METHOD, SYSTEM, COMPUTER DEVICE, AND STORAGE MEDIUM FOR NON-CONTACT DETERMINATION OF A SENSING BOUNDARY

Abstract

A method and a system are for non-contact determination of a detection boundary. A Wi-Fi receiving device simultaneously receives data packets by using any two antennas, and measures channel state information; conjugate multiplication is performed on the CSI measured at a moment corresponding to reception of the two antennas, to obtain a new CSI signal stream; a time window with the length M is selected from the new CSI signal stream, and M CSI samples in the time window are collected, to form a conjugate set; the mean of the conjugate set is subtracted from all the samples in the conjugate set, to obtain a zero-mean conjugate set; N zero-mean conjugate sets are obtained for CSI on N sub-carriers simultaneously transmitted; and probability distribution fitting is performed on the absolute values of all elements in the N zero-mean conjugate sets, to calculate a boundary detection feature.

Claims

1. A method for non-contact determination of a sensing boundary, comprising: S1. a Wi-Fi receiving device simultaneously receiving data packets by using two antennas, and measuring Channel State Information (CSI); S2. performing conjugate multiplication on the CSI of the two antennas at each sampling time, to obtain a new CSI signal stream; S3. selecting a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collecting M CSI samples in the time window, to form a conjugate set S(t.sub.0); S4. subtracting the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set; S5. repeating said S2-S4 for CSI on N sub-carriers simultaneously transmitted, to obtain N zero-mean conjugate sets; and performing probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0); and S6. accurately classifying, according to the boundary sensing feature σ(t.sub.0), activities inside and outside a sensing area by means of a sensing boundary formed by walls and extension lines thereof.

2. The method for non-contact determination of the sensing boundary according to claim 1, wherein the new CSI obtained after the conjugate multiplication is: $H_{conj}=H_1\left(t_0+t\right)*H_2^{*}\left(t_0+t\right)\approx A_{s1}{A}_{s2}{e}^{j\left({\phi }_{s1}-{\phi }_{s2}\right)}+A_{s1}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j\left({\phi }_{s1}-{\phi }_d^2\left(t_0+t\right)\right)}+A_{s2}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j\left({\phi }_d^1\left(t_0+t\right)-{\phi }_{s2}\right)}$ in the formula, H.sub.conj represents the new CSI obtained after the conjugate multiplication, H.sub.1(t.sub.0+t) represents CSI of a first antenna at the instant t.sub.0+t, H.sub.2(t.sub.0+t) represents CSI of a second antenna at the instant t.sub.0+t, H.sub.2*(t.sub.0+t) represents a conjugate of H.sub.2(t.sub.0+t), A.sub.s1 represents amplitude of a combined static signal of the first antenna, A.sub.s2 represents amplitude of a combined static signal of the second antenna, α.sub.d(t.sub.0) represents amplitude attenuation of a d-th dynamic signal at the instant t.sub.0, φ.sub.s1 represents phase offset of the combined static signal of the first antenna, φ.sub.s2 represents phase offset of the combined static signal of the second antenna, φ.sub.d.sup.1(t.sub.0+t) represents phase offset of a dynamic signal of the first antenna at the instant t.sub.0+t, and φ.sub.d.sup.1(t.sub.0+t) represents phase offset of a dynamic signal of the second antenna at the instant t.sub.0+t.

3. The method for non-contact determination of the sensing boundary according to claim 2, wherein the conjugate set S(t.sub.0) is:
S(t.sub.0)=[H.sub.1(t.sub.0)*H.sub.2*(t.sub.0),H.sub.1(t.sub.0+Δt)*H.sub.2*(t.sub.0+Δt), . . . H.sub.1(t.sub.0+MΔt)*H.sub.2*(t.sub.0+MΔt)] wherein [0, Δt, . . . , MΔt] are CSI sampling intervals relative to the instant t.sub.0, and Δt is the sampling time.

4. The method for non-contact determination of the sensing boundary according to claim 2, wherein a k-th element x(t.sub.0+kΔt) in the zero-mean conjugate set S(t.sub.0) is: $x\left(t_0+k\Delta t\right)=A_{s1}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j{\Psi }_{d\left(t_0+k\Delta t\right)}}+A_{s2}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j{\Omega }_d\left(t_0+k\Delta t\right)}=R\left(t_0+k\Delta t\right)+i*I\left(t_0+k\Delta t\right)$ wherein S(t.sub.0) is the zero-mean conjugate set of S(t.sub.0), Ψ.sub.d(t.sub.0+kΔt)=ω.sub.s1−φ.sub.d.sup.2(t.sub.0+kΔt), Ω.sub.d(t.sub.0+kΔt)=φ.sub.d.sup.1(t.sub.0+kΔt)−φ.sub.s2, and R(t.sub.0+kΔt) and I(t.sub.0+kΔt) are respectively the real part and imaginary part of x(t.sub.0+kΔt), where x(t.sub.0+kΔt) is a complex number.

5. The method for non-contact determination of the sensing boundary according to claim 2, wherein the boundary sensing feature σ(t.sub.0) is obtained by performing probability distribution fitting on the absolute values |x(t.sub.0+kΔt)| of all the elements in the sets, and a mode of the probability distribution fitting is Rayleigh distribution, Rice distribution, or lognormal distribution.

6. The method for non-contact determination of the sensing boundary according to claim 2, wherein in said S6, the boundary sensing feature σ(t.sub.0) is compared with a preset boundary determination threshold δ.sub.th; when σ(t.sub.0)>δ.sub.th, it is determined that the current activity is within the sensing area; and when σ(t.sub.0)<δ.sub.th, it is determined that the current activity is outside the sensing area.

7. A system for non-contact determination of a sensing boundary, comprising a receiving and measuring module, a conjugate multiplication module, a conjugate set module, a zero-mean conjugate set module, a boundary sensing feature calculation module, and a boundary determination module, wherein in the receiving and measuring module, a Wi-Fi receiving device simultaneously receives data packets by using two antennas, and measures Channel State Information (CSI); the conjugate multiplication module performs conjugate multiplication on the CSI of the two antennas at each instant, to obtain a new CSI signal stream; the conjugate set module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0); the zero-mean conjugate set module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set; the boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set module, and the zero-mean conjugate set module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0); and the boundary determination module accurately classifies, according to the boundary sensing feature σ(t.sub.0), activities inside and outside a sensing area by means of a sensing boundary formed by walls and extension lines thereof.

8. An indoor intrusion detection system based on the method for non-contact determination of the sensing boundary according to claim 1, comprising a receiving and measuring module, a conjugate multiplication module, a conjugate set construction module, a zero-mean conjugate set construction module, a boundary sensing feature calculation module, a moving and stationary detection module, and a feature-fusion intrusion determination module, wherein in the receiving and measuring module, a Wi-Fi receiving device simultaneously receives data packets by means of two antennas, and measures Channel State Information(CSI); the conjugate multiplication module performs conjugate multiplication on the CSI of the two antennas at each instant, to obtain a new CSI signal stream; the conjugate set construction module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0); the zero-mean conjugate set construction module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set; the boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set construction module, and the zero-mean conjugate set construction module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0); the moving and stationary detection module extracts a feature Dmotion(t.sub.0) for detecting a moving/stationary state of a target from the new CSI signal stream obtained after CSI conjugate multiplication/division transformation or an original CSI signal; and the feature-fusion intrusion determination module determines whether there is an intruder indoors by combining the boundary sensing feature σ(t.sub.0) and the moving and stationary detection feature Dmotion(t.sub.0).

9. The indoor intrusion detection system according to claim 8, wherein in the feature-fusion intrusion determination module, a method for determining whether there is an intruder indoors is as follows: when σ(t.sub.0)−σ(t.sub.−1)>ρ.sub.th1, it indicates that an intruder intrudes into a room, wherein σ(t.sub.−1) represents the magnitude of a boundary sensing feature at a previous instant of t.sub.0, and ρ.sub.th1 is a threshold for measuring the magnitude of a rise jump; when σ(t.sub.0)−σ(t.sub.−1)<ρ.sub.th2 and Dmotion(t.sub.0)>μ.sub.th, wherein ρ.sub.th2 is a threshold for measuring the magnitude of a fall jump and μ.sub.th is a threshold for moving and stationary determination, it indicates that the intruder leaves the room; and when ρ.sub.th2<σ(t.sub.0)−σ(t.sub.−1)<n.sub.th1, a moving/stationary state of the intruder is determined by combining the moving and stationary detection feature, that is, when Dmotion(t.sub.0)<μ.sub.th, it indicates that the detected target is in a static state indoors, and when Dmotion(t.sub.0)>μ.sub.th, it indicates that the intruder is in a moving state indoors.

10. An area detection system based on the method for non-contact determination of the sensing boundary according to claim 1, comprising a receiving and measuring module, a conjugate multiplication module, a conjugate set construction module, a zero-mean conjugate set construction module, a boundary sensing feature calculation module, an area code table building module, a global area code obtaining module, and an area detection module, wherein in the receiving and measuring module, a Wi-Fi receiving device simultaneously receives data packets by using two antennas, and measures Channel State Information(CSI); the conjugate multiplication module performs conjugate multiplication on the CSI of the two antennas at each instant and, to obtain a new CSI signal stream; the conjugate set construction module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0); the zero-mean conjugate set construction module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set; the boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set construction module, and the zero-mean conjugate set construction module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0); the area code table building module obtains an area code table according to an area structure defined by walls and states of Wi-Fi receiving devices pre-deployed in different areas, as a reference for online area detection; the global area code obtaining module compares the boundary sensing feature σ.sub.d(t.sub.0) of each device with a threshold value δ.sub.th, to obtain a current global area code; and the area detection module queries the built area code table for a current global area code, to identify the target area.

11. The area detection system according to claim 10, wherein in the area code table building module, a method for building the code table is as follows: when a target moves in an area A, states of all Wi-Fi devices in the area A are set to 1, states of Wi-Fi devices in other areas are set to 0, and accordingly, 0/1 code situations of states of different devices in a case where the target moves in all areas are obtained, to form the area code table.

12. The area detection system according to claim 10, wherein in the global area code obtaining module, a method for obtaining a global area code is as follows: when σ.sub.d(t.sub.0)>δ.sub.th, a current state of the device is set to 1, when σ.sub.d(t.sub.0)<δ.sub.th, the current state of the device is set to 0, and finally, states of different devices are combined to obtain the current global area code, wherein σ.sub.d(t.sub.0) is a boundary sensing feature of a d-th device for distinguishing between activities inside and outside walls.

13. The area detection system according to claim 10, wherein in the area detection module, a method for identifying a target current area is as follows: when the global area code can be found in the area code table, the target current area is updated to a corresponding area in the area code table; and when the global area code cannot be found in the area code table, the target is assumed to be in the same area of the previous instant.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG. 1 is a schematic diagram of Wi-Fi signal transmission; and

[0028] FIG. 2 is a block diagram of an example implementation process of a method for determining a sensing boundary provided by the present application.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0029] The present application provides a method for non-contact determination of a sensing boundary. In an ordinary home environment, the partition of a sensing area is generally at a room-level, and rooms are usually separated from each other by walls. In this method, by capturing differences in Wi-Fi signals respectively corresponding to activities of a moving target outside and inside walls, the walls in the daily environment can be necessarily used to partition the sensing area. Accordingly, in the present application, a boundary sensing feature extracted from a CSI conjugate multiplied signal of two antennas of a Wi-Fi receiving device reflects the boundary sensing feature.

[0030] the reflected signal strength of the moving target, effectively distinguishing between a direct dynamic signal (not blocked) and an indirect dynamic signal (blocked by a wall) generated by reflection of the moving target, and thereby determining an accurate sensing boundary by means of the walls in the daily environment and extension lines thereof. The present application is described in detail below with reference to the drawings and embodiments.

[0031] A transmitting terminal of the present invention is a Wi-Fi signal transmitter in a Wi-Fi transceiving environment, and a receiving terminal is a Wi-Fi signal receiver corresponding to the transmitting terminal. A natural wall in the environment assists in determining a sensing boundary, and a detected target does not need to wear any device. The sensing boundary is determined by means of Wi-Fi transceiver devices and the walls in the environment, by a method including the following steps:

[0032] S1. A Wi-Fi receiving device simultaneously receives data packets by means of two antennas, and measures Channel State Information (CSI).

[0033] The data packets are transmitted by a Wi-Fi transmitting device using one antenna. The Wi-Fi transmitting device can be a conventional commercial Wi-Fi device provided with one or more antennas. The Wi-Fi receiving device can be a conventional commercial Wi-Fi device provided with two or more antennas. The Wi-Fi transmitting device can be any device that includes a Wi-Fi network card, such as a mobile phone, a tablet, a computer, and a smart watch. The Wi-Fi receiving device and the Wi-Fi transmitting device adopt the same frequency, and common commercial devices generally support connection to three antennas, so the requirement of using one antenna at the transmitting terminal and two antennas at the receiving terminal can be met.

[0034] The Channel State Information (CSI) is used to describe changes in the amplitude and phase of a signal after being transmitted via a wireless channel. For example, on a commercial Wi-Fi device, the CSI reflects amplitude attenuation and phase drift of a wireless signal transmitted from a transmitting terminal to a receiving terminal via a wireless channel. At each sampling instant t.sub.0+t, each receiving antenna collects the respective Channel State Information, and the Channel State Information of one antenna is expressed as follows:

H(t.sub.0+t)=e.sup.jθ.sup.offset(A.sub.se.sup.jφ.sup.s+Σ.sub.d.sup.Dα.sub.d(t.sub.0+t))e.sup.jφ.sup.d.sup.(t.sup.0.sup.+t)  (1)

[0035] In the formula, θ.sub.offset represents a phase error of a phase of the CSI resulting from clock asynchronism between the Wi-Fi transmitting terminal and receiving terminal; A.sub.se.sup.jφ.sup.s represents a static signal generated by the superposition of a direct path in the environment and reflection from a static object (furniture, a ceiling, etc.) (A.sub.s represents amplitude of the static signal, and φ.sub.s represents phase shift of the static signal); Σ.sub.d.sup.Dα.sub.d(t.sub.0+t))e.sup.jφ.sup.d.sup.(t.sup.0.sup.+t) represents the superposition of all dynamic signals generated by reflection from a moving target, wherein when the moving target moves inside walls (in a sensing area), these signals are direct dynamic signals that are not blocked by the walls, and when the moving target moves outside the walls (outside the sensing area), these signals are blocked by the walls and thus become indirect dynamic signals; t.sub.0 represents the initial instant, and t represents a time period; α.sub.d(t.sub.0+t) represents amplitude attenuation of a d-th dynamic signal at the instant t.sub.0+t, φ.sub.d(t.sub.0+t) represents phase shift of the d-th dynamic signal, and D represents the total number of dynamic paths.

[0036] S2. Conjugate multiplication is performed on the CSI of the two antennas at each instant, to eliminate a phase error and obtain a new CSI signal stream.

[0037] In formula 1, due to the clock asynchronism between the Wi-Fi transmitting terminal and receiving terminal, the phase error θ.sub.offset is brought to the phase of the CSI, and the phase error changes with the time, so a boundary sensing feature cannot be directly extracted from the CSI. Since phase errors between different antennas on the same network card are the same, the phase error resulting from the asynchronism between the receiving and transmitting devices can be eliminated by means of conjugate multiplication on the CSI measured at the same sampling time by two antennas. The new CSI obtained after the conjugate multiplication is expressed as follows:

[00003]$\begin{array}{cc}{H}_{conj}=H_1\left(t_0+t\right)*H_2^{*}\left(t_0+t\right)\approx A_{s1}{A}_{s2}{e}^{j\left({\phi }_{s1}-{\phi }_{\mathrm{s2}}\right)}+A_{s1}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j\left({\phi }_{s1}-{\phi }_d^2\left(t_0+t\right)\right)}+A_{s2}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j\left({\phi }_d^1\left(t_0+t\right)-{\phi }_{s2}\right)}& \left(2\right)\end{array}$

[0038] In the formula, H.sub.conj represents the new CSI obtained after the conjugate multiplication, H.sub.1(t.sub.0+t) represents CSI of a first antenna measured at the instant t.sub.0+t, H.sub.2(t.sub.0+t) represents CSI of a second antenna measured at the instant t.sub.0+t, H.sub.2*(t.sub.0+t) represents a conjugate of H.sub.2(t.sub.0+, A.sub.s1 represents amplitude of the merged static signal of the first antenna, A.sub.se represents amplitude of the merged static signal of the second antenna, α.sub.d(t.sub.0) represents amplitude attenuation of a d-th dynamic signal at the instant t.sub.0, φ.sub.s1 represents phase offset of the merged static signal of the first antenna, φ.sub.s2 represents phase offset of the merged static signal of the second antenna, φ.sub.d.sup.1(t.sub.0+t) represents phase offset of a dynamic signal of the first antenna at the instant t.sub.0+t, and φ.sub.d.sup.2(t.sub.0+t) represents phase offset of a dynamic signal of the second antenna at the instant t.sub.0+t.

[0039] The new CSI signal H.sub.conj obtained after the conjugate multiplication eliminates the phase error, and on the other hand, enhances the energy attenuation of a target reflected signal due to wall blocking, thereby amplifying a difference between a direct dynamic signal and an indirect dynamic signal.

[0040] S3. A time window with the length M is selected from the new CSI signal stream at the instant t.sub.0, and M CSI samples in the time window are collected, to form a conjugate set S(t.sub.0):

S(t.sub.0)=[H.sub.1(t.sub.0)*H.sub.2*(t.sub.0),H.sub.1(t.sub.0+Δt)*H.sub.2*(t.sub.0+Δt), . . . H.sub.1(t.sub.0+MΔt)*H.sub.2*(t.sub.0+MΔt)]  (3)

[0041] wherein [0, Δt, . . . , MΔt] are CSI sampling intervals relative to the instant t.sub.0, and Δt is an interval sampling time.

[0042] S4. The mean of S(t.sub.0) is subtracted from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set, wherein x(t.sub.0+kΔt) is a k-th element in the zero-mean conjugate set obtained after the mean subtraction:

[00004]$\begin{array}{cc}x\left(t_0+k\Delta t\right)\in S\left(t_0\right)-\stackrel{\_}{S\left(t_0\right)}x\left(t_0+k\Delta t\right)=A_{s1}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j{\Psi }_d\left(t_0+k\Delta t\right)}+A_{s2}\underset{d}{\overset{D}{.Math.}}{\alpha }_d\left(t_0\right)e^{j{\Omega }_d\left(t_0+k\Delta t\right)}=R\left(t_0+k\Delta t\right)+i*I\left(t_0+k\Delta t\right)& \left(3\right)\end{array}$

[0043] wherein, S(t.sub.0) is the mean of the conjugate set S(t.sub.0), Ψ.sub.d(t.sub.0+kΔt)=ω.sub.s1−φ.sub.d.sup.2(t.sub.0+kΔt), Ω.sub.d(t.sub.0+kΔt)=φ.sub.d.sup.1(t.sub.0+kΔt)−φ.sub.s2, R(t.sub.0+kΔt) and I(t.sub.0+kΔt) are respectively the real part and imaginary part of x(t.sub.0+kΔt) when x(t.sub.0+kΔt) is a complex number, and i represents a symbol of the imaginary part of the complex number.

[0044] S5. Steps S2-S4 are repeated for CSI on N sub-carriers simultaneously transmitted, to obtain N zero-mean conjugate sets; and probability distribution fitting is performed on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0).

[0045] A commercial Wi-Fi device performs communication transmission in the manner of Orthogonal Frequency Division Multiplexing (OFDM), which can support simultaneous transmission of a plurality of sub-carriers. Therefore, the CSI conjugate signal streams of the N sub-carriers can be obtained by performing conjugate multiplication on each corresponding sub-carrier of the two antennas, and the N zero-mean conjugate sets are obtained by repeating steps S2-S4 for each sub-carrier.

[0046] The boundary sensing feature σ(t.sub.0) is obtained by performing probability distribution fitting on the absolute values |x(t.sub.0+kΔt)| of all the elements in the sets, and a mode of the probability distribution fitting can be Rayleigh distribution, Rice distribution, or lognormal distribution. In this embodiment, the Rayleigh distribution is preferably used to obtain the boundary sensing feature σ(t.sub.0):

[00005]$\begin{array}{cc}\sigma \left(t_0\right)=\sqrt{\frac{1}{2M*N}\underset{f=1}{\overset{N}{.Math.}}\underset{k=1}{\overset{M}{.Math.}}{.Math.x_f\left(t_0+k\Delta t\right).Math.}^2}& \left(4\right)\end{array}$

[0047] In the formula, f is a serial number of a subcarrier.

[0048] S6. According to the boundary sensing feature σ(t.sub.0), activities inside and outside a sensing area are accurately classified by means of a sensing boundary formed by walls and extension lines thereof: the boundary sensing feature σ(t.sub.0) is compared with a preset boundary determination threshold δ.sub.th; if σ(t.sub.0)>δ.sub.th, it is determined that a current activity is within the sensing area (within the walls); and if σ(t.sub.0)<δ.sub.th, it is determined that the current activity is outside the sensing area (outside the walls) (as shown in FIG. 1), thereby the activities inside and outside the sensing area are accurately classified by means of the sensing boundary formed by the walls and the extension lines thereof.

Embodiment 1

[0049] In the embodiment based on the present application, a daily Wi-Fi device is used as a receiving terminal (such as a mobile phone, a router, a computer, and a television, and in this embodiment, a minicomputer is taken as an example, but it is not limited to the minicomputer), and two antennas are used to receive a signal. A common Wi-Fi device (such as a mobile phone, a router, a computer, and a television, in this embodiment, a router is taken as an example, but it is not limited to the router) is used as a signal transmitting terminal, and one antenna is used to transmit a signal. The method for determining a sensing boundary provided by the present application is as follows:

[0050] 1) System Building

[0051] A router is used as a Wi-Fi transmitting device, and a minicomputer equipped with a commercial Wi-Fi network card supporting two antenna interfaces is used as a receiving device, wherein the two antenna interfaces are respectively connected to two antennas. In an example of the present application, an Intel 5300 network card is used to build the system, and the Wi-Fi transmitting and receiving devices work at a frequency of 5 GHz and communicate by means of a bandwidth of 20 MHz. This example consists of one transmitting terminal, one receiving terminal, and one server.

[0052] A flow of the method of determining a sensing boundary by using the built system is shown in FIG. 2, and includes the following steps:

[0053] 2) The Wi-Fi signal transmitting device transmits data packets by means of 30 sub-carriers at a rate of 200 Hz per second; two antennas of the Wi-Fi receiving device simultaneously receive the data packets on the plurality of sub-carriers, and measure CSI of each data packet transmitted by each sub-carrier.

[0054] 3) For each sub-carrier of the 30 sub-carriers, conjugate multiplication is performed on CSI measured by the two antennas of the receiving device at the same moment, to eliminate a phase error and obtain CSI conjugate signal streams of the 30 sub-carriers.

[0055] 4) Sliding processing is performed on the CSI conjugate signal streams of the 30 sub-carriers by using a sliding window of the size of 400, with a sliding step of 10.

[0056] 5) The mean of samples in a current window is subtracted from each of 400 CSI conjugate signal samples in the sliding window for each subcarrier.

[0057] 6) The CSI conjugate signal samples of all the 30 sub-carriers are fused, to obtain a total of 30*400=12000 samples, and probability distribution fitting is performed on the absolute values of all the samples, to obtain a boundary sensing feature σ(t.sub.0), wherein a mode of the probability distribution fitting can be Rayleigh distribution, Rice distribution, or lognormal distribution. Preferably, a formula for fitting by using the Rayleigh distribution is as follows:

[00006]$\sigma \left(t_0\right)=\sqrt{\frac{1}{2*12000}\underset{f=1}{\overset{30}{.Math.}}\underset{k=1}{\overset{400}{.Math.}}{.Math.x_f\left(t_0+\Delta t_k\right).Math.}^2}$

[0058] 7) The receiving device transmits the calculated boundary sensing feature in each window to a designated port on a server in real time by means of socket connection.

[0059] 8) The server monitors the designated port in real time and compares σ(t.sub.0) with a boundary sensing threshold δ.sub.th=40 once receiving σ(t.sub.0), wherein if σ(t.sub.0)>40, it indicates that a target moves in a sensing area within walls, thereby determining the sensing boundary of the area where the target is located by means of the walls and extension lines thereof.

[0060] The present application further provides a system for non-contact determination of a sensing boundary, including a receiving and measuring module, a conjugate multiplication module, a conjugate set construction module, a zero-mean conjugate set construction module, a boundary sensing feature calculation module, and a boundary determination module.

[0061] In the receiving and measuring module, a Wi-Fi receiving device simultaneously receives data packets by two antennas, and measures Channel State Information (CSI).

[0062] The conjugate multiplication module performs conjugate multiplication on the CSI of the two antennas at each instant, to obtain a new CSI signal stream.

[0063] The conjugate set construction module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0).

[0064] The zero-mean conjugate set construction module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set.

[0065] The boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set module, and the zero-mean conjugate set module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0).

[0066] The boundary determination module accurately classifies, according to the boundary sensing feature σ(t.sub.0), activities inside and outside a sensing area by means of the boundary formed by walls and extension lines thereof.

Embodiment 2

[0067] The present application further provides an application based on the method for non-contact determination of the sensing boundary of the present application, which is specifically an indoor intrusion detection system based on the method for non-contact determination of the sensing boundary. The system is built on the basis of the above system for non-contact determination of the sensing boundary of the present application and includes a receiving and measuring module, a conjugate multiplication module, a conjugate set construction module, a zero-mean conjugate set construction module, a boundary sensing feature calculation module, a moving and stationary detection module, and a feature-fusion and intrusion determination module.

[0068] In the receiving and measuring module, a Wi-Fi receiving device simultaneously receives data packets by two antennas, and measures Channel State Information.

[0069] The conjugate multiplication module performs conjugate multiplication on the CSI of the two antennas at each instant, to obtain a new CSI signal stream.

[0070] The conjugate set construction module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0).

[0071] The zero-mean conjugate set construction module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set.

[0072] The boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set module, and the zero-mean conjugate set module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0).

[0073] The moving and stationary detection module extracts a feature Dmotion(t.sub.0) for detecting a moving/stationary state of an active target from the new CSI signal stream obtained after CSI conjugate multiplication/division transformation or an original CSI signal. Preferably, a Doppler energy feature Dmotion(t.sub.0) is extracted by means of a DopplerMUSIC algorithm, or an amplitude autocorrelation feature Dmotion(t.sub.0) is extracted by means of a CSI amplitude autocorrelation algorithm.

[0074] The feature-fusion and intrusion determination module determines whether there is an intruder indoors by combining the boundary sensing feature σ(t.sub.0) and the moving and stationary detection feature Dmotion(t.sub.0). Specifically, if the boundary sensing feature undergoes an obvious jump, that is, if σ(t.sub.0)−σ(t.sub.−1)>ρ.sub.th1(σ(t.sub.−1) represents the magnitude of a boundary sensing feature at a previous instant of t.sub.0 and ρ.sub.th1 is a threshold for measuring the magnitude of a rise jump), it indicates that an intruder intrudes into a room; if σ(t.sub.0)−σ(t.sub.−1)<ρ.sub.th2 and Dmotion(t.sub.0)>μ.sub.th (ρ.sub.th2 is a threshold for measuring the magnitude of a fall jump and μ.sub.th is a threshold for moving and stationary determination), it indicates that the intruder leaves the room; and if ρ.sub.th2<σ(t.sub.0)−σ(t.sub.−1)<ρ.sub.th1, a moving/stationary state of the intruder is determined by combining the moving and stationary detection feature, that is, if Dmotion(t.sub.0)<μ.sub.th, it indicates that the detected target is in a static state indoors, and if Dmotion(t.sub.0)>μ.sub.th, it indicates that the intruder is in an active state indoors.

Embodiment 3

[0075] The present application further provides an application based on the method for non-contact determination of the sensing boundary of the present application, which is specifically a non-contact indoor area detection system implemented by means of a sensing boundary determined by using a plurality of devices. The system is built on the basis of the above system for non-contact determination of the sensing boundary of the present application and includes, a receiving and measuring module, a conjugate multiplication module, a conjugate set construction module, a zero-mean conjugate set construction module, a boundary sensing feature calculation module, an area code table building module, a global area code obtaining module, and an area detection module.

[0076] The area code table building module obtains an area code table according to an area structure defined by walls and states of the Wi-Fi receiving devices pre-deployed in different areas. A method for building the area code table is as follows: a unique label is configured for each device in all areas, when a detected target moves in an area A, states of all Wi-Fi devices (RX 1) in the area A are set to 1, states of Wi-Fi devices (RX 2, RX 3 . . . RX4) in other areas are set to 0, and accordingly, 0/1 code situations of states of different devices in a case where the target moves in all areas are obtained by combining the states of all the devices, to form the area code table (as shown in Table 1), which is used as a reference for online area detection.

TABLE-US-00001 TABLE 1 Area code table RX 1 RX2 RX 3 RX 4 Area A 1 0 0 0 Area B 0 1 0 0 Area C 0 0 1 0 Area D 0 0 0 1

[0077] In the receiving and measuring module, a Wi-Fi transmitting device transmits data packets by using one antenna; and one or more receiving devices in different areas receive the data packets from the transmitting terminal at the same time by using two antennas, and measure CSI information therefrom.

[0078] The conjugate multiplication module performs conjugate multiplication on the CSI measured of the two antennas at each instant, to obtain a new CSI signal stream.

[0079] The conjugate set construction module selects a time window with the length M from the new CSI signal stream at the instant t.sub.0, and collects M CSI samples in the time window, to form a conjugate set S(t.sub.0).

[0080] The zero-mean conjugate set construction module subtracts the mean of S(t.sub.0) from all the samples in the conjugate set S(t.sub.0), to obtain a zero-mean conjugate set.

[0081] The boundary sensing feature calculation module obtains, by means of the conjugate multiplication module, the conjugate set module, and the zero-mean conjugate set module, N zero-mean conjugate sets for CSI on N sub-carriers simultaneously transmitted, and performs probability distribution fitting on the absolute values of all elements in the N zero-mean conjugate sets, to obtain a boundary sensing feature σ(t.sub.0).

[0082] The global area code obtaining module compares the boundary sensing feature σ.sub.d(t.sub.0) of each device with a threshold value δ.sub.th, to obtain a current global area code. Specifically, if σ.sub.d(t.sub.0)>δ.sub.th, it indicates that a target moves in an area where the current device is located (movement/activity inside walls), so a state of the device is set to 1; if σ.sub.d(t.sub.0)<δ.sub.th, it indicates that the target moves outside the area where the current device is located (movement outside walls), so the current state of the device is set to 0; and finally, states of different devices are combined to obtain a current global area code.

[0083] The area detection module queries the area code table for a current global area code, to identify the target area. Specifically, if the global area code can be found in the area code table, the target current area is updated to a corresponding area in the area code table; and if the global area code cannot be found in the area code table, the target is assumed to be in the same area of the previous instant.

[0084] The present application further provides a computer device including a memory, a processor, and a computer program stored on the memory and executable by the processor, wherein the method for non-contact determination of the sensing boundary in the above embodiments is implemented when the processor executes the computer program.

[0085] The present application further provides a computer-readable storage medium with a computer program stored thereon, wherein the method for non-contact determination of the sensing boundary in the above embodiments is implemented when a processor executes the program.

[0086] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application may adopt a form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to a disk storage, a CD-ROM, an optical storage, etc.) containing computer-usable program codes.

[0087] The present application is described with reference to the flowcharts and block diagrams of the method, system, and computer program product according to the embodiments of the present application. It should be understood that each process and block in the flowcharts and block diagrams, and the combination of the process and block in the flowcharts and block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or a processor of other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in a flowchart and one or more blocks in a block diagram.

[0088] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction apparatus. The instruction apparatus implements the functions specified in one or more processes in a flowchart and one or more blocks in a block diagram.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, to execute a series of operation steps on the computer or other programmable devices to produce computer-implemented processing, so that the instructions executed on a computer or other programmable devices provide steps for implementing the functions specified in one or more processes in a flowchart and one or more blocks in a block diagram.

[0090] The above embodiments of the present application are not used to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application fall within the scope of the pending claims of the present application.