MIMO through-the-wall radar imaging method based on propagation compensation and coherence factor
Through the MIMO wall-through radar imaging method based on propagation compensation and coherence factors, the problem of gate lobe side lobe and signal weakening is solved, and fast and efficient wall-through radar imaging is achieved, which is suitable for multiple transmission and multi-collection and single-static arrays.
Patent Information
- Application Number
- CN202510670058.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-23
- Publication Date
- 2025-08-22
AI Technical Summary
The existing wall-through radar imaging methods have gate lobe side lobe problems in multiple transmission and multiple collection systems, the signal is weakened and the calculation complexity is high, making it difficult to process quickly, and the deep learning methods rely on a large amount of labeled data and have high cost and poor generalization of the scene.
Using MIMO wall-through radar imaging method based on propagation compensation and coherence factors, we use data reconstruction, dimensionality reduction, wavenumber domain processing and weighted imaging, and use coherence factors to deduce parallel calculations to correct signal attenuation and refractive errors to achieve rapid imaging.
Effectively suppress the side lobe of the gate and enhance the target energy. It is suitable for MIMO and single static arrays, maintaining fast solution characteristics and improving imaging quality.
Smart Images
Figure CN120522702A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of radar signal processing, and in particular relates to through-wall radar imaging of indoor stationary targets. Background Art
[0002] As a nondestructive detection device, through-the-wall radar (TWR) has significant practical value in specialized operational environments, such as disaster emergency rescue and counter-terrorism tactical raids. Its core technology relies on the strong penetration of low-frequency electromagnetic waves, using signal processing techniques to extract the real-time position coordinates, morphological characteristics, vital sign parameters, and internal building structural data of targets obscured by walls. For multi-input multi-output (MIMO) through-the-wall radars, a sparse array design is typically used to reduce system complexity. However, this produces grating lobes and sidelobes that reduce imaging quality, and it is also necessary to overcome signal weakening caused by wall attenuation.
[0003] Conventional through-wall radar imaging methods, such as the backprojection algorithm, have problems such as high grating lobe sidelobes and need to be combined with grating lobe suppression methods such as the coherence factor method, but the calculation speed is slow; although sparse reconstruction methods such as compressed sensing suppress sidelobes through optimization models, the sparsity assumption of the echo signal fails in dense multi-target scenarios, and the computational complexity is high, making it difficult to process quickly; although deep learning methods can adaptively extract features, they rely on a large amount of labeled data, and the acquisition cost of through-wall radar measured data is high and the scene generalization is poor, which restricts its engineering application. Summary of the Invention
[0004] In response to the above problems, the present invention proposes a MIMO through-wall radar imaging method based on propagation compensation and coherent factor (CF). The present invention addresses the problems of target signal attenuation, imaging accompanied by grating lobes and side lobes, and low imaging efficiency. First, based on the data reconstructed from the coherence factor, the echo data is deformed. The reconstructed data can be calculated in parallel with the original echo data to obtain the incoherent superposition part of the coherence factor; then, based on the equivalent phase center, the MIMO array echo data is reduced to a single static array echo data; based on the scalar diffraction theory, a specific propagation compensation value is quantitatively derived to compensate for the fixed free space propagation attenuation when the non-fixed wall attenuation is difficult to obtain, completing the wavenumber domain processing; a wall compensation factor in the wavenumber domain is designed to directly correct the additional time delay introduced by through-wall refraction in the wavenumber domain; finally, the incoherent superposition components and coherent superposition components obtained by parallel calculation are used to complete weighted imaging to obtain the final imaging result.
[0005] Figure 1This is a signal processing flow chart of an embodiment of the present invention. To achieve the purpose of the present invention, the technical solution adopted is: a MIMO through-wall radar imaging method based on propagation compensation and coherence factor, comprising the following steps:
[0006] Step 1: Acquire echo data and preprocess:
[0007] Obtain the frequency domain data of the echo received by the through-wall radar system and save it in the form of a three-dimensional matrix, with frequency dimension K and transmitting antenna dimension N. T , receiving antenna dimension N R The dimension is K×N T ×N R . Remove wall clutter from the echo through preprocessing.
[0008] Step 2: Reconstruct echo data:
[0009] like Figure 2 As shown, the reconstructed echo data G(k s ):
[0010] G(k s )=∑ i-j＝K-1-n s(k 1(i) )s * (k 2(j) ),n=0,...,2K-2,k s =k s min +nΔk,
[0011] where k s is the wave number of the reconstructed echo data, k is the wave number of the original echo data, s(k 1(i) ) and s(k 2(j) ) are two independent representations of the original echo data s(k), the second subscripts represent the (i+1)th wavenumber and the (j+1)th wavenumber i, j=0,...,K-1, and Δk is the wavenumber step.
[0012] Step 3: Reduce the dimension of echo data:
[0013] Taking the original echo as an example, calculate the relative transmit and receive array element (x T ,y T ,0),(x R ,y R ,0)The reference phase term of the scattered echo of the target point (x,y,z)
[0014]
[0015] in
[0016] Calculate the equivalent phase center coordinates (xc ,y c ,0)
[0017]
[0018] Get the reference phase term about the equivalent phase center
[0019]
[0020] in
[0021] Based on the equivalent phase center principle, the MIMO array echo is compared with the two calculated phase terms to obtain the echo of the equivalent phase center:
[0022]
[0023] Since different transmitting and receiving antennas may correspond to the same equivalent phase center, the echo of the equivalent phase center Accumulate the echo data at the same position to obtain the echo data with the same coordinates in three-dimensional form That is, the MIMO data is reduced in dimension and converted into single static uniform array echo data:
[0024]
[0025] in is the equivalent phase center echo, with the same dimension as the original echo s(x T ,y T ,x R ,y R ,k) consistent, It is a single static uniform array echo obtained after dimensionality reduction, and its dimension is only related to the number of points and frequency of the equivalent phase center in the x and y directions.
[0026] Similarly, the dimensionality reduction equivalent echo after data reconstruction can be obtained as
[0027] Step 4: Get the wall compensation factor:
[0028] Based on the estimated wall thickness D and relative dielectric constant ε r , the wall compensation factor in the wave number domain is obtained as
[0029]
[0030] where k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
[0031] Step 5: Complete distance compensation:
[0032] Based on the scalar diffraction model, the echo signal is equivalently transformed into the wave number domain. After multiplying the wall compensation factor in the wave number domain, it is converted back to the spatial domain through Fourier transform. The expression is as follows
[0033]
[0034] Among them, σ0(x,y,z) is the coherent superposition result, k c is the central wave number, λ is the wavelength, z'=z-Z0, Z0 is the vertical distance from the array to the wall, F is the Fourier transform, F 2D and is the two-dimensional Fourier transform and its inverse transform, s * (x c ,y c ,k) is the original echo data s(x c ,y c ,k) conjugate, k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
[0035] Similarly, the incoherent superposition result I can be obtained by using the echo data reconstructed and reduced in dimension. CF (x,y,z) is
[0036] Among them G * (x c ,y c ,k s ) is the dimensionality reduction equivalent echo G(x c ,y c ,k s ) conjugate, k sx ,k sy ,k sz is the wave number k s Components in the x, y, and z directions.
[0037] Step 6: Weighted Imaging:
[0038] Based on the coherent superposition and incoherent superposition results, CF is calculated.
[0039]
[0040] The weighted imaging is completed using the CF weight, and the final imaging result is
[0041] σ f (x,y,z)=σ0(x,y,z)⊙CF(x,y,z),
[0042] where ⊙ is the Hadamard product.
[0043] At this point, a MIMO through-wall radar imaging method based on propagation compensation and coherence factor has been completed.
[0044] Beneficial effects:
[0045] The present invention is applied to through-wall radar imaging processing. By systematically considering factors such as enhancing target echo energy and suppressing grating lobe sidelobes, the method achieves good imaging effects. It is applicable to array formats such as MIMO and monostatic, and is an effective and practical through-wall radar imaging method. Specifically, it includes:
[0046] 1. The present invention systematically achieves energy enhancement of remote targets, suppression of grating lobe sidelobes, and correction of refraction errors;
[0047] 2. The present invention maintains the characteristic of rapid solution;
[0048] 3. The present invention is applicable to MIMO arrays and monostatic arrays. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 is a signal processing flow chart of an embodiment of the present invention;
[0050] Figure 2 It is a data reconstruction schematic diagram of the present invention;
[0051] Figure 3 It is a schematic diagram of the actual measurement verification scenario of the present invention;
[0052] Figure 4 This is a schematic diagram of the array structure for actual measurement verification of the present invention;
[0053] Figure 5 It is the imaging result of BP directly after preprocessing;
[0054] Figure 6 It is the step-by-step process flow and intermediate results of the present invention;
[0055] Figure 7 This is the imaging result of the MIMO through-wall radar proposed in the present invention. DETAILED DESCRIPTION
[0056] The present invention is described in detail below with reference to the accompanying drawings and embodiments.
[0057] Step 1: Acquire echo data and preprocess:
[0058] Obtain the frequency domain data of the echo received by the through-wall radar system and save it in the form of a three-dimensional matrix, with frequency dimension K and transmitting antenna dimension N. T , receiving antenna dimension N RThe dimension is K×N T ×N R . Remove wall clutter from the echo through preprocessing.
[0059] Obtain frequency domain data of the echo received by the through-wall radar system and save it as a three-dimensional matrix with a frequency dimension of 256, a transmit antenna dimension of 10, and a receive antenna dimension of 6, resulting in a dimension of 256 × 10 × 6. Remove wall clutter from the echo by background cancellation.
[0060] Step 2: Reconstruct echo data:
[0061] like Figure 2 As shown, the reconstructed echo data G(k s ):
[0062] G(k s )=∑ i-j＝K-1-n s(k 1(i) )s * (k 2(j) ),n=0,...,2K-2,k s =k s min +nΔk,
[0063] where k s is the wave number of the reconstructed echo data, k is the wave number of the original echo data, s(k 1(i) ) and s(k 2(j) ) are two independent representations of the original echo data s(k), the second subscripts represent the (i+1)th wavenumber and the (j+1)th wavenumber i, j=0,...,K-1, and Δk is the wavenumber step.
[0064] Calculate the reconstructed echo data G(k s ), with dimensions of 511×10×6.
[0065] Step 3: Reduce the dimension of echo data:
[0066] Taking the original echo as an example, calculate the relative transmit and receive array element (x T ,y T ,0),(x R ,y R ,0)The reference phase term of the scattered echo of the target point (x,y,z)
[0067]
[0068] in
[0069] Calculate the equivalent phase center coordinates (x c ,y c ,0)
[0070]
[0071] Get the reference phase term about the equivalent phase center
[0072]
[0073] in
[0074] Based on the equivalent phase center principle, the MIMO array echo is compared with the two calculated phase terms to obtain the echo of the equivalent phase center:
[0075]
[0076] Since different transmitting and receiving antennas may correspond to the same equivalent phase center, the echo of the equivalent phase center Accumulate the echo data at the same position to obtain the echo data with the same coordinates in three-dimensional form That is, the MIMO data is reduced in dimension and converted into single static uniform array echo data:
[0077]
[0078] Similarly, the dimensionality reduction equivalent echo after data reconstruction can be obtained as
[0079] The compensation is completed by calculating the phase term based on the coordinates of the transmitting and receiving array elements and the equivalent phase center. Since the number of transmitting and receiving antennas in this measurement is small, the original echo data after dimensionality reduction and the equivalent echo after data reconstruction are The dimensions remain unchanged, namely 256×10×6 and 511×10×6, thus converting the MIMO data into single static uniform array echo data, and directly completing the intermediate results of frequency domain imaging. Figure 6 shown.
[0080] Step 4: Get the wall compensation factor:
[0081] Based on the estimated wall thickness D and relative dielectric constant ε r , the wall compensation factor in the wave number domain is obtained as
[0082]
[0083] where k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
[0084] Based on the estimated wall thickness of 20 cm and relative dielectric constant of 4, the wall compensation factor in the wavenumber domain is obtained. The intermediate result after the wall compensation factor is applied is as follows: Figure 6 shown.
[0085] Step 5: Complete distance compensation:
[0086] Based on the scalar diffraction model, the echo signal is equivalently transformed into the wave number domain. After multiplying the wall compensation factor in the wave number domain, it is converted back to the spatial domain through Fourier transform. The expression is as follows
[0087]
[0088] Among them, σ0(x,y,z) is the coherent superposition result, k c is the central wave number, λ is the wavelength, z'=z-Z0, Z0 is the vertical distance from the array to the wall, F is the Fourier transform, F 2D and is the two-dimensional Fourier transform and its inverse transform, s * (x c ,y c ,k) is the original echo data s(x c ,y c ,k) conjugate, k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
[0089] Similarly, the incoherent superposition result I can be obtained by using the echo data reconstructed and reduced in dimension. CF (x,y,z) is
[0090] Among them G * (x c ,y c ,k s ) is the dimensionality reduction equivalent echo G(x c ,y c ,k s ) conjugate, k sx ,k sy ,k sz is the wave number k s Components in the x, y, and z directions.
[0091] Based on the scalar diffraction model, the echo signal is equivalently transformed into the wave number domain. After multiplying the wall compensation factor in the wave number domain, it is converted back to the spatial domain through Fourier transform to obtain the coherent superposition result σ0(x, y, z) and the incoherent superposition result I CF (x,y,z), the intermediate result after distance compensation is as follows Figure 6 shown.
[0092] Step 6: Weighted Imaging:
[0093] Based on the coherent superposition and incoherent superposition results, CF is calculated.
[0094]
[0095] The weighted imaging is completed using the CF weight, and the final imaging result is
[0096] σ f (x,y,z)=σ0(x,y,z)⊙CF(x,y,z),
[0097] where ⊙ is the Hadamard product.
[0098] The CF weights are calculated to complete weighted imaging. The final imaging result is as follows: Figure 7 shown.
[0099] At this point, a MIMO through-wall radar imaging method based on propagation compensation and coherence factor has been completed.
[0100] Example
[0101] To verify the proposed MIMO through-wall radar imaging method based on propagation compensation and coherence factor, a field experiment was designed for analysis. The field experiment parameters are shown in Table 1.
[0102] Table 1 Measured experimental parameter settings
[0103]
[0104]
[0105] Construct a typical through-wall radar application scenario such as Figure 3 As shown, the detection targets are two diagonally placed corners, and the wall to be penetrated is a 20cm brick wall. The actual measurement radar is a commercial MIMO through-wall radar, and its schematic diagram is as follows Figure 4 As shown, the through-wall radar data collection can be completed well.
[0106] For the frequency domain data received by the radar, back projection (BP) imaging is performed directly after preprocessing, and the results are as follows: Figure 5 As shown in the figure, the horizontal axis represents azimuth, and the vertical axis represents range. In the imaging results, in addition to the target, there are also strong sidelobes and grating lobes. At the same time, after preprocessing, wall clutter also remains, which is not conducive to subsequent target detection and recognition.
[0107] After applying the through-wall radar imaging method proposed by the present invention, the results are as follows: Figure 7Compared to BP imaging, image quality is significantly improved. Areas previously characterized by grating sidelobes are suppressed, while targets further away are enhanced, effectively reducing the probability of missed detection. Targets are clearly visible and accurately positioned in the image, effectively enabling MIMO through-wall radar target imaging.
[0108] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A MIMO through-wall radar imaging method based on propagation compensation and coherence factor, characterized in that The steps of the method include: Step 1: Acquire echo data and pre-process to suppress wall clutter; Step 2: Reconstruct the echo data to calculate the incoherent superposition part; Step 3: Reduce the dimensionality of the echo data, converting the echo data from a matrix with the same transmitting and receiving antennas and wave numbers to a matrix with the same uniform coordinates and wave numbers; Step 4: Obtain the wall compensation factor and complete the refraction correction in the wavenumber domain; Step 5: Complete distance compensation, enhance target energy, and obtain coherent superposition and incoherent superposition parts; Step 6: Complete CF calculation and weighted imaging to obtain the final imaging result.
2. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 1, the frequency domain data of the echo received by the through-wall radar system is obtained and saved in the form of a three-dimensional matrix, and the wall clutter in the echo is removed by preprocessing.
3. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 2, the reconstructed echo data G(k s ): G(k s )=∑ i-j＝K-1-n s(k 1(i) )s * (k 2(j) ),n=0,…,2K-2,k s =k smin +nΔk, where k s is the wave number of the reconstructed echo data, k is the wave number of the original echo data, s(k 1(i) ) and s(k 2(j) ) are two independent representations of the original echo data s(k), the second subscripts represent the (i+1)th wavenumber and the (j+1)th wavenumber i, j=0,...,K-1, and Δk is the wavenumber step.
4. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 3, the echo at the equivalent phase center is: Where s(x T ,y T ,x R ,y R ,k) is the original echo, s ref (x T ,y T ,x R ,y R ,k) is the reference phase term of the original echo, is the reference phase term about the equivalent phase center, and the coordinates of the equivalent phase center are (x c ,y c ,0), the coordinates of the transmitting and receiving elements are (x T ,y T ,0),(x R ,y R ,0).
5. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 3, the single static uniform array echo data is: in is the equivalent phase center echo, with the same dimension as the original echo s(x T ,y T ,x R ,y R ,k) consistent, It is a single static uniform array echo obtained after dimensionality reduction, and its dimension is only related to the number of points and frequency of the equivalent phase center in the x and y directions.
6. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 3, the dimensionality-reduced equivalent echo after data reconstruction is:
7. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 4, the wall compensation factor in the wavenumber domain is: Where D is the wall thickness, ε r is the relative dielectric constant, k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
8. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 5, after multiplying the wall compensation factor in the wave number domain and then converting it back to the spatial domain through Fourier transform, the expression is as follows Among them, σ0(x,y,z) is the coherent superposition result, k c is the central wave number, λ is the wavelength, z'=z-Z0, Z0 is the vertical distance from the array to the wall, F is the Fourier transform, F 2D and is the two-dimensional Fourier transform and its inverse transform, s * (x c ,y c ,k) is the original echo data s(x c ,y c ,k) conjugate, k x ,k y ,k z are the components of wave number k in the x, y, and z directions.
9. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 5, the incoherent superposition result I CF (x,y,z) is: Among them G * (x c ,y c ,k s ) is the dimensionality reduction equivalent echo G(x c ,y c ,k s ) conjugate, k sx ,k sy ,k sz are the components of the wave number ks in the x, y, and z directions.
10. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 6, CF is calculated based on the coherent superposition and incoherent superposition results:
11. The MIMO through-wall radar imaging method based on propagation compensation and coherence factor according to claim 1, characterized in that: In step 6, weighted imaging is performed using the CF weights, and the final imaging result is: σ f (x,y,z)=σ0(x,y,z)⊙CF(x,y,z), where ⊙ is the Hadamard product.