CS positioning and deformation monitoring method for underground tunnels based on millimeter wave sensing assistance

Abstract

The present invention discloses a method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance, including: receiving millimeter wave orthogonal frequency division multiplexing signals from a transmitter based on an intelligent reflection unit; receiving the reflected signal of the intelligent reflection unit through a receiver and using a super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflection unit, and determining the rough position of the intelligent reflection unit according to a triangular relationship; using a compressed sensing tool to refine the angle according to the signal departure angle and arrival angle provided by the rough position of the intelligent reflection unit, and obtaining a refined departure angle and arrival angle estimate; using a triangular relationship, finally solving the position information of the intelligent reflection unit, and uploading the position information to a database, updating the position information at intervals, and judging the deformation of the tunnel according to the comparison of the previous and next position information. The present invention can achieve millimeter-level positioning accuracy while having a low computational complexity, and can effectively monitor subtle tunnel changes.

Description

Technical Field

[0001] The present invention belongs to the field of tunnel monitoring, and in particular relates to an underground tunnel CS positioning and deformation monitoring method based on millimeter wave sensing assistance. Background Art

[0002] With the increasing complexity of the electromagnetic environment and the increasing number of large-scale, complex and diverse access devices, the perception needs of local positioning systems, especially in underground tunnels, for ultra-reliable, low-latency and ultra-high-density connections have become an important problem that needs to be solved in the future development of positioning. The Internet of Things has been recognized by academia and industry as an important part of future wireless networks. Applications such as smart industry, smart transportation, and smart health will become rapidly growing areas in future wireless networks; with the explosive growth of smart devices and ultra-large-scale connections, the Internet of Things faces great challenges in perception. Positioning technology supported by interawareness integration is expected to become a key technology to solve the perception problem of future wireless networks. Applying backscatter positioning technology to the Internet of Things is a major driving force for promoting positioning-related research.

[0003] In real-world application scenarios, the millimeter-wave multiple-input multiple-output wireless communication system can provide finer spatial resolution due to its high frequency characteristics. The signal can be received and spatial signal processing technology can be used to calculate the signal's arrival angle and delay information to achieve rough positioning. Given the sparse characteristics of the millimeter-wave channel, this feature can be used to refine the angle within a certain angle range of the target in combination with the rough positioning result to obtain the angle refinement result, thereby achieving the positioning of the backscatterer. Summary of the invention

[0004] The main content of the present invention is to propose a method for CS positioning and deformation monitoring of underground tunnels based on millimeter wave sensing assistance, which can realize deformation monitoring of underground tunnels by positioning the scatterer of intelligent reflection unit.

[0005] In order to solve the above technical problems, the present invention provides a method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance, comprising:

[0006] Based on the intelligent reflection unit installed on the wall of the tunnel, the millimeter wave orthogonal frequency division multiplexing signal from the transmitter is received and transmitted to the receiver;

[0007] The receiver receives the millimeter wave orthogonal frequency division multiplexing signal reflected by the intelligent reflection unit and uses a super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflection unit, and determines the rough position of the intelligent reflection unit according to the trigonometric relationship;

[0008] According to the signal departure angle and arrival angle provided by the rough position of the intelligent reflector unit, a compressed sensing tool is used to perform angle refinement to obtain refined departure angle and arrival angle estimation;

[0009] By using the trigonometric relationship, the position information of the intelligent reflection unit is finally solved and uploaded to the database. The position information is updated at intervals, and the deformation of the tunnel is determined based on the comparison of the previous and next position information.

[0010] Preferably, the millimeter wave orthogonal frequency division multiplexing signal reflected by the intelligent reflection unit and received by the receiver is expressed as:

[0011] Define the coordinate as P T (x t ,y t ) T N t Antenna transmitter, coordinates P R (x r ,y r ) T N r Antenna receiving device, L single-antenna intelligent reflection unit backscattering devices and environmental scatterers. Assuming that the direct link and the environmental scatterer reflection link are known, the OFDM millimeter wave signal received at the receiving end is expressed as:

[0012] y q [k]=h[k]F q [k]x q [k]+n q [k]

[0013] where k represents the kth subcarrier, F[k] is an arbitrary transmit beamforming matrix, and x q [k] is the qth transmitted signal, h[k] is the discrete frequency domain channel representation, n q [k] means the mean is 0 and the variance is N 0 Gaussian noise.

[0014] Preferably, the process of using a super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflector unit and determining the rough position of the intelligent reflector unit according to the triangular relationship includes:

[0015] Receiving the reflected signal of the intelligent reflection unit by the receiver and using the super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflection unit to obtain the arrival angle and delay information of the reflected signal;

[0016] Obtaining the coordinates of the reflection device of the intelligent reflection unit according to the trigonometric relationship between the arrival angle and delay information of the reflection signal and the known transceiver position coordinates;

[0017] The reflection device coordinates are used to solve the departure angle of the transmitted signal. According to the obtained target departure angle and arrival angle, the angle information is refined using a compressed sensing tool to regain the precise coordinates of the reflection device.

[0018] Preferably, the process of estimating the angle of arrival and delay of the intelligent reflection unit using a super-resolution algorithm includes:

[0019] The continuously transmitted time domain signal is expressed as:

[0020]

[0021] Among them, M t is the number of OFDM symbols, K is the number of subcarriers, Δf is the subcarrier spacing, f c is the center carrier frequency, Rect(t / T) is a rectangular pulse window function with a duration of T; for the qth signal, when M t OFDM symbols, the qth transmitted discrete time signal is:

[0022]

[0023] Without considering the direct link and the reflection link of the environmental scatterer, the time domain channel is expressed as:

[0024]

[0025] Among them, θ l and are the departure angle and arrival angle of the lth path respectively, is the direction-steering vector, assuming the antenna spacing is d 1 , d 1 =λ / 2, wavelength is λ=c / f c , L is the number of intelligent reflection units, α l is the channel coefficient;

[0026] After sampling, the time domain channel is transformed into the frequency domain through K-point FFT transformation. The frequency domain channel at the kth subcarrier is expressed as:

[0027] h[k]=A Rx,Lχ [k]A Tx,L T

[0028] Among them, A Rx,L is the receiving direction steering matrix, X Tx,L is the transmission direction steering matrix, χ[k]=diag[γ k,l ],Bundle Rect k,l , N cpis the number of channel taps, which is also equal to the number of cyclic prefixes. Note as γ k,l [k];

[0029] Channel coefficient α l It also includes the path loss, radar cross-section area and delay information of the lth intelligent reflector unit, then in is the path loss, and h is the channel gain.

[0030] Preferably, the process of obtaining the arrival angle of the reflected signal includes:

[0031] According to the signal model, channel model and received signal model, when sending M t When there is an OFDM signal, the discrete frequency domain channel is represented as:

[0032]

[0033] where k,m∈[0,1,...,L],[0,1,...,M t ], β m,k =diag(β m,k,l ), β m,k Contains other channel information besides angle information:

[0034]

[0035] R l and f l is the distance and Doppler information of the lth intelligent reflection unit, G l represents the attenuation factor related to the path loss and radar cross section of the lth smart reflector unit;

[0036] The received signal in the frequency domain is expressed as:

[0037]

[0038] F (:,m) is the mth column of the beamforming matrix F, Indicates that the mean is 0 and the variance is N 0 Gaussian noise vector;

[0039] By definition The received signal in the frequency domain is again expressed as:

[0040]

[0041] By using M t OFDM symbols and K subcarriers are stacked, and the resulting signal is represented as:

[0042] in

[0043] Preferably, the process of obtaining the arrival angle of the reflected signal further includes:

[0044] Perform eigenvalue decomposition based on the covariance matrix of the received signal to obtain the eigenvalues ​​of the correlation matrix;

[0045] Obtaining a noise subspace matrix according to the number L of intelligent reflection units, and obtaining a peak value of a spatial spectrum function based on the noise subspace of the noise subspace matrix as the arrival angle information;

[0046] Wherein, the covariance matrix of the received signal is expressed as:

[0047]

[0048] The eigenvalues ​​and corresponding eigenvectors of the correlation matrix are expressed as:

[0049] [U y ,Σ y ]=eig(R y )

[0050] U y and∑ y is an orthogonal eigenvalue matrix and a diagonal matrix with eigenvalues ​​in descending order;

[0051] The noise subspace matrix U N It is expressed as:

[0052] U N =U y [:,L+1:N r ]

[0053] The spatial spectrum function is expressed as:

[0054]

[0055] Preferably, the process of obtaining the reflected signal delay information includes:

[0056] Based on the arrival angle of the reflected signal, the received signal at the kth subcarrier of the mth OFDM symbol is represented by the receive beamforming vector:

[0057] Right now

[0058] is the receive beamforming vector, which is generated by low-complexity least squares according to the arrival angle information, that is, where cR is any complex value with modulus 1;

[0059] Will Bring in

[0060] Get the new received signal:

[0061]

[0062] in The mean is still 0 and the variance is N 0 Gaussian noise;

[0063] The representation of Doppler and delay steering vectors along subcarriers and symbols is defined as follows:

[0064]

[0065] Ignore Doppler information when locating the target object, and use the delay steering vector to obtain the multi-target delay steering matrix, that is, A R,K =[a R,K (R 1 ),a R,K (R 2 ),......,a R,K (R L )];

[0066] Then all OFDM symbols with K subcarriers are stacked together to obtain:

[0067]

[0068] Preferably, the process of obtaining the reflected signal delay information further includes:

[0069] Removing the sent communication pilot data from the received signal to obtain channel information, and calculating the channel information based on a super-resolution algorithm to obtain a delay estimate;

[0070] The process of calculating the channel information based on the super-resolution algorithm to obtain the delay estimation includes:

[0071] Perform eigendecomposition based on the covariance matrix of the channel information to obtain the eigenvalues ​​of the correlation matrix;

[0072] A noise subspace matrix is ​​obtained according to the number L of intelligent reflection units, and a peak value of a spatial spectrum function is obtained based on the noise subspace of the noise subspace matrix, which is the delay information;

[0073] Wherein, the covariance matrix of the channel information is expressed as:

[0074]

[0075] The eigenvalues ​​and corresponding eigenvectors of the covariance matrix are expressed as:

[0076] [U τ ,Σ τ ]=eig(R τ )

[0077] The spatial spectrum function is expressed as:

[0078]

[0079] Preferably, the roughly estimated coordinates of the intelligent reflection unit are expressed as:

[0080] Based on the arrival angle θ 1 and delay information τ 1 , transmitter location coordinates (x t ,y t ), receiving device location coordinates (x r ,y r ), the rough estimated coordinates of the intelligent reflection unit are expressed as:

[0081]

[0082] Where c is the speed of light, The distance information between the transceiver and the receiver;

[0083] The departure angle of the signal transmitted by the transmitter is obtained based on the known transceiver coordinates and the roughly estimated coordinates. for:

[0084]

[0085] Preferably, the precise coordinates of the intelligent reflection unit are expressed as:

[0086] Assume that the transmitter location coordinates (x t ,y t ), receiving device location coordinates (x r ,y r ), the angle of arrival is θ 1 ′ and departure angle Then the precise coordinates of the intelligent reflection unit are obtained as

[0087]

[0088] Compared with the prior art, the present invention has the following advantages and technical effects:

[0089] The present invention places the intelligent reflection unit on the wall of the underground tunnel, calculates the arrival angle and delay information of the signal by using the spatial signal processing technology to obtain a rough positioning, and then obtains the positioning of the intelligent reflection unit with refined angle within a certain angle range of the target based on the sparse characteristics of the millimeter wave signal channel and the rough positioning result. The present invention is suitable for mines, has a simple implementation plan and significant practical value, and more importantly, can achieve millimeter-level positioning accuracy while having a low computational complexity, and can effectively monitor subtle tunnel changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0091] Figure 1 A schematic diagram of a system model of an embodiment of the present invention;

[0092] Figure 2 A schematic diagram of a signal processing process according to an embodiment of the present invention;

[0093] Figure 3 A schematic diagram of a road deformation monitoring process according to an embodiment of the present invention;

[0094] Figure 4 This is a schematic diagram of the relationship between the mean square error value of the angle of arrival obtained by spatial domain signal processing and the angle of departure and angle of arrival obtained by compressed sensing and the link signal-to-noise ratio in an embodiment of the present invention. DETAILED DESCRIPTION

[0095] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0096] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0097] Embodiment 1

[0098] like Figure 1-4 As shown, this embodiment provides a method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance, including:

[0099] Based on the intelligent reflection unit installed on the wall of the tunnel, the millimeter wave orthogonal frequency division multiplexing signal from the transmitter is received and transmitted to the receiver;

[0100] The reflected signal of the intelligent reflection unit is received by the receiver and the arrival angle and delay of the intelligent reflection unit are estimated by using a super-resolution algorithm, and the rough position of the intelligent reflection unit is determined according to the trigonometric relationship;

[0101] Based on the signal departure angle and arrival angle provided by the rough position of the intelligent reflector unit, the compressed sensing tool is used to refine the angle to obtain refined departure angle and arrival angle estimates;

[0102] By using the trigonometric relationship, the position information of the intelligent reflection unit is finally solved and uploaded to the database. The position information is updated at intervals, and the deformation of the tunnel is judged based on the comparison of the previous and subsequent position information.

[0103] Furthermore, the arrival angle and delay of the intelligent reflector unit are estimated by using a super-resolution algorithm, and the process of determining the rough position of the intelligent reflector unit according to the triangular relationship includes:

[0104] The reflected signal of the intelligent reflection unit is received by the receiver and the arrival angle and delay information of the intelligent reflection unit are obtained by estimating the arrival angle and delay of the intelligent reflection unit using a super-resolution algorithm;

[0105] Obtaining the coordinates of the reflection device of the intelligent reflection unit according to the trigonometric relationship between the arrival angle and delay information of the reflection signal and the known transceiver position coordinates;

[0106] The departure angle of the transmitted signal is solved using the coordinates of the reflecting device. Based on the obtained target departure angle and arrival angle, the angle information is refined using a compressed sensing tool to regain the precise coordinates of the reflecting device.

[0107] Furthermore, if Figure 1 As shown, the positioning system of the present invention takes into account the N t The antenna transmitter, whose coordinates are P T (x t ,y t ) T 、N r Antenna receiving device, its coordinates are P R (x r ,y r ) TA millimeter wave underground tunnel deformation system consisting of L single-antenna intelligent reflection unit backscattering devices and some environmental scatterers; in the tunnel, the intelligent reflection unit with a single antenna tag is installed on the tunnel wall, and the intelligent reflection unit reflection device transmits information to the receiver by receiving the millimeter wave orthogonal frequency division multiplexing signal from the transmitter, and the receiver jointly estimates the position of all reflection devices. It is assumed that the transceiver shares the necessary signal structure and pilot information, and due to the severe attenuation of the millimeter wave signal, only a single reflection process is considered. It is considered that in addition to the reflection of the intelligent reflection unit reflection device, the reflection of other scatterers in the environment can also be received at the receiver, and due to the severe attenuation of the millimeter wave signal, only a single reflection process is considered, which is called the environmental link signal. Under the premise of perfect estimation of the environmental link channel, the receiver receives the intelligent reflection unit reflection signal, and the arrival angle and delay information of the reflection signal can be obtained by using signal processing technology. According to the arrival angle, delay and known position coordinates of the transceiver of the backscattering signal, the coordinates of the backscattering device of the intelligent reflection unit can be roughly obtained. The departure angle of the transmitted signal is solved using the coordinates of the reflector. Based on the obtained departure angle and arrival angle of the target, the angle information is refined using the compressed sensing tool to obtain the precise coordinates of the reflector. The deformation of the lane is determined based on the position information at different times.

[0108] Figure 1 The direct signal and the reflected signal of the environmental scatterer in the channel can be used to estimate the direct link and the reflected link of the environmental scatterer in advance. It is assumed that the environmental reflection link including the direct link is perfectly known, so only the Figure 1 The intelligent reflection unit reflection link in the system is used to locate the object carrying the intelligent reflection unit.

[0109] Assuming that the direct link and the environmental scatterer reflection link are known, the OFDM millimeter wave signal received at the receiving end is expressed as:

[0110] y q [k]=h[k]F q [k]x q [k]+n q [k]

[0111] where k represents the kth subcarrier, F[k] is an arbitrary transmit beamforming matrix, and x q [k] is the qth transmitted signal, h[k] is the discrete frequency domain channel representation, n q [k] means the mean is 0 and the variance is N 0 Gaussian noise.

[0112] For sending continuous time domain signals, it can be expressed as:

[0113]

[0114] Among them, M t is the number of OFDM symbols, K is the number of subcarriers, Δf is the subcarrier spacing, f c is the center carrier frequency, Rect(t / T) is a rectangular pulse window function with a duration of T. For the qth signal, when M t OFDM symbols, the qth transmitted discrete time signal is:

[0115]

[0116] Without considering the direct link and the reflection link of the environmental scatterer, the time domain channel is expressed as:

[0117]

[0118] Among them, θ l and are the departure angle and arrival angle of the lth path respectively, is the direction-steering vector, assuming the antenna spacing is d 1 , d 1 =λ / 2, wavelength is λ=c / f c , L is the number of intelligent reflection units, α l is the channel coefficient. After sampling, the time domain channel can be transformed from the time domain to the frequency domain through K-point FFT transformation.

[0119] The frequency domain channel at the kth subcarrier is expressed as: h[k] = A Rx,L χ[k]A Tx,L T , where A Rx,L is the receiving direction steering matrix, A Tx,L is the transmission direction steering matrix, χ[k]=diag[γ k,l ],Bundle Recorded as Rect k,l , N cp is the number of channel taps, which is also equal to the number of cyclic prefixes. Note as γ k,l [k]. Note that the channel coefficient α l It also includes the path loss, radar cross-section area and delay information of the lth intelligent reflector unit, so in is the path loss, and h is the channel gain.

[0120] According to the signal model, channel model and received signal model, when sending M t When there is an OFDM signal, the discrete frequency domain channel can be re-expressed as:

[0121]

[0122] where k,m∈[0,1,...,K],[0,1,...,M t ], β m,k =diag(β m,k,l ), β m,k Contains other channel information besides angle information:

[0123]

[0124] R l and f l is the distance and Doppler information of the lth intelligent reflection unit, G l Represents the attenuation factor related to the path loss and radar cross section of the lth smart reflector unit.

[0125] Therefore, the received signal in the frequency domain can be expressed as: F (:,m) is the mth column of the beamforming matrix F, Indicates that the mean is 0 and the variance is N 0 Gaussian noise vector.

[0126] By definition The received signal in the frequency domain can again be expressed as:

[0127]

[0128] By using M t OFDM symbols and K subcarriers are stacked, and the resulting signal is represented as: in

[0129] The arrival angle information and delay information can be obtained by using the super-resolution algorithm for the stacked received signals. The specific steps for obtaining the arrival angle information are as follows:

[0130] The covariance matrix of the received signal is: Then the eigenvalue decomposition of the correlation matrix can be obtained.

[0131] Then we can get the eigenvalue decomposition of the correlation matrix: y ,∑ y ]=eig(R y ), where U y and∑ y is an orthogonal eigenvalue matrix and a diagonal matrix with the eigenvalues ​​in descending order.

[0132] Since we already know that the number of smart reflection units is L, we can get the noise subspace matrix U N , U N =U y [:,L+1:N r ].

[0133] Finally, the spatial spectrum function is obtained according to the noise subspace The peak value is the desired angle of arrival information.

[0134] Only the arrival angle information is not enough to obtain the positioning result, so it is necessary to solve the delay information, that is, the distance information, of the received signal through beamforming. The specific detailed steps are as follows:

[0135] After obtaining the arrival angle, the received signal at the kth subcarrier of the mth OFDM symbol is expressed by the receive beamforming vector as:

[0136]

[0137] Right now

[0138] here is the receive beamforming vector, which is generated by low-complexity least squares according to the arrival angle information, that is, Among them C R is any complex value modulo 1.

[0139] Will Bring in

[0140] Get the new received signal:

[0141]

[0142] in The mean is still 0 and the variance is N 0 Gaussian noise.

[0143] It can be seen that Doppler and delay exist as independent complex exponential functions with separate indices along subcarriers and symbols. Therefore, the representation of Doppler and delay steering vectors along subcarriers and symbols is defined as follows:

[0144]

[0145]

[0146] The present invention locates the target object, thus ignoring the Doppler information, and using the delay steering vector to obtain the multi-target delay steering matrix, namely A R,K =[a R,K(R 1 ),a R,K (R 2 ),......,a R,K (R L )].

[0147] Then, similar to estimating the arrival angle before, by stacking all OFDM symbols with K subcarriers together, we get:

[0148] The transmitted communication pilot data is removed from the received signal to obtain a representation of the channel information. Similar to the arrival angle information estimation, the delay estimation can be obtained based on the super-resolution algorithm for the obtained channel information.

[0149] The covariance matrix of the channel information is: Decompose its features;

[0150] Perform eigendecomposition on the covariance matrix, that is, [U τ ,∑ τ ]=eig(R τ ), follow the angle information solving steps, and finally get the spatial spectrum function of the time delay information: The peak value is the target delay estimate.

[0151] Next, take a single target as an example to solve the target coordinate information, and use the arrival angle θ obtained in the above implementation process 1 and delay information τ 1 , transmitter location coordinates (x t ,y t ), receiving device location coordinates (x r ,y r ), the rough estimated coordinates of the intelligent reflection unit are obtained as

[0152]

[0153] Where c is the speed of light, The distance information between the transceiver and the receiver;

[0154] Since the coordinates of the transceiver are known and the rough coordinates of the target are estimated, the angle of departure of the signal transmitted by the transmitter is obtained. for:

[0155]

[0156] The above is the process of using the super-resolution algorithm to solve the coordinates of the intelligent reflector unit for the received millimeter-wave OFDM signal. However, in order to achieve a more accurate positioning purpose and to obtain centimeter positioning accuracy, the compressed sensing tool is used on the basis of the above process. As we all know, the millimeter-wave channel has a sparse characteristic in the beam domain, and this property can be used to obtain more accurate departure angle and arrival angle information of the signal. On the other hand, since we have already obtained a rough estimate of the departure angle and the arrival angle, when using the compressed sensing tool, the measurement matrix does not need to be divided within the entire angle range of 180 degrees, but only needs to be divided into finer grids within a certain angle range of the roughly estimated target. The use of compressed sensing tools not only reduces the overall computational complexity, but also makes the parameter estimation results more ideal, that is, the positioning effect is better. The specific implementation plan is:

[0157] According to the received signal model, when K subcarriers are sent, assuming that K subcarriers represent pilot information, the received signal of the K subcarrier pilot information is:

[0158] In the refinement phase, the received signal Perform vectorization operations and use the results of the previous positioning phase as the initial estimate of the angle. The SWOMP algorithm assumes that there are L paths for initialization, and the initial departure angle and arrival angle are:

[0159] The process of the initial coarse positioning aiding the SWOMP algorithm to estimate the refined departure angle and arrival angle in the present invention is as follows.

[0160] The inputs of the SWOMP algorithm are: received signal Y, transmitted signal X, initial arrival angle and departure angle and Angle division step d θ and And the initialization angle variation error σ θ , .

[0161] Angle refinement process:

[0162] Retrieve the direction-steering matrix based on the refined angle information:

[0163] According to the SWOMP algorithm, the refined angle information is obtained:

[0164] Finally, unlike the super-resolution algorithm positioning method, which uses the obtained arrival angle and delay information for positioning, the precise positioning is based on the refined departure angle and arrival angle. The precise coordinates of the final target are estimated to be

[0165]

[0166] According to the above description, if Figure 3 That is the overall implementation process diagram of the present invention, and finally the deformation situation of the tunnel is obtained.

[0167] like Figure 4 As shown in the figure, when the center frequency is set to 70GHz, the subcarrier spacing is 120KHz, the number of subcarriers and symbols is 256 and 1 respectively, the number of transmitting antennas is equal to the number of receiving antennas, that is, the performance simulation results of 32 antennas, through several Monte Carlo angle estimation mean square errors under different signal-to-noise ratios, according to the simulation results, it can be seen that the angle estimation error in the proposed positioning algorithm is less affected by the signal-to-noise ratio and has a higher estimation accuracy. Compared with other direct use of compressed sensing algorithms, it has a good estimation effect under low signal-to-noise ratios. For example, when the departure angle and arrival angle estimation error are 0.01 degrees, the positioning error can be at the millimeter level. This positioning effect is particularly suitable for scenarios such as monitoring of small wall changes in mines.

[0168] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance, characterized in that: include: Based on the intelligent reflection unit installed on the wall of the tunnel, the millimeter wave orthogonal frequency division multiplexing signal from the transmitter is received and transmitted to the receiver; Receiving the reflected signal of the intelligent reflection unit through the receiver and using the super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflection unit, and determining the rough position of the intelligent reflection unit according to the trigonometric relationship; According to the signal departure angle and arrival angle provided by the rough position of the intelligent reflector unit, a compressed sensing tool is used to refine the angle to obtain a refined departure angle and arrival angle estimation; Using the trigonometric relationship, the position information of the intelligent reflection unit is finally solved, and the position information is uploaded to the database, the position information is updated at intervals, and the deformation of the roadway is judged based on the comparison of the previous and next position information; The process of using the super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflector unit and determining the rough position of the intelligent reflector unit based on the triangular relationship includes: Receiving the reflected signal of the intelligent reflection unit by the receiver and using the super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflection unit to obtain the arrival angle and delay information of the reflected signal; Obtaining the coordinates of the reflection device of the intelligent reflection unit according to the trigonometric relationship between the arrival angle and delay information of the reflection signal and the known transceiver position coordinates; The departure angle of the transmitted signal is solved by using the coordinates of the reflection device, and according to the obtained target departure angle and arrival angle, the angle information is refined by using a compressed sensing tool to obtain the precise coordinates of the reflection device again; The process of using the super-resolution algorithm to estimate the arrival angle and delay of the intelligent reflector unit includes: The continuously transmitted time domain signal is expressed as: Among them, M t is the number of OFDM symbols, K is the number of subcarriers, Δf is the subcarrier spacing, f c is the center carrier frequency, Rect(t / T) is a rectangular pulse window function with a duration of T; for the qth signal, when M t OFDM symbols, the qth transmitted discrete time signal is: Without considering the direct link and the reflected link of the environmental scatterer, the time domain channel is expressed as: Among them, θ l and are the departure angle and arrival angle of the lth path respectively, is the direction-steering vector, assuming that the antenna spacing is d1, d1 = λ / 2, and the wavelength is λ = c / f c , L is the number of intelligent reflection units, α l is the channel coefficient; After sampling, the time domain channel is transformed into the frequency domain through K-point FFT transformation. The frequency domain channel at the kth subcarrier is expressed as: h[k]=A Rx,L x[k]A Tx,L T Among them, A Rx,L is the receiving direction steering matrix, A Tx,L is the transmission direction steering matrix, χ[k]=diag[γ k,l ],Bundle Rect k,l , N cp is the number of channel taps, which is also equal to the number of cyclic prefixes. Note as γ k,l [k]; Channel coefficient α l It also includes the path loss, radar cross-section area and delay information of the lth intelligent reflector unit, then in is the path loss, Indicates the channel gain.

2. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The millimeter-wave orthogonal frequency division multiplexing signal reflected by the intelligent reflection unit received by the receiver is expressed as: Define the coordinate as P T (x t ,y t ) T N t Antenna transmitter, coordinates P R (x r ,y r ) T N r Antenna receiving device, L single-antenna intelligent reflection unit backscattering devices and environmental scatterers. Assuming that the direct link and the environmental scatterer reflection link are known, the OFDM millimeter wave signal received at the receiving end is expressed as: y q [k]=h[k]F q [k]x q [k]+n q [k] where k represents the kth subcarrier, F[k] is an arbitrary transmit beamforming matrix, and x q [k] is the qth transmitted signal, h[k] is the discrete frequency domain channel representation, n q [k] represents Gaussian noise with mean 0 and variance N0.

3. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The process of obtaining the arrival angle of the reflected signal includes: According to the signal model, channel model and received signal model, when sending M t When there are OFDM signals, the discrete frequency domain channel is expressed as: where k,m∈[0,1,...,K],[0,1,...,M t ], β m,k =diag(β m,k,l ), β m,k Indicates that it contains other channel information besides angle information: R l and f l is the distance and Doppler information of the lth intelligent reflection unit, G l represents the attenuation factor related to the path loss and radar cross section of the lth smart reflector unit; The received signal in the frequency domain is expressed as: F (:,m) is the mth column of the beamforming matrix F, represents a Gaussian noise vector with a mean of 0 and a variance of N0; By definition The received signal in the frequency domain is again expressed as: By using M t OFDM symbols and K subcarriers are stacked, and the resulting signal is represented as: in 4. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The process of obtaining the arrival angle of the reflected signal also includes: Perform eigenvalue decomposition based on the covariance matrix of the received signal to obtain the eigenvalues ​​of the correlation matrix; Obtaining a noise subspace matrix according to the number L of intelligent reflection units, and obtaining a peak value of a spatial spectrum function based on the noise subspace of the noise subspace matrix as the arrival angle information; Wherein, the covariance matrix of the received signal is expressed as: The eigenvalues ​​and corresponding eigenvectors of the correlation matrix are expressed as: [U y ,∑ y ]=eig(R y ) U y and∑ y is an orthogonal eigenvalue matrix and a diagonal matrix with eigenvalues ​​in descending order; The noise subspace matrix U N It is expressed as: U N =U y [:,L+1:N r ] The spatial spectrum function is expressed as:

5. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The process of obtaining the reflected signal delay information includes: Based on the arrival angle of the reflected signal, the received signal at the kth subcarrier of the mth OFDM symbol is represented by the receive beamforming vector: Right now is the receive beamforming vector, which is generated by low-complexity least squares according to the arrival angle information, that is, where c R is any complex value with modulus 1; Will Substitution Get the new received signal: in It is still Gaussian noise with mean 0 and variance N0; The representation of Doppler and delay steering vectors along subcarriers and symbols is defined as follows: Ignore Doppler information when locating the target object, and use the delay steering vector to obtain the multi-target delay steering matrix, that is, A R,K =[a R,K (R1),a R,K (R2),......,a R,K (R L )]; Then all OFDM symbols with K subcarriers are stacked together to obtain:

6. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The process of obtaining the reflected signal delay information also includes: Removing the sent communication pilot data from the received signal to obtain channel information, and calculating the channel information based on a super-resolution algorithm to obtain a delay estimate; The process of calculating the channel information based on the super-resolution algorithm to obtain the delay estimation includes: Perform eigendecomposition based on the covariance matrix of the channel information to obtain the eigenvalues ​​of the correlation matrix; A noise subspace matrix is ​​obtained according to the number L of intelligent reflection units, and a peak value of a spatial spectrum function is obtained based on the noise subspace of the noise subspace matrix, which is the delay information; Wherein, the covariance matrix of the channel information is expressed as: The eigenvalues ​​and corresponding eigenvectors of the covariance matrix are expressed as: [U τ ,Σ τ ]=eig(R τ ) The spatial spectrum function is expressed as:

7. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The rough estimated coordinates of the intelligent reflection unit are expressed as: Based on the arrival angle θ1 and the delay information τ1, the transmitter position coordinates (x t ,y t ), receiving device location coordinates (x r ,y r ), the rough estimated coordinates of the intelligent reflection unit are expressed as: Where c is the speed of light, The distance information between the transceiver and the receiver; The departure angle of the signal transmitted by the transmitter is obtained based on the known transceiver coordinates and the roughly estimated coordinates. for:

8. The method for underground tunnel CS positioning and deformation monitoring based on millimeter wave sensing assistance according to claim 1 is characterized in that: The precise coordinates of the intelligent reflection unit are expressed as: Assume that the transmitter location coordinates (x t ,y t ), receiving device location coordinates (x r ,y r ), the arrival angle is θ′1, and the departure angle is Then the precise coordinates of the intelligent reflection unit are obtained as

Images