difference vector from the multiple groups of time domain difference vectors, and determining phase differences between other time domain difference vectors and the reference time domain difference vector; determining first index values of other time domain difference vectors based on the phase difference, and determining second index values of other time domain difference vectors based on other time domain difference vectors and the reference time domain difference vector; according to the first index value and the second index value, selecting a time domain difference vector subset which meets a first index threshold value and a second index threshold value at the same time, has continuous carrier waves and is counteracted in the dynamic interference multipath from other time domain difference vectors; and finally, accurately determining a target path ToF according to the static and dynamic multipath removed time domain differential vector quantum sets.

Description

Technical Field

[0001] The present disclosure generally relates to the technical field of time of flight (ToF) estimation, and more specifically, to a ToF estimation method based on RF signals and related products. Background Art

[0002] ToF (Time of Flight) indicates the time required for a signal to propagate from the transmitter to the receiver through a channel. In Wi-Fi sensing applications, the application of ToF technology is becoming more and more extensive. For example, in smart home systems, ToF can be used to identify the number of people and locations in a room to automatically adjust home appliances, improve living experience and energy efficiency. In robot navigation and drone flight, ToF can be used to detect the distance to obstacles to avoid collisions with obstacles and improve driving reliability.

[0003] ToF can generally be measured through CSI (Channel State Information) signals. Furthermore, CSI signals can be obtained through widely used RF (Radio Frequency Signal) devices such as Wi-Fi network interface cards (NICs), and Wi-Fi network interface cards can be Intel 5300, Atheros 9580, etc. In theory, when there is only a single propagation path in the environment, the CSI signal can be subjected to an inverse fast Fourier transform (IFFT) to find out the time delay of the CSI signal reaching the receiving end through the single propagation path, i.e., ToF.

[0004] However, in real-world indoor environments, after a signal is sent from the transmitter, it will travel through different propagation paths to reach the receiver. Figure 4 , Figure 4 This is a schematic diagram of the multipath effect of signal reflection in a real scene. Figure 4As shown, among the signals emitted by the transmitter, in addition to the signals reflected to the receiving end by targets such as users, there are also some signals that propagate through a direct path or are reflected by static objects such as walls to reach the receiving end, and the length of this propagation path does not change with time. In addition, some signals are reflected by surrounding moving objects to reach the receiving end. Among the aforementioned multiple propagation paths, the propagation path that is reflected to the receiving end by a single reflection from the target is called the target path, the propagation path that is reflected by direct propagation or static objects is called static multipath, and the propagation path that is reflected by moving objects is called interference dynamic multipath. Among the above-mentioned multiple propagation paths, the ToF of the signal propagated through the target path is what we need to solve, and the signal performance of the signal propagated through the static multipath and the interference dynamic multipath is similar to that of the signal propagated through the target path, which will interfere with the ToF estimation of the target path, thereby causing the ToF estimation of the target path to become very difficult, and it belongs to the interference path.

[0005] In order to find the delay ToF introduced by the target path from multiple propagation paths. One method is to expand the bandwidth of the transmitted signal. Specifically, a larger bandwidth can enhance the ability to distinguish multiple paths ToF. For example, multiple channels in the 2.4G and 5G frequency bands can be combined to achieve a bandwidth of nearly 1Gbps to obtain decimeter-level resolution. However, this method is incompatible with the existing Wi-Fi protocol and requires changes to hardware and software. In addition, the Wi-Fi protocol specifies the bandwidth resource range, which limits the available bandwidth size and cannot be applied to Wi-Fi devices. Another method is to distinguish multiple propagation paths by using multiple parameter dimensions, including ToF, AoA (Angle of Arrival), DFS (Dynamic Frequency Selection, Doppler shift) and AoD (Angle of Departure, signal departure angle) to improve multipath discrimination. When the limitation of signal bandwidth leads to low ToF resolution, if the signals arriving via different paths arrive at the receiving end with a large difference in incident angles, they can be distinguished by the AoA dimension. However, since the polarization directions and installation locations of many Wi-Fi devices are not designed to meet the requirements for calculating AoA, it is difficult to obtain accurate AoA information on the Wi-Fi devices. Therefore, this method is also not applicable to Wi-Fi devices.

[0006] In view of this, there is an urgent need to provide a ToF estimation method based on RF signals and related products, so that they can be applied to RF devices with bandwidth limitations, such as Wi-Fi devices, to improve the breadth of ToF applications. Summary of the invention

[0007] In order to at least solve one or more of the technical problems mentioned above, the present disclosure proposes a ToF estimation method based on RF signals and related products in multiple aspects.

[0008] In the first aspect, the present disclosure provides a ToF estimation method based on RF signals, which may include: acquiring multiple groups of RF signals of different subcarriers collected by an RF receiving device, and parsing multiple groups of CSI signals of different subcarriers from the RF signals; differentiating the CSI readings of each group of the CSI signals at a time point to obtain a time domain differential vector corresponding to each group of the CSI signals; selecting a reference time domain differential vector from the multiple groups of the time domain differential vectors, and determining the phase difference between other time domain differential vectors and the reference time domain differential vector; based on the phase difference, determining a first index value of the other time domain differential vector; based on the other time domain differential vector and the reference time domain differential vector, determining a second index value of the other time domain differential vector; based on the first index value and the second index value, selecting a subset of time domain differential vectors from the other time domain differential vectors that simultaneously meets the first index threshold and the second index threshold and has continuous carriers; and determining the ToF of the target path based on the subset of time domain differential vectors.

[0009] In some embodiments, the first indicator value is expressed as:

[0010]

[0011] Wherein, R(Δf,Δθ(Δf)) represents the first index value, Δf represents the frequency interval between other time domain differential vectors of different subcarriers and the reference time domain differential vector, σ Δf represents the standard deviation of Δf, Δθ(Δf) represents the phase difference between other time-domain difference vectors and the reference time-domain difference vector, and cov(Δf,Δθ(Δf)) represents the covariance of Δf and Δθ(Δf).

[0012] In some embodiments, the second indicator value is expressed as:

[0013]

[0014] Wherein, D(f1,f2) represents the second index value, E(f1,t) represents the reference time domain difference vector with frequency f1, Represents the other time-domain difference vector with frequency f2.

[0015] In some embodiments, the first indicator threshold is close to 1 and the second indicator threshold is close to 0.

[0016] In some embodiments, the determining of the ToF of the target path according to the time domain differential vector subset includes: selecting a reference time domain differential vector from the time domain differential vector subset, determining a phase difference between other time domain differential vectors and the reference time domain differential vector; calculating a slope of the time domain differential vector subset according to the phase difference; calculating the ToF of the target path using the slope; wherein the slope is Represents the average phase difference between other time domain differential vectors of different subcarriers and the reference time domain differential vector, Represents the frequency interval average of other time domain differential vectors of different subcarriers and the reference time domain differential vector, and the ToF is

[0017] In some embodiments, selecting a reference time domain differential vector from multiple groups of time domain differential vectors and determining the phase difference between other time domain differential vectors and the reference time domain differential vector includes: selecting a reference time domain differential vector from multiple groups of time domain differential vectors; rotating the other time domain differential vectors by different angles on the complex plane; after the other time domain differential vectors are rotated by different angles, calculating the sum of the distances between the other time domain differential vectors and the reference time domain differential vector sampled at corresponding moments on the complex plane; and taking the rotation angle corresponding to the minimum value of the sum of the distances as the phase difference between the other time domain differential vector and the reference time domain differential vector.

[0018] In some embodiments, selecting a reference time domain differential vector from the time domain differential vector subset and determining the phase difference between other time domain differential vectors and the reference time domain differential vector includes: selecting a reference time domain differential vector from the time domain differential vector subset; rotating other time domain differential vectors in the time domain differential vector subset by different angles on the complex plane; after the other time domain differential vectors are rotated by different angles, calculating the sum of the distances between the other time domain differential vectors and the reference time domain differential vector sampled at corresponding moments on the complex plane; and taking the rotation angle corresponding to the minimum value of the sum of the distances as the phase difference between the other time domain differential vectors and the reference time domain differential vector.

[0019] In some embodiments, the step of acquiring multiple groups of RF signals of different subcarriers collected by an RF receiving device and parsing the multiple groups of CSI signals of different subcarriers from the RF signals includes: collecting RF signals under the same subcarriers using different antennas of the RF receiving device; parsing at least two groups of initial CSI signals of the same subcarriers from the RF signals; and dividing the two groups of initial CSI signals of the same subcarriers to obtain the CSI signals.

[0020] In some embodiments, after determining the ToF of the target path according to the time domain differential vector subset, the method further includes: calculating the distance of the target path using the ToF of the target path and the propagation speed of the CSI signal.

[0021] In a second aspect, the present disclosure provides an RF device, comprising: an RF transmitting device and an RF receiving device; wherein the RF transmitting device is provided with at least 1 antenna for transmitting RF signals; and the RF receiving device is provided with at least 2 antennas for receiving RF signals.

[0022] In some embodiments, the antenna is a vertically polarized omnidirectional antenna.

[0023] In some embodiments, the antenna is perpendicular or parallel to the ground.

[0024] Through a ToF estimation method based on RF signals and related products provided above, the disclosed embodiment differentiates the CSI readings of CSI signals of different subcarriers at time points to obtain a time domain differential vector with static multipath removed, and then further selects a subset of time domain differential vectors in which the interference dynamic multipath is offset and the carrier is continuous in the time domain differential vector. The interference dynamic multipath of the time domain differential vector subset is removed and is determined only by the target path, and then the ToF of the target path can be accurately determined based on the time domain differential vector subset. The ToF of the target path calculated by the above method can be directly applied to RF devices with bandwidth limitations such as Wi-Fi devices, without the need to expand its bandwidth or modify the device hardware, which can improve the breadth of ToF applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the detailed description below with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0026] Figure 1 is a flowchart of steps of a ToF estimation method based on RF signals according to some embodiments of the present disclosure;

[0027] Figure 2 is a flowchart of steps of a ToF estimation method based on RF signals according to some embodiments of the present disclosure;

[0028] Figure 3 is a flowchart of steps of a ToF estimation method based on RF signals according to some embodiments of the present disclosure;

[0029] Figure 4 This is a schematic diagram of the multipath effect of signal reflection in a real scenario;

[0030] Figure 5 This is a schematic diagram of the superposition of multipath and environmental noise in a real scene;

[0031] Figure 6 It is a schematic diagram of time domain differential vectors at different moments of some embodiments of the present disclosure. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0033] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0034] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0035] As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0036] The specific implementation of the present disclosure is described in detail below with reference to the accompanying drawings.

[0037] See also Figure 1 , Figure 1 1 is a flowchart of the steps of a ToF estimation method 10 based on RF signals according to some embodiments of the present disclosure. Figure 1As shown, method 10 includes steps S101 to S107. First, at step S101, method 10 obtains multiple groups of RF signals of different subcarriers collected by the RF receiving device, and parses the RF signals to obtain multiple groups of CSI signals of different subcarriers. The disclosed embodiment extracts CSI signals from the collected RF signals of different subcarriers for ToF estimation of the target path. The CSI signal represents the characteristics and state of the wireless channel at a specific time and frequency, which includes information such as the fading characteristics, propagation delay, and multipath effect of the signal, which can help understand how the signal propagates in a wireless environment. It should be noted that the multiple groups of RF signals collected by the disclosed embodiment are signals with continuous carriers. In addition, the disclosed embodiment does not limit the number of RF signals collected. It can be understood that the more RF signals are collected, the richer the data can be, which helps to improve the accuracy and reliability of the ToF estimation of the target path. As an example, 30 groups of RF signals with continuous carriers can be collected.

[0038] Furthermore, the present disclosure does not specifically limit how to extract the CSI signal from the RF signal. Those skilled in the art may refer to the specific implementation methods of extracting the CSI signal in the prior art. For example, the gain and phase of the channel may be estimated by comparing the known pilot signal with the pilot signal in the collected RF signal, and then the CSI signal may be extracted from the RF signal based on the channel estimation result.

[0039] In some embodiments, before step S101, method 10 may first deploy an RF transmitting device and an RF receiving device to transmit an RF signal through the RF transmitting device, and collect the RF signal transmitted by the RF transmitting device and propagated by the target through the RF receiving device. The RF transmitting device and the RF receiving device may be any device that supports radio frequency communication protocols and frequency bands, especially devices with bandwidth limitations such as Wi-Fi devices. Among them, the Wi-Fi device is a device with the ability to transmit and receive Wi-Fi signals, and the Wi-Fi signal it transmits is a type of RF signal. Specifically, the Wi-Fi device may be a 4G / 5G mobile terminal device or a LoRa (LongRange) device. As an example, the 4G / 5G mobile terminal device may be a laptop, a mobile phone, a MiniPC, a router, an IoT device, etc.

[0040] In some embodiments, the RF transmitting device and the RF receiving device may be provided with antennas for transmitting and receiving RF signals, respectively. Further, the antennas provided in the RF transmitting device and the RF receiving device may be vertically polarized omnidirectional antennas. The antennas may be provided vertically or parallel to the ground.

[0041] In some embodiments, the RF transmitting device may be provided with at least one antenna for transmitting RF signals. The RF receiving device may be provided with at least two antennas for receiving RF signals, which is not specifically limited in the embodiments disclosed herein. As an example, a WiFi signal with a center frequency of 5.32 GHz may be used as the RF signal, and a MiniPC equipped with an 8 db omnidirectional antenna and an Intel 5300 network card may be used as the RF transmitting device and the RF receiving device.

[0042] In some other embodiments, the RF receiving device is provided with three antennas for receiving RF signals. It is understandable that the more antennas the RF receiving device is provided with, the more RF signals it can collect, and the rich amount of RF signals helps to extract the desired target signal from it for determining the ToF of the target path. In addition, multiple antennas receive RF signals, which can also reduce the impact of multipath effects and improve the quality of RF signals. When configuring the antennas of the RF receiving device, those skilled in the art can determine the number of antennas actually configured based on the number of antenna interfaces of the RF receiving device and the actual RF signal collection requirements.

[0043] The process then proceeds to step S102, where method 10 differentiates the CSI readings of each group of CSI signals at a point in time to obtain a time domain differential vector corresponding to each group of CSI signals. Static multipath refers to a propagation path whose length does not change with time, such as a direct path or a path reflected by a static object in the environment. Its impact on the signal is manifested as a static vector with constant amplitude and phase, which does not change with time. The dynamic vector introduced by the target path changes with time. In the disclosed embodiment, the characteristic that the static vector does not change with time is utilized, and the adjacent CSI readings of each group of CSI signals at a point in time are differentiated so that the static vector in the obtained new time domain differential vector is removed, thereby removing the impact of static multipath. As an example, see Figure 6 , Figure 6 Schematic diagram of time domain differential vectors at different times in some embodiments of the present disclosure. Figure 6 As shown in (a) and (b), (a) and (b) respectively show the CSI signals of two different subcarriers and the time domain difference vector obtained after the CSI signals are differentiated. H1 and H2 are respectively the CSI signals of different subcarriers extracted, and E1 and E2 are respectively the time domain difference vectors formed by the difference of the CSI readings of CSI signal H1 and CSI signal H2 at a time point.

[0044] It should be noted that the present disclosure embodiment does not specifically limit the value of the time interval between adjacent CSI readings, as long as the static multipath component remains unchanged. In addition, when the sampling interval of the CSI signal is short enough, that is, the sampling rate is high enough, the requirement that the static multipath component remains unchanged can be easily met. In addition, when the target moves within a small range, the sampling rate of the CSI signal can be low, which can also meet the requirement that the static multipath component remains unchanged, and the present disclosure embodiment does not specifically limit this.

[0045] To facilitate understanding of the elimination of static vectors in the disclosed embodiment, the process of removing static vectors is modeled below:

[0046] First, in order to focus on the discussion of static multipath, let's simplify the scenario. Assume that there is no interfering dynamic multipath or noise in the CSI signal. In this case, the CSI signal is only affected by the static multipath and the target path, and the CSI signal can be expressed as the following formula (1):

[0047]

[0048] Where H(f,t) represents the CSI signal, part 1 represents the target path, and part 2 represents the static multipath. ρ(.) represents the amplitude attenuation, C represents the speed of light, d(t) represents the target path length, S represents the number of static propagation paths, s is the index of each static path, and d s Represents the static propagation path length.

[0049] Next, the adjacent CSI readings obtained at adjacent time points t and t0 of the CSI signal are subtracted, and the difference between the adjacent CSI readings forms a new time domain difference vector, which can be expressed as the following formula (2):

[0050]

[0051] Where E(f,t) represents the time domain difference vector.

[0052] According to the above formula (2), the time domain difference vector E(f, t) is only determined by the dynamic vector of the target path, and the static vector (ie, static interference multipath) can be removed.

[0053] The time domain differential vector obtained by executing step S102 has removed the influence of static multipath, but the interference dynamic multipath has not been removed. Moreover, both the interference dynamic multipath and the target path will change with the movement of the target, and the properties of the two are very similar. In the case of limited bandwidth, it is difficult to remove the interference dynamic multipath from the target path. Therefore, in the disclosed embodiment, the goal is not to remove the interference dynamic multipath, but to find signals in multiple groups of time domain differential vectors that make the interference dynamic multipath weaker due to multipath cancellation, that is, subsets of time domain differential vectors. The interference dynamic multipaths on these signals are weak in strength due to mutual cancellation, which has a negligible effect on the target path.

[0054] Therefore, after executing step S102, the process reaches step S103. At step S103, method 10 selects a reference time domain differential vector from multiple groups of time domain differential vectors, and determines the phase difference between other time domain differential vectors and the reference time domain differential vector. In some embodiments, the reference time domain differential vector can be randomly selected, for example, the time domain differential vector with the lowest carrier frequency is selected as the reference time domain differential vector.

[0055] Since the CSI signals of different subcarriers are affected by static multipath interference, the phase difference of the target path cannot be accurately obtained. The disclosed embodiment utilizes the characteristic that the phase difference between the time domain differential vectors is equal to the phase difference between the CSI signals of different subcarriers, and improves the determination of the ToF of the target path based on the phase difference introduced by the target path in the CSI signals of different subcarriers to the determination of the ToF of the target path based on the phase difference between the time domain differential vectors of different subcarriers. The time domain differential vector is not affected by static interference multipath, so the accuracy of the extracted phase difference can be improved, thereby achieving the purpose of improving the accuracy of the ToF estimation of the target path.

[0056] As an example, Figure 6 As shown in Figures (a) and (b), and They correspond to the target path phase angles of different subcarrier CSI signals H1 and H2 respectively. The phase difference introduced by the target path in CSI signal H1 and CSI signal H2 on different carriers is like Figure 6 As shown in Figure (c), θ1 and θ2 correspond to the phase angles of the time domain differential vectors E1 and E2 of different subcarriers, respectively, and the phase difference between the time domain differential vector E1 and the time domain differential vector E2 is θ2-θ1. It is easy to prove that

[0057] Next, in order to verify the viewpoint proposed in some embodiments of the present disclosure that the phase difference of the time domain differential vectors of different subcarriers is equal to the phase difference introduced by the target path in the CSI signals of different subcarriers, a proof process is given.

[0058] First, the target path length is divided into two parts, namely d(t) = d(t0) + Δd(t). d(t0) represents the path length at time t0, while Δd(t) represents the change of the path length at time t relative to the path length at time t0. Based on the above target path length, the time domain difference vector can be expressed as the following formula (3):

[0059]

[0060] According to formula (3), the phase of the time domain difference vector consists of two parts: and in is determined by the target path length at time t0, and It is related to the movement of the target.

[0061] Furthermore, when the target moves within a small range, such as waving or breathing, the signal phase is introduced by the tiny movement Δd(t). The time domain differential vectors of different subcarriers are almost the same. For example, when the target path length varies within a range of 10cm, the maximum phase changes of the two subcarriers at 5.3GHz and 5.32GHz are 11.1003 radians and 11.1422 radians, respectively. If the phase difference is ignored, the maximum error of ToF will be about 0.33ns, which is equivalent to a path length of 10cm. Since multiple CSI signals are collected during the movement, some of the CSI signals are collected when the target is closer to the RF receiving device, and some of the CSI signals are collected when the target is farther away from the RF receiving device, resulting in an average error of about 5cm. In addition, if the subject performs periodic motion within a small range, such as breathing, and the target path length fluctuates around the path length at time t0, the average error will be close to 0. Based on the above analysis, it can be seen that the smaller the target's motion amplitude, the smaller the phase difference between different subcarriers. The smaller the effect of the alignment, the more accurate the estimate. When estimating ToF during slight body movement or breathing, the theoretical error is only a few millimeters or centimeters.

[0062] Furthermore, the phase difference between the time domain differential vectors from different subcarriers can be expressed as the following equation (4):

[0063]

[0064] In summary, the phase difference between the time domain differential vectors from different subcarriers is mainly dominated by the target path, and the aforementioned phase difference introduced by the target path in the CSI signals from different subcarriers is also dominated by the target path. This proves that the phase difference between the time domain differential vectors from different subcarriers is equal to the phase difference introduced by the target path in the CSI signals from different subcarriers.

[0065] Then the process proceeds to step S104, at which method 10 determines a first index value of other time domain differential vectors based on the phase difference. In some embodiments, the first index value represents the linear relationship between the phases of the time domain differential vectors of different subcarriers, which can be expressed as the following formula (5):

[0066]

[0067] Wherein, R(Δf,Δθ(Δf)) represents the first index, Δf represents the frequency interval between other time domain differential vectors of different subcarriers and the reference time domain differential vector, σ Δf represents the standard deviation of Δf, Δθ(Δf) represents the phase difference between other time-domain difference vectors and the reference time-domain difference vector, and cov(Δf,Δθ(Δf)) represents the covariance of Δf and Δθ(Δf).

[0068] In the disclosed embodiment, the carrier frequency of the time-domain differential vector is a known variable, so the frequency interval Δf between different subcarrier time-domain differential vectors and the standard deviation σ of Δf are Δf It can be obtained by pre-calculation as a known variable. The phase difference of the time domain differential variable has been obtained by the aforementioned step S103. In this step S04, method 10 substitutes the phase difference into the equation (5) corresponding to the above first index value to obtain the first index value of each group of other time domain differential vectors.

[0069] Then, the process reaches step S105, at which step S105, method 10 determines the second index value of the other time domain differential vector based on the other time domain differential vector and the reference time domain differential vector. In some embodiments, the second index value is the sum of the distances of the corresponding sampling points of the other time domain differential vectors of different subcarriers and the reference time domain differential vector on the complex plane, which characterizes the signal pattern similarity of the other time domain differential vectors of different subcarriers and the reference time domain differential vector on the complex plane. The second index value can be expressed as the following formula (6):

[0070]

[0071] Where E(f1,t) represents the reference time domain difference vector with frequency f1, Represents the other time-domain difference vector with frequency f2.

[0072] Further, the process reaches step S106. In step S106, method 10 selects a subset of time domain differential vectors that simultaneously meet the first indicator threshold and the second indicator threshold and have continuous carriers from other time domain differential vectors based on the first indicator value and the second indicator value.

[0073] In the disclosed embodiment, a time domain differential vector whose first index value satisfies the first index threshold and whose second index value satisfies the second index threshold is selected from other time domain differential vectors, and then a time domain differential vector with continuous carrier is selected from the multiple time domain differential vectors as a time domain differential vector to form a subset of time domain differential vectors. For example, the time domain differential vectors of different subcarriers are numbered according to the carrier order, and are 1 to 30 time domain differential vectors respectively. Among them, the time domain differential vectors whose first index value satisfies the first index threshold are numbered 9 to 11, 15, and 16 respectively. The time domain differential vectors whose second index value satisfies the second index threshold are numbered 8 to 10, 12, and 15 respectively. Then the time domain differential vector that satisfies both the first index threshold and the second index threshold and has continuous carrier is the time domain differential vector numbered 9 to 10, so the time domain differential vectors numbered 9 to 10 are listed as time domain differential vectors to form a subset of time domain differential vectors.

[0074] Further, when there are signals in the time domain differential vectors from different subcarriers that have undergone dynamic multipath mutual cancellation, the first indicator value is close to 1. Therefore, the first indicator threshold can be set to a value close to 1. Those skilled in the art can set the first indicator threshold according to actual accuracy requirements. For example, the first indicator threshold can be 0.85 or 0.92. When the time domain differential vectors of two different subcarriers exhibit the same pattern, their samples at any given moment will overlap on the complex plane, so that the total distance obtained by accumulating the sampling points at multiple moments, that is, the second indicator value, has a minimum value close to 0. Therefore, the second indicator threshold can be set to a value close to 0. Those skilled in the art can set it according to actual accuracy requirements. For example, the second indicator threshold can be 0.05 or 0.02.

[0075] In order to facilitate verification of the situation in which some signals interfering with dynamic multipath cancellation exist in the time domain differential vectors of different subcarriers proposed in some embodiments of the present disclosure, an explanation is given below.

[0076] The time domain differential vector obtained by step S102 of the embodiment of the present disclosure has its signal in which the static multipath has been removed. Assuming that only the interfering dynamic multipath and the target path remain in the time domain differential vector at this time, the time domain differential vector can be expressed as the following formula (7):

[0077]

[0078] Among them, part 1 represents the target path, and part 3 represents the interference dynamic multipath.

[0079] For a given subcarrier, its wavelength is λ, and the propagation path length at time t is d1(t), which will introduce a phase change Considering that there are many propagation paths in the environment, there may be another path at time t whose length is The phase change introduced by this path is The phase change introduced by the second path is opposite to the phase change introduced by the first path. When the RF signal propagates along these two paths and reaches the RF receiving device, the signal changes introduced by them at time t will cancel each other out, thus not affecting the signal changes introduced by the target path.

[0080] Rich spatial signal diversity (samples collected on different antennas) and frequency signal diversity (samples collected on different subcarriers) greatly increase the possibility of the above situation. The position difference between antennas causes the signal to reach different antennas with different propagation path lengths, so the phase of the signal reaching different antennas is different. The same applies to frequency diversity. The signals on different subcarriers of the same antenna will also introduce different phases on different carriers due to the difference in wavelength. Because of the differences in the arrival of signals on different antennas and carriers, many superposition combinations occur, some of which happen to superimpose and cancel each other, making the interference dynamic multipath weaker. In addition, when the target moves in a small range, such as breathing, the signal propagation path length changes with time due to the movement of the target. However, because the path length changes very little, the superposition remains roughly unchanged, and the interference dynamic multipath components still cancel each other out at different times.

[0081] In summary, it is shown that there are some signals in the time domain difference vectors of different subcarriers that interfere with dynamic multipath cancellation.

[0082] Finally, the process reaches step S107, where method 10 determines the ToF of the target path based on the time domain differential vector subset. In some embodiments, a reference time domain differential vector can be selected from the time domain differential vector subset to determine the phase difference between other time domain differential vectors and the reference time domain differential vector. Then, the slope of the time domain differential vector subset is calculated based on the phase difference. Further, the ToF of the target path is calculated using the slope.

[0083] In some embodiments, the slope is represents the mean phase difference of the time domain difference vector, represents the mean frequency interval between other time domain differential vectors and the reference time domain differential vector, ToF is

[0084] In some embodiments, the reference time domain differential vector may be randomly selected, for example, the time domain differential vector with the lowest carrier frequency is selected as the reference time domain differential vector.

[0085] Next, in order to verify the view that the phase difference between time-domain differential vectors of different subcarriers proposed in some embodiments of the present disclosure can be used to calculate ToF, the reasoning process is given:

[0086] When the CSI signal with carrier frequency f propagates from the RF transmitting device to the RF receiving device through a path of length d(t) at time t, a certain delay, namely ToF, will be generated. This delay ToF can be expressed as the following formula (8):

[0087]

[0088] Where τ represents ToF, d(t) represents the target path length, and c represents the speed of light.

[0089] Furthermore, to focus on the discussion of the target path ToF, it is assumed that in an ideal environment, the CSI signal is not affected by static multipath, interfering dynamic multipath, and noise, and is only determined by the target path d(t). Then the CSI signal H(f,t) can be expressed as the following formula (9):

[0090]

[0091] According to the above formula (9), the phase change caused by the target path is:

[0092] Furthermore, the combined equation (8) and the phase change The following calculation expression (10) for ToFτ can be obtained:

[0093]

[0094] According to formula (10), if the phase change θ(f) occurring during the propagation of the CSI signal can be obtained, then the ToFτ of the target path can be determined based on the phase change θ(f), and the target path length d(t)=cτ can be further determined based on ToFτ.

[0095] However, the phase wrapping problem causes the phase change obtained from the CSI signal to be unable to be directly used for ToF estimation. The phase change can be expressed as the following equation (11):

[0096]

[0097] The phase winding problem causes the phase change extracted from the CSI signal based on the above formula (11) to lose the periodic information, so that the signal phase change and the reflection path length are no longer one-to-one corresponding. Specifically, when the phase change falls within the interval [0,2π], the above formula (10) can be used Accurately measure ToF. However, if the phase change exceeds 2π, the phase size will jump and start accumulation again, so that the phase change always falls within the range of [0,2π], making it impossible to extract the true phase change from the CSI signal of a single subcarrier.

[0098] As an example, for a Wi-Fi signal with a frequency of 2.4GHz, multiple propagation paths with a propagation path length difference of 12cm, such as 3cm, 15cm, 27cm, 39cm, etc., will all produce the same channel phase change, that is, π / 2. Therefore, based only on one carrier phase change information, it is impossible to distinguish these propagation paths of different lengths. Moreover, in actual scenarios, the length of the signal propagation path usually covers several meters or even tens of meters. Measuring ToF and reflection path length using the phase change of a single subcarrier obviously faces significant distance ambiguity. In order to ensure that the path length can still be estimated unambiguously when the propagation path length is large, one method is to reduce the signal frequency. According to the basic principles of electromagnetism, the phase change accumulated by the signal during propagation is related to its frequency. The phase change of signals with lower frequencies accumulates more slowly. However, under the 802.11 protocol, the frequency of the Wi-Fi subcarrier is fixed, and it is impossible to directly obtain a lower frequency signal from an existing Wi-Fi device, so it is not applicable to RF devices with fixed frequencies such as Wi-Fi devices. Therefore, the disclosed embodiment uses existing subcarriers to construct a virtual signal to achieve the possibility of equivalently reducing the frequency.

[0099] In the field of Wi-Fi communications, under the 20MHz bandwidth limit, the minimum frequency interval between adjacent subcarrier CSI signals (denoted as f1 and f2) is 312.5kHz. Based on this fact, the disclosed embodiment creates a virtual "subcarrier" by taking the difference between two subcarrier CSI signals with a smaller frequency interval, denoted as Δf = f2-f1. The virtual signal propagates along the same path as the original subcarrier, and its phase change θ (Δf) is equal to the phase difference between subcarriers f1 and f2, which can be written as the following formula (12):

[0100]

[0101] When this virtual signal propagates along a given path to a receiving device, its lower frequency prevents phase wrapping, thereby removing ambiguity in the distance estimate.

[0102] Based on the above formula (12), taking the Wi-Fi protocol as an example, under the 20MHz bandwidth limit, the minimum frequency interval Δf between adjacent subcarriers is 312.5kHz. When the target path length d reaches 960 meters, the phase change θ(Δf) corresponds to 2π. This means that within the range of 960 meters of the target path length d, there is a unique phase difference at any position, eliminating ambiguity, that is, there is no phase wrapping problem. This is sufficient to meet the needs of most daily applications. Therefore, according to the above content, it is verified that there is no phase wrapping problem between CSI signals, and the real phase difference can be obtained from the CSI signal, which further shows that the phase difference calculated from the two subcarrier CSI signals can be used to estimate ToF and the target path length.

[0103] The above verifies that the phase difference θ(f) introduced by the target path in different subcarrier CSI signals can be used to calculate the ToFτ of the target path and the target path length d(t). The phase difference between different subcarrier time domain differential vectors is equal to the phase difference θ(f) introduced by the target path in different subcarrier CSI signals, so the phase difference between different subcarrier time domain differential vectors can be used to calculate the ToFτ of the target path and the target path length d(t).

[0104] In some embodiments, after executing step S107 to obtain the ToF of the target path, method 10 can further use the ToF of the target path and the propagation speed of the CSI signal to calculate the distance of the target path. Wherein the propagation speed of the CSI signal is the speed of light C, then the distance of the target path d(t) = cτ.

[0105] Combination of the above Figure 1 and Figure 6 The ToF estimation method 10 based on RF signals of some embodiments of the present disclosure is generally described. The disclosed embodiment obtains a time domain differential vector with static multipath removed by differentiating the CSI readings of CSI signals of different subcarriers at time points. Then, a reference time domain differential vector is selected from multiple groups of time domain differential vectors to determine the phase difference between other time domain differential vectors and the reference time domain differential vector. Further based on the phase difference, the first index value of other time domain differential vectors is determined, and based on other time domain differential vectors and the reference time domain differential vector, the second index value of other time domain differential vectors is determined. Thus, according to the first index value and the second index value, a subset of time domain differential vectors that simultaneously meet the first index threshold and the second index threshold and whose carrier continuous interference dynamic multipath is offset is selected from other time domain differential vectors, and the time domain differential vector is determined only by the target path, and then the ToF of the target path can be accurately determined according to the subset of time domain differential vectors. The ToF of the target path calculated by the above method can be directly applied to RF devices with bandwidth limitations, such as Wi-Fi devices, without the need to expand the bandwidth or modify the device hardware, which can improve the wide application of ToF. However, those skilled in the art should understand that Figure 1 The methods shown are exemplary rather than limiting, and those skilled in the art can make adjustments as needed.

[0106] Next, we will combine Figure 2 Multiple embodiments and variations of the ToF estimation method 10 based on RF signals according to some embodiments of the present disclosure are further described.

[0107] See also Figure 2 , Figure 2 2 is a flowchart of the steps of a ToF estimation method 20 based on RF signals according to some embodiments of the present disclosure. Figure 2 As shown, method 20 includes steps S201 to S210. First, at step 201, method 20 obtains multiple groups of RF signals of different subcarriers collected by the RF receiving device, and parses the RF signals to obtain multiple groups of CSI signals of different subcarriers. Then step S202 is executed. At step S202, method 20 differentiates the CSI readings of each group of CSI signals at a time point to obtain a time domain differential vector corresponding to each group of CSI signals. In the embodiment of the present disclosure, by utilizing the characteristic that static vectors do not change with time, the time domain differential vectors obtained by differentiating the CSI readings can remove static multipath interference. Steps S201 and S202 are the same or similar to steps S101 and S102 of method 20 mentioned above. The specific implementation method thereof can refer to the corresponding content above, and will not be repeated here.

[0108] In addition to being affected by static multipath and interfering dynamic multipath, CSI signals may also be affected by environmental noise. Figure 5 , Figure 5 This is a schematic diagram of the superposition of multipath and environmental noise in a real scene. Figure 5 As shown, H3 and H4 are CSI signals of different subcarriers obtained by analysis, which are superimposed signals including the target path, static multipath, interfering dynamic multipath and environmental noise. Then, the time domain differential vectors E3 and E4 obtained after the CSI signals H3 and H4 are differentiated and the static multipath interference has been removed are superimposed signals including the target path, interfering dynamic multipath and environmental noise. Therefore, the time domain differential vector after removing the static multipath interference may be affected by environmental noise in addition to the interfering dynamic multipath. At this time, combined with the above formula (8), the time domain differential vector can be expressed as the following formula (13):

[0109]

[0110] Among them, ∈(f,t) represents the environmental noise.

[0111] Under the influence of environmental noise, the phase difference between other time domain differential vectors and the reference time domain differential vector cannot be directly determined. Therefore, in order to obtain an accurate phase difference, after executing step S202, the process proceeds to steps S203 to S206 in sequence.

[0112] At step S203, method 20 selects a reference time domain differential vector from multiple groups of time domain differential vectors. The reference time domain differential vector can be randomly selected, for example, the time domain differential vector with the lowest carrier frequency is selected as the reference time domain differential vector. Then, the process reaches step S204, at which method 20 rotates other time domain differential vectors at different angles on the complex plane. The angle of each rotation of other time domain differential vectors is not specifically limited in the disclosed embodiment. As an example, the rotation angle can be increased by 10 degrees each time. Then the process reaches step S205, at which method 20 calculates the sum of the distances sampled at the corresponding moments of other time domain differential vectors and the reference time domain differential vector on the complex plane after the other time domain differential vectors are rotated at different angles. Then step S206 is executed. At step S206, method 20 uses the rotation angle corresponding to the minimum value of the sum of distances as the phase difference between the other time domain differential vectors and the reference time domain differential vector. In the disclosed embodiment, if the sum of the distances is the smallest, it indicates that the phase angles of the other time domain differential vectors after the current rotation are flush with or equal to the reference time domain differential vectors, and the environmental noise is greatly reduced and can be ignored. The rotation angle corresponding to the minimum sum of the distances is equal to the phase difference introduced by the target diameter ToF.

[0113] To facilitate understanding of the technical idea of ​​using the phase rotation in the complex plane and the minimum sum of the distances at the sampling moment to indicate that the ambient noise is significantly reduced and the corresponding rotation angle is the phase difference in the embodiment of the present disclosure, the principle explanation is given below:

[0114] For a subcarrier with frequency f, ∈(f,t) can be approximated as having a mean of zero and a variance of σ 2 (f) is the additive white noise, that is, When measuring the CSI signal, for any two different subcarriers f1≠f2 or any two different time instants t1≠t2, the noise ∈(f1,t1) and ∈(f2,t2) are independent.

[0115] Inspired by the characteristics of noise, in the past, ToF was usually estimated through a single CSI signal, and then multiple ToF estimates were averaged to reduce the impact of noise. However, when the noise level is high, environmental noise may introduce a large phase offset, causing the signal phase to exceed π or -π, thereby causing phase wrapping. In this case, the average value of multiple samples is not equal to the phase difference of the time domain differential vectors of different subcarriers. And as the noise level increases, more samples will encounter phase wrapping problems, causing the error between the average and the true value to increase. Therefore, the above ToF estimation method will fail when the noise level is high.

[0116] In order to more effectively reduce the impact of noise, the disclosed embodiment improves the method of reducing errors by sampling signals at multiple times. According to the above formula (4), the time domain differential vectors of different subcarriers have the same signal pattern, but different initial phases. The difference between the initial phases is (corresponding to the phase difference of the time domain differential vector), which can be used for ToF estimation of the target path.

[0117] If the signals from different subcarriers are rotated The samples collected from different subcarriers at the same time will overlap on the complex plane. When there is no environmental noise in the signal, the sum of the distances of the samples at the corresponding time on the complex plane is zero. Therefore, the sum of the distances of the samples at the corresponding time on the complex plane can be used to determine whether the phases of different subcarriers have been adjusted to be the same, and whether the rotation angle is equal to the phase difference.

[0118] Furthermore, it can be proved that even in the case of high environmental noise level, the embodiment of the present disclosure can still use the sum of the distances between the samples of the time domain differential vectors of different subcarriers at the corresponding moments as a measurement indicator to determine whether the rotation angle is equal to That is, the phase difference introduced by the target diameter ToF. This process can be formulated as the following formula (14):

[0119]

[0120] According to Kolmogorov's strong law of large numbers, when the number of samples is sufficient, the noise term The correlation distance hardly changes with the signal rotation. i,j equal hour, When the minimum value 0 is reached, the formula as a whole also reaches the minimum value. Based on this, the disclosed embodiment proposes to use the sum of the distances between the samples at the corresponding time as a measurement indicator, so as to extract the accurate phase difference for ToF calculation when the samples are sufficient.

[0121] Further, after executing step S206, the process proceeds to step S207, where method 20 determines the first index value of other time domain differential vectors based on the phase difference. Then continue to execute step S208, where method 20 determines the second index value of other time domain differential vectors based on other time domain differential vectors and the reference time domain differential vector. After determining the first index value and the second index value of other time domain differential vectors, the process arrives at step S209. At step S209, method 20 selects a subset of time domain differential vectors that simultaneously meet the first index threshold and the second index threshold and have continuous carriers from other time domain differential vectors based on the first index value and the second index value. In the disclosed embodiment, the second index value

[0122] ^

[0123] In the labeled value, E(f2,t) indicates that the other time domain differential vector with frequency f2 refers to the other time domain differential vector after angle rotation.

[0124] ^

[0125] Domain differential vector. Except for the difference in the second index E(f2, t), the above steps S207 to S209 are the same or similar to the steps S104 to S106 in the method 10 described above. Therefore, the specific embodiments of the above steps S207 to S209 can refer to the steps S104 to S106 in the method 10 described above, and will not be repeated here.

[0126] Finally, the process reaches step S210, where the method 20 determines the ToF of the target path according to the time domain differential vector subset. Similarly, step S210 is the same or similar to step S107 of the above method 10, and will not be repeated here.

[0127] In addition, it should be noted that the above method 10 selects a reference time domain differential vector from the time domain differential vector subset, and the step flow of determining the phase difference between other time domain differential vectors and the reference time domain differential vector is the same as step S203 to step S206 of the embodiment disclosed in the present invention. That is, the step of selecting a reference time domain differential vector from the time domain differential vector subset and determining the phase difference between other time domain differential vectors and the reference time domain differential vector includes: selecting a reference time domain differential vector from the time domain differential vector subset. Then, the other time domain differential vectors in the time domain differential vector subset are rotated at different angles on the complex plane. After the other time domain differential vectors are rotated at different angles, the sum of the distances sampled at the corresponding moments of the other time domain differential vectors and the reference time domain differential vector on the complex plane is calculated. Thus, the rotation angle corresponding to the minimum value of the sum of the distances is used as the phase difference between the other time domain differential vectors and the reference time domain differential vector. Thus, the slope of the time domain differential vector subset is calculated based on the phase difference, and the ToF of the target path is calculated using the slope. The slope is θ(f) represents the average phase difference between other time domain differential vectors of different subcarriers and the reference time domain differential vector. Represents the frequency interval average of other time domain differential vectors of different subcarriers and the reference time domain differential vector, ToF is

[0128] Combination of the above Figure 2 and Figure 5 The ToF estimation method 20 based on RF signals of some embodiments of the present disclosure is generally described. The disclosed embodiment selects a reference time domain differential vector from the time domain difference vector, and then rotates other time domain differential vectors by different angles on the complex plane. After the other time domain differential vectors are rotated by different angles, the sum of the distances sampled on the complex plane at the corresponding moments of the other time domain differential vectors and the reference time domain differential vector is calculated. Thus, the rotation angle corresponding to the minimum value of the sum of distances is used as the phase difference between the other time domain differential vectors and the reference time domain differential vector. By rotating the other time domain differential vectors by different angles, not only can the phase difference between the other time domain differential vectors and the reference time domain differential vector be accurately obtained, but the rotated other time domain differential vectors can also significantly reduce the environmental noise, thereby improving the accuracy of ToF estimation of the target path. However, it should be understood by those skilled in the art that Figure 2 The methods shown are exemplary rather than limiting, and those skilled in the art can make adjustments as needed.

[0129] Next, we will combine Figure 3 The present disclosure further describes multiple embodiments and variations of the RF signal-based ToF estimation methods 10 and 20 of some embodiments. The RF signal-based ToF estimation method 30 described below can be applied to the methods 10 and 20 described above.

[0130] If the CSI signal obtained by analyzing the RF signal is only affected by the environmental noise, then the CSI reading of the CSI signal at the time point can be directly differentiated to obtain the time domain differential vector. However, due to the device limitations of the RF receiving device, the CSI signal is not only affected by the environmental noise, but also by the impulse noise in the CSI signal amplitude and the random offset in the CSI signal phase caused by the device. At this time, the CSI signal can be expressed as the following formula (15):

[0131]

[0132] Among them, ∈(f,t) represents the environmental noise, A noise (f,t) is the impulse noise in amplitude, θ offset (f,t) is the random phase offset, H air(f, t) is the wireless channel state information under ideal conditions.

[0133] The impulse noise in the CSI signal amplitude and the random offset in the CSI signal phase will also affect the phase difference of the target path, resulting in the inability to obtain an accurate phase difference later. Therefore, in order to reduce the impulse noise in the CSI signal amplitude and the random offset in the CSI signal phase, the disclosed embodiment proposes a ToF estimation method 30 based on RF signals.

[0134] See also Figure 3 , Figure 3 is a flowchart of the steps of a ToF estimation method 30 based on RF signals according to some embodiments of the present disclosure. Figure 3 As shown, method 30 includes steps S301 to S309. First, at step S301, method 30 uses different antennas of an RF receiving device to collect RF signals under the same subcarrier. In the embodiment of the present disclosure, different antennas are used to simultaneously collect a group of RF signals of the same subcarrier, thereby obtaining at least two groups of RF signals with the same subcarrier. Then, at step S302, method 30 parses the RF signal to obtain at least two groups of initial CSI signals of the same subcarrier. After executing step S302, the process proceeds to step S303, in which method 30 divides the two groups of initial CSI signals of the same subcarrier to obtain a CSI signal. In the embodiment of the present disclosure, the amplitude pulse noise and phase offset of the initial CSI signal are consistent between different antennas on the same RF receiving device, and the amplitude pulse noise and phase offset of the initial CSI signal are eliminated by dividing the two groups of initial CSI signals of the same subcarrier collected by different antennas. The result of dividing the initial CSI signal is the CSI signal, in which the impulse noise in the amplitude and the random offset in the phase of the CSI signal are eliminated. At this time, the CSI signal is only affected by the environmental noise and can be expressed as the following formula (16):

[0135] H(f,t)=H air (f,t)+∈(f,t) (16)

[0136] It should be noted that when two sets of initial CSI signals with the same subcarrier are divided, the initial CSI signal on the denominator will affect the initial CSI signal on the numerator. However, the distance between the RF transmitting device and the RF receiving device is close, and the static component of the denominator is relatively strong, so that the denominator can be approximated as a constant, which can be eliminated by calibration.

[0137] Further, the process reaches step S304, where method 30 differentiates the CSI readings of each group of CSI signals at a time point to obtain a time domain differential vector corresponding to each group of CSI signals. In the disclosed embodiment, adjacent CSI readings of CSI signals of each group of different subcarriers at a time point are differentiated, and the result forms a time domain differential vector, which can remove the interference of static multipath. This step S304 is the same or similar to step S102 of method 10 and step S202 of method 20 in the above text. The specific embodiments of these steps can be referred to and will not be repeated here. After step S304, the process executes step S305, where method 30 selects a reference time domain differential vector from multiple groups of time domain differential vectors to determine the phase difference between other time domain differential vectors and the reference time domain differential vector. Similarly, in the disclosed embodiment, since the time domain differential vector is also affected by environmental noise ∈(f, t), the phase difference cannot be directly obtained from the time domain differential vector. In order to obtain an accurate phase difference, the specific implementation manner of step S305 may refer to steps S203 to S206 of the above method 20, which will not be described in detail here.

[0138] To select a subset of time domain differential vectors in which the interfering dynamic multipath is offset. The process then reaches step S306 to step S309 in sequence. At step S306, method 30 determines the first index value of other time domain differential vectors based on the phase difference. Then at step S307, method 30 determines the second index value of other time domain differential vectors based on other time domain differential vectors and the reference time domain differential vector. Further, at step S308, method 30 selects a subset of time domain differential vectors that simultaneously meet the first index threshold and the second index threshold and have continuous carriers from other time domain differential vectors based on the first index value and the second index value.

[0139] Finally, step S309 is executed. At step S309, method 30 determines the ToF of the target path based on the time domain differential vector subset. In the disclosed embodiment, the noise, static interference multipath, and interference dynamic multipath of the time domain differential vector subset are all reduced and can be ignored. It is only determined by the target path, so the ToF of the target path can be accurately obtained therefrom. The specific implementation methods of step S306 and step S309 can refer to method 10 or method 20 described above, which will not be repeated here.

[0140] Combination of the above Figure 3The present disclosure generally describes a ToF estimation method 30 based on RF signals in some embodiments. The initial CSI signals of two groups of the same subcarriers collected by different antennas are divided to eliminate the amplitude pulse noise and phase offset of the initial CSI signal. The result of the division of the initial CSI signals is the CSI signal, in which the pulse noise in the amplitude and the random offset in the phase of the CSI signal are eliminated, thereby further improving the accuracy of the ToF estimation of the target path.

[0141] Based on the above, some embodiments of the present disclosure further provide an RF device, including an RF transmitting device and an RF receiving device. The RF transmitting device is provided with at least one antenna for transmitting RF signals, and the RF receiving device is provided with at least two antennas for receiving RF signals. Specifically, the RF transmitting device and the RF receiving device can be any device that supports radio frequency communication protocols and frequency bands, especially RF devices with bandwidth limitations such as Wi-Fi devices.

[0142] It is understood that a Wi-Fi device is a device that has the ability to send and receive Wi-Fi signals, and the Wi-Fi signals it sends are a type of RF signal. Specifically, a Wi-Fi device can be a 4G / 5G mobile terminal device, a LoRa (LongRange) device. Further, as an example, a 4G / 5G mobile terminal device can be a laptop, a mobile phone, a MiniPC, a router, an IoT device, etc.

[0143] In some other embodiments, the RF receiving device is provided with three antennas for receiving RF signals. It is understandable that the more antennas the RF receiving device is provided with, the more RF signals it can collect, and the rich amount of RF signals helps to extract the desired target signal from it for determining the ToF of the target path. In addition, multiple antennas receive RF signals, which can also reduce the impact of multipath effects and improve the quality of RF signals. When configuring the antennas of the RF receiving device, those skilled in the art can determine the number of antennas actually configured based on the number of antenna interfaces of the RF receiving device and the actual RF signal collection requirements.

[0144] In some embodiments, the antennas provided in the RF transmitting device and the RF receiving device may be vertically polarized omnidirectional antennas. Further, the antennas may be provided vertically or parallel to the ground, which is not specifically limited in the embodiments disclosed herein.

[0145] Taking Wi-Fi signal as RF signal as an example, the WiFi signal with a center frequency of 5.32GHz can be used as the RF signal, and a MiniPC equipped with an 8db omnidirectional antenna and an Intel5300 network card can be used as the RF transmitting device and the RF receiving device.

[0146] In some embodiments, the RF device, as a device supporting radio frequency communication protocols and frequency bands, is originally configured with an RF signal interface. Therefore, an antenna feeder can be used to directly connect the RF device to the antenna, wherein one end of the antenna feeder is connected to the RF signal interface of the RF device and the other end is connected to the antenna, so that the RF device can transmit and receive RF signals through the antenna connected thereto.

[0147] In some embodiments, the antenna connected to the RF device may be installed on the top or side of the RF device, which helps to improve the receiving and transmitting effects of the RF signal.

[0148] The above generally describes the RF devices of some embodiments of the present disclosure. The embodiments of the present disclosure implement the transceiver of RF signals by setting an antenna on the RF device so as to be used in a ToF estimation method based on RF signals described above, so as to improve the wide application of ToF. For example, the RF transceiver device provided in the embodiments of the present disclosure can be used to transmit and receive RF signals and / or execute the ToF estimation method based on RF signals described above in the present disclosure.

[0149] Based on the above, the present disclosure also discloses a computer-readable storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the methods described in the above multiple embodiments or implementation methods are implemented.

[0150] Although multiple embodiments of the present disclosure have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art may think of many changes, modifications, and alternatives without departing from the thought and spirit of the present disclosure. It should be understood that in the process of practicing the present disclosure, various alternatives to the embodiments of the present disclosure described herein may be adopted. The attached claims are intended to define the scope of protection of the present disclosure, and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A ToF estimation method based on RF signal, characterized in that: include: Acquire multiple groups of RF signals of different subcarriers collected by an RF receiving device, and parse the RF signals to obtain multiple groups of CSI signals of different subcarriers; Differentiating the CSI readings of each group of the CSI signals at a time point to obtain a time domain differential vector corresponding to each group of the CSI signals; Selecting a reference time domain differential vector from the multiple groups of time domain differential vectors, and determining a phase difference between other time domain differential vectors and the reference time domain differential vector; Based on the phase difference, determining a first index value of the other time-domain differential vector; Determine a second indicator value of the other time-domain difference vector based on the other time-domain difference vector and the reference time-domain difference vector; According to the first indicator value and the second indicator value, selecting a subset of time domain difference vectors that simultaneously meet the first indicator threshold and the second indicator threshold and have continuous carriers from the other time domain difference vectors; The ToF of the target path is determined according to the time domain difference vector subset.

2. The method according to claim 1, characterized in that The first indicator value is expressed as: Wherein, R(Δf, Δθ(Δf)) represents the first index, Δf represents the frequency interval between other time domain differential vectors of different subcarriers and the reference time domain differential vector, σ Δf represents the standard deviation of Δf, Δθ(Δf) represents the phase difference between the other time-domain difference vectors and the reference time-domain difference vector, and cov(Δf, Δθ(Δf)) represents the covariance of Δf and Δθ(Δf).

3. The method according to claim 1, characterized in that The second index value is expressed as: Wherein, D(f1,f2) represents the second index value, E(f1,t) represents the reference time domain difference vector with frequency f1, Represents the other time-domain difference vector with frequency f2.

4. The method according to claim 1, characterized in that: The first indicator threshold is close to 1, and the second indicator threshold is close to 0.

5. The method according to claim 1, characterized in that Determining the ToF of the target path according to the time domain difference vector subset includes: Selecting a reference time domain differential vector from the time domain differential vector subset, and determining a phase difference between other time domain differential vectors and the reference time domain differential vector; Calculating the slope of the time domain difference vector subset according to the phase difference; Calculating the ToF of the target path using the slope; The slope is Represents the average phase difference between other time domain differential vectors of different subcarriers and the reference time domain differential vector, Represents the frequency interval average of other time domain differential vectors of different subcarriers and the reference time domain differential vector, and the ToF is 6. The method according to claim 1, characterized in that The selecting a reference time domain differential vector from the multiple groups of time domain differential vectors and determining the phase difference between other time domain differential vectors and the reference time domain differential vector comprises: Selecting a reference time-domain difference vector from a plurality of groups of the time-domain difference vectors; Rotating the other time-domain difference vectors by different angles on the complex plane; After the other time domain differential vectors are rotated by different angles, calculating the sum of the distances between the other time domain differential vectors and the reference time domain differential vectors sampled at corresponding moments on the complex plane; The rotation angle corresponding to the minimum value of the sum of the distances is used as the phase difference between the other time-domain difference vector and the reference time-domain difference vector.

7. The method according to claim 5, characterized in that The selecting a reference time domain differential vector from the time domain differential vector subset and determining a phase difference between other time domain differential vectors and the reference time domain differential vector comprises: Selecting a reference time-domain difference vector from the time-domain difference vector subset; Rotating other time-domain difference vectors in the time-domain difference vector subset by different angles on a complex plane; After the other time domain differential vectors are rotated by different angles, calculating the sum of the distances between the other time domain differential vectors and the reference time domain differential vectors sampled at corresponding moments on the complex plane; The rotation angle corresponding to the minimum value of the sum of the distances is used as the phase difference between the other time-domain difference vector and the reference time-domain difference vector.

8. The method according to claim 1, characterized in that The acquiring of multiple groups of RF signals of different subcarriers collected by the RF receiving device and parsing the multiple groups of CSI signals of different subcarriers from the RF signals comprises: Using different antennas of the RF receiving device to collect RF signals under the same subcarrier respectively; Parsing the RF signal to obtain at least two groups of initial CSI signals of the same subcarriers; The initial CSI signals of two groups of identical subcarriers are divided to obtain the CSI signal.

9. The method according to claim 1, characterized in that: After determining the ToF of the target path according to the time domain differential vector subset, the method further includes: The distance of the target path is calculated using the ToF of the target path and the propagation speed of the CSI signal.

10. An RF device, characterized in that: include: RF transmitting equipment and RF receiving equipment; The RF transmitting device is provided with at least one antenna for transmitting RF signals; The RF receiving device is provided with at least two antennas for receiving RF signals.

11. The RF device according to claim 11, characterized in that: The antenna is a vertically polarized omnidirectional antenna.

12. The RF device according to claim 11, characterized in that: The antenna is perpendicular or parallel to the ground.

Images