Two-channel radar foresight imaging method
By using a single-shot and two-receive configuration and a time-frequency and space-oriented matrix construction in dual-channel radar, combined with a regularized iterative adaptive method, the problems of left/right blur and low resolution in dual-channel radar forward vision imaging are solved, and the problem of left/right blur and low resolution is achieved More efficient imaging performance.
Patent Information
- Application Number
- CN202411267808.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-11
AI Technical Summary
In forward vision imaging, dual-channel radar has problems such as poor left/right fuzzy solution performance and low imaging resolution in areas near the route, and the prior art is difficult to effectively solve with limited resources.
Using a single send and two receiving channel configuration, distance pulse compression and distance migration correction are performed by obtaining echo data, time-frequency and space-oriented matrix is constructed, and a regularized iterative adaptive method is used to solve left/right blur and realize super-resolution Imaging.
Under the same channel resource conditions, two-channel forward SAR left/right defuzzing and super-resolution imaging performance is achieved better than the existing image domain methods, improving imaging quality.
Smart Images

Figure CN118938224A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of radar imaging, and in particular relates to a dual-channel radar forward-looking imaging method. Background Art
[0002] Radar forward imaging has important applications in autonomous landing, autonomous navigation and other fields. However, conventional single-channel SAR or Doppler beam sharpening technology has a forward imaging blind spot due to Doppler symmetric blur and small Doppler variation in the forward-looking area.
[0003] In the literature "Lu Jingyue, Zhang Lei, Wang Guanyong. Forward-looking multi-channel synthetic aperture radar deambiguation imaging method. Journal of Electronics and Information Technology, 2018, 40(12): 2820-2825", the platform Doppler information is first used to obtain a forward-looking image with left / right ambiguity, and then the data of multiple channels are combined to deambiguate the left / right ambiguity. This scheme can obtain good resolution in areas deviating from the flight path. However, in the area near the platform route, its azimuth resolution is limited by the small change in viewing angle and is still poor. The document "W.Li, R.Chen, J.Yang, J.Wu, Y.Zhang, and Y.Huang, A hybrid real / synthetic aperture scheme for multichannel radar forward-looking superresolution imaging, IEEE Geoscience and Remote Sensing Letters, vol.20, pp.1-5, 2023" studies the hybrid real / synthetic aperture scheme for multichannel radar forward-looking super-resolution imaging. It can obtain high-quality forward-looking images by fusing the synthetic aperture imaging results and the real aperture super-resolution results. However, the above scheme requires a large number of array elements to realize forward-looking imaging with left and right ambiguity, which increases the complexity of the system and limits its application in certain situations where channel resources are limited.
[0004] Theoretically, the left / right ambiguity in forward-looking SAR imaging is a binary problem that can be solved with only two channels. The document "Bao Min, Zhou Peng, Shi Lin. Research on Deambiguation Algorithm for Dual-antenna Forward-looking Missile-borne SAR. Journal of Electronics and Information Technology, 2013, 35(12): 2857-2862" studied a dual-channel forward-looking SAR imaging scheme, which first used synthetic aperture processing to obtain a forward-looking image with left / right ambiguity, and then used the dual-channel path difference information and beamforming technology to resolve the left / right ambiguity. However, due to limited channel resources, the performance of solving left / right ambiguity is also limited. The document "J. Lu, L. Zhang, S. Wei, and Y. Li, Resolution enhancement for forward looking multi-channel sar imagery with exploiting space-time sparsity, IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1-17, 2023" studies the use of sparse Bayesian reconstruction methods to resolve left / right ambiguity from the image domain and improve deambiguation imaging performance. However, in practice, the left / right blur resolution performance is always affected by the focusing result. Especially in the area near the flight path, the azimuth main lobe becomes significantly wider, and the adjacent azimuth units in the image domain will be strongly interfered with each other, resulting in a decrease in the left / right blur resolution performance. Summary of the invention
[0005] In order to solve the above technical problems, the present invention provides a dual-channel radar forward-looking imaging method, which aims to overcome the problems of poor left / right fuzzy solution performance of dual-channel radar and low imaging resolution in the area near the route.
[0006] The technical solution adopted by the present invention is: a dual-channel radar forward imaging method, the specific steps are as follows:
[0007] Step 1: Acquire echo data of the area to be imaged;
[0008] The dual-channel radar forward imaging adopts a single-transmit and two-receive channel configuration, that is, one channel transmits signals and two channels receive signals. Suppose there is any point target P in the scene 0 (x 0 ,y 0 ,0)=(r 0 sinα 0 cosθ 0 ,r 0 sinα 0 sinθ 0 ,0), then its distance history to the transmitting channel is R pt The distance history R of the receiving channelpr The expressions are as follows:
[0009]
[0010] Among them, v r represents the flight speed of the platform, h represents the flight altitude of the platform, r 0 and θ 0 Represents point target P 0 Relative to the transmitting antenna T xa The slant range and azimuth, α 0 represents the angle between the target line of sight and the vertical direction, t s Indicates slow time, y k represents the azimuth coordinate of the kth receiving antenna (k=1,2).
[0011] For point target P 0 , its two-way distance history R p (t s ,y k ; r 0 ,θ 0 ) is expressed as follows:
[0012] R p (t s ,y k ; r 0 ,θ 0 )=R pt (t s ; r 0 ,θ 0 )+R pr (y k ,t s ; r 0 ,θ 0 ) (3)
[0013] Assuming the transmitted signal is a linear frequency modulation pulse, the echo signal S received by the two channels e (y k ,t r ,t s ) is expressed as follows:
[0014]
[0015] Among them, σ p represents the scattering coefficient of the target, K r represents the distance modulation frequency, c represents the speed of light, λ represents the wavelength of the transmitted signal, and t r Indicates fast time.
[0016] Step 2, performing range pulse compression on the data obtained in step 1;
[0017] Set the matching function of pulse compression to S ref (t r )=exp(-jπK r t r 2 ), then the pulse compressed signal S c (y k ,t r ,t s ) is expressed as follows:
[0018]
[0019] Among them, IFFT represents an inverse Fourier transform operator, and FFT represents a Fourier transform operator.
[0020] Step 3, performing range migration correction on the data after the range pulse compression in step 2;
[0021] First, perform the range Fourier transform FT on equation (5) to obtain the echo signal S in the range frequency domain: 1 (y k ,f r ,t s ; r 0 ,θ 0 ), the expression is as follows:
[0022]
[0023] Among them, S(f r )=FT[sinc(B r t r )],B r represents the bandwidth of the LFM pulse, f 0 Indicates the carrier frequency, f r Indicates the distance frequency.
[0024] Then select the azimuth as 0° and the slant distance as r p The point target is taken as the reference target, and its coordinates are P r (x r ,y r ,0)=(r p sinα 0 ,0,0), then the echo signal S after phase compensation 2 (y k ,f r ,t s ; r 0 ,θ 0 ) is expressed as follows:
[0025]
[0026] Then, the first-order keystone transformation is used to realize the linear range unit migration RCM correction, that is, the scale transformation of the slow time is performed, and the expression is as follows:
[0027]
[0028] Where η represents the slow time after scale transformation.
[0029] Substitute equation (8) into equation (7) and perform inverse Fourier transform to obtain the echo S after RCM: 3 (y k ,t r ,t s ; r 0 ,θ 0 ), the expression is as follows:
[0030]
[0031] Then the echo S of a range unit 4 (y k ,t s ; r 0 ,θ 0 ) is expressed as follows:
[0032]
[0033] For R in formula (10), p (t s ,y k ; r 0 ,θ 0 )-R p (t s ,y k ; r p ,0) Perform Taylor expansion and retain the first-order term. The expression is as follows:
[0034] R p (t s ,y k ; r 0 ,θ 0 )-R p (t s ,y k ; r p ,0)≈
[0035] 2(r 0 -r p )-2v r sinα 0 (1-cosθ 0 )·t s -sinα 0 sinθ0 ·y k (11)
[0036] Substituting equation (11) into equation (10), the echo S of a range unit is 4 (y k ,t s ; r 0 ,θ 0 ) is expressed as follows:
[0037]
[0038] Among them, the first term represents a constant that does not affect azimuth focusing, the second term represents the phase of the path difference related to the spatial channel, and the third term represents the phase related to the synthetic aperture.
[0039] Step 4: Based on step 3, construct the echo domain time-frequency-space steering matrix, and establish a linear mapping equation between the echo and the original scene;
[0040] According to formula (12), 2M narrowband signals are assumed, that is, M targets are symmetrically distributed on the left and right sides of the route. Then, the received signal expressions of the nth (n = 1, 2…N) sampling point in each channel are as follows:
[0041]
[0042] Among them, α l represents the pitch angle of the lth range unit, l = (1, 2, ..., L), L represents the number of sampling points in the range direction, and They represent the noise and echo of the nth sample in the kth channel (k=1,2), respectively. Indicates the nth sampling point of slow time. and denote the scattering coefficients of left and right symmetric targets, respectively, represents the azimuth angle of the mth target.
[0043] Equations (13) and (14) fully describe the Doppler variation and path difference between the target and multiple channels. By combining the results of N sampling points in each channel, they are written in matrix form. The expressions are as follows:
[0044]
[0045] Among them, (·) represents the dot product of the matrix, s k =[s k1 ,s k2 ,...,s kN ] H represents the echo matrix of the kth channel, (·) HIndicates the transposition operation of the matrix. represents the azimuth of the target, n k =[n k1 ,n k2 ,...,n kN ] H represents the noise matrix of the kth channel, and Respectively represent the scattering coefficient matrix of the left or right target in the scene. A represents the guidance matrix formed by Doppler change, and the expression is as follows:
[0046]
[0047] in,
[0048] It represents the steering matrix formed by the spatial path difference, and its expression is as follows:
[0049]
[0050] Solve equations (15) and (16) at the same time to resolve the left / right ambiguity. The expression is as follows:
[0051]
[0052] Then rewrite equation (19), that is, construct the linear mapping equation between the echo and the original scene, the expression is as follows:
[0053] Z 2N×1 =H 2N×2M ×g 2M×1 +N 2N×1 (20)
[0054] Among them, Z 2N×1 =[s 1 ,s 2 ] H , g 2M×1 =[σ p ,σ s ] H , N 2N×1 =[n 1 ,n 2 ] H , H represents the echo domain time-frequency-space steering matrix, and its expression is as follows:
[0055]
[0056] Step 5: Jointly solve the time-frequency-space model, that is, use the regularized iterative adaptive method to solve the linear mapping equation between the echo in step 4 and the original scene, and simultaneously realize the left / right ambiguity resolution and super-resolution imaging of the dual-channel forward-looking SAR;
[0057] According to formula (20), the weighted least squares cost function is defined as The expression is as follows:
[0058]
[0059] Among them, h m represents the mth column of the steering matrix H, R represents the covariance matrix, and R = HPH H , P represents a diagonal matrix, the diagonal elements Represents the target scattering coefficient.
[0060] For formula (22) Taking the partial derivative and setting the result equal to zero, we obtain:
[0061]
[0062] According to the matrix inversion lemma, we can get: Substituting it into formula (23), we can get:
[0063]
[0064] Then the diagonally loaded regularized covariance matrix is used for solution, that is, in the covariance matrix R, the diagonal matrix μI is introduced 2M×2M The solution is as follows:
[0065] A. Initialization: identity matrix I, regularization parameter μ, number of iterations T;
[0066] B. For the echo data Z, obtain the data of its lth range unit;
[0067] C. Construct the steering matrix H according to formula (20);
[0068] D、P=|H H Z| 2 / (H H H) 2 ;
[0069] E, for i = 1 to T do;
[0070] F, R = HPH H +μI;
[0071] G, for m = 1 to 2M do;
[0072] H.
[0073] I, end for;
[0074] J.
[0075] K, end for.
[0076] Finally, all distance units are traversed, the operations in steps AK are executed, and the imaging results are output to realize left / right ambiguity resolution and super-resolution imaging of dual-channel forward-looking SAR.
[0077] Beneficial effects of the present invention: The method of the present invention first acquires echo data of the area to be imaged, performs range pulse compression and range migration correction on the acquired data, constructs a steering matrix that considers time-frequency-space information, establishes a linear equation between the echo and the original scene, and then uses a regularized iterative adaptive method to solve the equation, thereby simultaneously realizing dual-channel forward-looking SAR left / right ambiguity resolution and super-resolution imaging. The method of the present invention directly realizes dual-channel forward-looking SAR left / right ambiguity resolution and super-resolution imaging in the echo domain at the same time, and under the same channel resource conditions, can obtain better imaging performance than the existing image domain method. BRIEF DESCRIPTION OF THE DRAWINGS
[0078] Figure 1 The present invention is a flow chart of a dual-channel radar forward-looking imaging method.
[0079] Figure 2 4 is a geometric model diagram of dual-channel radar forward imaging in an embodiment of the present invention.
[0080] Figure 3 Schematic diagram of an observation scene in an embodiment of the present invention.
[0081] Figure 4 Schematic diagram of a single channel synthetic aperture dimensional pulse compression echo in an embodiment of the present invention.
[0082] Figure 5 A schematic diagram of a single distance unit steering matrix design in an embodiment of the present invention.
[0083] Figure 6 It is a schematic diagram of imaging results of an observation scene using the method of the present invention in an embodiment of the present invention. DETAILED DESCRIPTION
[0084] The method of the present invention is further described below with reference to the accompanying drawings and embodiments.
[0085] like Figure 1 As shown, a flow chart of a dual-channel radar forward imaging method of the present invention, the specific steps are as follows:
[0086] Step 1: Acquire echo data of the area to be imaged;
[0087] The dual-channel radar forward imaging adopts a single-transmit and two-receive channel configuration, that is, one channel transmits signals and two channels receive signals. In this embodiment, the forward dual-channel radar geometric configuration is as follows Figure 2 The parameters of the forward-looking dual-channel radar are shown in Table 1.
[0088] Table 1
[0089] parameter symbol Numeric unit Aircraft flight altitude h 10000 m Aircraft flight speed <![CDATA[v r ]]> 70 m / s Pulse duration <![CDATA[T r ]]> 1 μs Signal bandwidth B 120 MHz Sampling rate <![CDATA[F s ]]> 144 MHz Array antenna length <![CDATA[L e ]]> 0.0365 m Pulse repetition modulation frequency PRF 1000 Hz Transmitted signal wavelength λ 0.0167 m SNR after range pulse compression SNR 20 dB Synthetic aperture length <![CDATA[L s ]]> 140 m Distance sampling points L 128 Number of channels <![CDATA[N a ]]> 2 Synthetic aperture dimension sampling points N 2000 Azimuth angle of observation scene division M 512
[0090] like Figure 2 As shown, in an xyz space coordinate system, O represents the origin of the coordinate system, and the dual-channel forward-looking radar is moving at a speed v r =70m / s, height h=10000m, fly in a straight line at a constant speed along the X-axis direction, forming a synthetic aperture with a length of 140m. The center position of the target scene is X c =10000m,Y c =0m, Z c = 0, that is, the pitch angle between the carrier and the center of the scene is 45°. On the platform, the launch channel T xa and receiving channel R xb Arranged at a spacing of 0.0365m along the Y axis, T xa The signal is transmitted at a pulse repetition frequency (PRF = 1000 Hz), T xa and R xb In this embodiment, the radar transmission wavelength is set to λ = 0.0167m, and the pulse width is T r = 1μs, a linear frequency modulation pulse signal with a bandwidth of B = 60MHz, a single snapshot sampling point number L is 128, and the number of channels N a is 2, the number of sampling points N in the synthetic aperture dimension is 2000, and the echo signal-to-noise ratio after pulse compression is 20dB.
[0091] like Figure 3 As shown, the azimuth angle range of the observation scene in this embodiment is -8.2° to 8.2°, the distance length is 160m, and the azimuth coordinate y of the receiving antenna 1 =0,y 2 = 0.0365m. For each channel, according to the number of synthetic aperture dimension sampling points and the number of distance sampling points N and L, the scene is evenly divided into N×L grids, and the two-way distance history R is calculated using formulas (1), (2), and (3) p (t s ,y k ; r 0 ,θ 0 ), and then use formula (4) to calculate the echo signal S e (y k ,t r ,t s ).
[0092] Step 2: Perform range pulse compression on the data obtained in step 1;
[0093] Based on the sampling rate F s and the distance sampling point L, calculate the fast time t r Sequence, and then get the matching function S of pulse compression ref (t r )=exp(-jπK r t r 2 ), based on the echo data generated in step 1, the result after echo pulse compression is obtained using formula (5), as Figure 4 shown.
[0094] Step 3, performing range migration correction on the data after the range pulse compression in step 2;
[0095] Based on formula (5), the fast time dimension is transformed by Fourier transform to obtain the range frequency echo signal S as shown in formula (6): 1 (y k ,f r ,t s ; r 0 ,θ 0 ). Then take the target at the center of the scene, i.e. the coordinates as X c =10000m,Y c =0m, Z c = 0 is used as the reference target, and the target signal after phase compensation is obtained using equation (7). Based on equation (7), considering that the range curvature is usually small, the first-order keystone transform is used to implement the linear range unit migration (RCM) correction, and the slow time is scaled using equation (8), and then combined with the range inverse Fourier transform, the echo signal S after RCM is obtained. 3 (y k ,t r ,t s ; r 0 ,θ 0 ) is shown in formula (9). In order to simplify formula (9), the echo signal S of each distance unit 4 (y k ,t s ; r 0 ,θ 0 ) is expressed by formula (12).
[0096] Step 4: Based on step 3, construct the echo domain time-frequency-space steering matrix, and establish a linear mapping equation between the echo and the original scene;
[0097] Based on the preprocessed echo, a steering matrix containing time-frequency-space information is designed according to formula (21).
[0098] Pitch angle information for each distance unit Where R l Indicates the slant distance corresponding to the lth distance unit. In the parameters in Table 1, the range of the pitch angle is calculated to be 44.64°~45.31°. In addition, the azimuth angles divided according to the observation scene range are also required. The forward observation area ranges from -8.2° to 8.2°, which is divided into M grid points. Equations (13) and (14) fully describe the Doppler changes and path differences between the target and multiple channels.
[0099] Then, according to the basic parameters provided in Table 1, the steering matrix A formed by the Doppler change shown in equation (17) and the steering matrix V formed by the path difference in the spatial domain shown in equation (18) are designed. Combining the two steering matrices, according to equation (21), the dual-channel time-frequency-space steering matrix H is obtained. H is the established steering matrix, which fully considers the time-frequency-space information of the echo data. By solving equation (20), super-resolution imaging and left / right ambiguity resolution can be achieved simultaneously.
[0100] In summary, the design diagram of the echo domain time-frequency-space steering matrix is as follows: Figure 5 shown.
[0101] Step 5: Jointly solve the time-frequency-space model, that is, use the regularized iterative adaptive method to solve the linear mapping equation between the echo in step 4 and the original scene, and simultaneously realize the left / right ambiguity resolution and super-resolution imaging of the dual-channel forward-looking SAR;
[0102] In practice, in order to overcome the influence of noise, a diagonally loaded regularized covariance matrix is used for solution, that is, in the covariance matrix R, a diagonal matrix μI is introduced 2M×2M To solve, in this embodiment, the regularization coefficient μ is set to 10, the number of iterations is T = 15, and for the echo signal Z of a distance unit, the steering matrix H of a distance unit constructed in step 4 is used, combined with the regularized iterative adaptive method in the specific step 5, to iteratively obtain the imaging result of a distance unit. Then, combined with the imaging results of all distance units, the final super-resolution imaging result can be obtained as follows Figure 6 shown.
[0103] In summary, the method of the present invention first obtains the echo data of the area to be imaged, performs range pulse compression and range migration correction on the acquired data, constructs a steering matrix that considers time-frequency-space information, establishes a linear equation between the echo and the original scene, and then uses a regularized iterative adaptive method to solve the equation, thereby simultaneously realizing dual-channel forward-looking SAR left / right ambiguity resolution and super-resolution imaging. The method of the present invention directly realizes dual-channel forward-looking SAR left / right ambiguity resolution and super-resolution imaging in the echo domain at the same time, and under the same channel resource conditions, better imaging performance can be obtained than the existing image domain method.
[0104] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.
Claims
1. A dual-channel radar forward imaging method, the specific steps are as follows: Step 1: Acquire echo data of the area to be imaged; The dual-channel radar forward imaging adopts a single-transmit and two-receive channel configuration, that is, one channel transmits signals and two channels receive signals; if there is any point target P0(x0,y0,0)=(r0sinα0cosθ0,r0sinα0sinθ0,0) in the scene, then its distance history R to the transmitting channel pt The distance history R of the receiving channel pr The expressions are as follows: in, v r represents the flight speed of the platform, h represents the flight altitude of the platform, r0 and θ0 represent the distance of the point target P0 relative to the transmitting antenna T xa The slant range and azimuth of the target, α0 represents the angle between the target line of sight and the vertical direction, t s Indicates slow time, y k represents the azimuth coordinate of the kth receiving antenna (k=1,2); For the point target P0, its two-way distance history R p (t s ,y k ; r0,θ0) are expressed as follows: R p (t s ,y k ;r0,θ0)=R pt (t s ;r0,θ0)+R pr (y k ,t s ;r0,θ0) (3) Assuming the transmitted signal is a linear frequency modulation pulse, the echo signal S received by the two channels e (y k ,t r ,t s ) is expressed as follows: Among them, σ p represents the scattering coefficient of the target, K r represents the distance modulation frequency, c represents the speed of light, λ represents the wavelength of the transmitted signal, and t r Indicates fast time; Step 2: Perform range pulse compression on the data obtained in step 1; Set the matching function of pulse compression to S ref (t r )=exp(-jπK r t r 2 ), then the pulse compressed signal S c (y k ,t r ,t s ) is expressed as follows: Among them, IFFT represents the inverse Fourier transform operator, and FFT represents the Fourier transform operator; Step 3, performing range migration correction on the data after the range pulse compression in step 2; First, perform the range Fourier transform FT on equation (5) to obtain the echo signal S1(y k ,f r ,t s ; r0,θ0), the expression is as follows: Among them, S(f r )=FT[sinc(B r t r )],B r represents the bandwidth of the LFM pulse, f0 represents the carrier frequency, and f r Indicates distance frequency; Then select the azimuth as 0° and the slant distance as r p The point target is taken as the reference target, and its coordinates are P r (x r ,y r ,0)=(r p sinα0,0,0), then the echo signal S2(y k ,f r ,t s ; r0,θ0) are expressed as follows: Then, the first-order keystone transformation is used to realize the linear range unit migration RCM correction, that is, the scale transformation of the slow time is performed, and the expression is as follows: Among them, η represents the slow time after scale transformation; Substitute equation (8) into equation (7) and perform inverse Fourier transform to obtain the echo S3(y k ,t r ,t s ; r0,θ0), the expression is as follows: Then the echo S4(y k ,t s ; r0,θ0) are expressed as follows: For R in formula (10), p (t s ,y k ;r0,θ0)-R p (t s ,y k ; r p ,0) Perform Taylor expansion and retain the first-order term. The expression is as follows: R p (t s ,y k ;r0,θ0)-R p (t s ,y k ;r p ,0)≈2(r0-r p )-2v r sinα0(1-cosθ0)·t s -sinα0sinθ0·y k (11) Substituting equation (11) into equation (10), the echo S4(y k ,t s ; r0,θ0) are expressed as follows: Among them, the first term represents a constant that does not affect azimuth focusing, the second term represents the phase of the path difference related to the spatial channel, and the third term represents the phase related to the synthetic aperture; Step 4: Based on step 3, construct the echo domain time-frequency-space steering matrix, and establish a linear mapping equation between the echo and the original scene; According to formula (12), 2M narrowband signals are assumed, that is, M targets are symmetrically distributed on the left and right sides of the route. Then, the received signal expressions of the nth (n = 1, 2…N) sampling point in each channel are as follows: Among them, α l represents the pitch angle of the lth range unit, l = (1, 2, ..., L), L represents the number of sampling points in the range direction, and They represent the noise and echo of the nth sample in the kth channel (k=1,2), respectively, sn Indicates the nth sampling point of slow time; and denote the scattering coefficients of left and right symmetric targets, respectively, represents the azimuth of the mth target; Equations (13) and (14) describe the Doppler variation and path difference between the target and multiple channels. By combining the results of N sampling points in each channel, they are written in matrix form. The expressions are as follows: Among them, (·) represents the dot product of the matrix, s k =[s k1 ,s k2 ,...,s kN ] H represents the echo matrix of the kth channel, (·) H Indicates the transposition operation of the matrix. represents the azimuth of the target, n k =[n k1 ,n k2 ,...,n kN ] H represents the noise matrix of the kth channel, and Respectively represent the scattering coefficient matrix of the left or right target in the scene; A represents the guidance matrix formed by Doppler change, and the expression is as follows: in, It represents the steering matrix formed by the spatial path difference, and its expression is as follows: Solve equations (15) and (16) at the same time to resolve the left / right ambiguity. The expression is as follows: Then rewrite equation (19), that is, construct the linear mapping equation between the echo and the original scene, the expression is as follows: Z 2N×1 =H 2N×2M ×g 2M×1 +N 2N×1 (20) Among them, Z 2N×1 =[s1,s2] H , g 2M×1 =[σ p ,σ s ] H , N 2N×1 =[n1,n2] H , H represents the echo domain time-frequency-space steering matrix, and its expression is as follows: Step 5: Jointly solve the time-frequency-space model, that is, use the regularized iterative adaptive method to solve the linear mapping equation between the echo in step 4 and the original scene, and simultaneously realize the left / right ambiguity resolution and super-resolution imaging of the dual-channel forward-looking SAR; According to formula (20), the weighted least squares cost function is defined as The expression is as follows: Among them, h m represents the mth column of the steering matrix H, R represents the covariance matrix, and R = HPH H , P represents a diagonal matrix, the diagonal elements represents the target scattering coefficient; For formula (22) Taking the partial derivative and setting the result equal to zero, we obtain: According to the matrix inversion lemma, we can get: Substituting it into formula (23), we can get: Then the diagonally loaded regularized covariance matrix is used for solution, that is, in the covariance matrix R, the diagonal matrix μI is introduced 2M×2M The solution is as follows: A. Initialization: identity matrix I, regularization parameter μ, number of iterations T; B. For the echo data Z, obtain the data of its lth range unit; C. Construct the steering matrix H according to formula (20); D、P=|H H Z| 2 / (H H H) 2 ; E. for i = 1 to T do; F、R=HPH H +μI; G, for m = 1 to 2M do; H、 I, end for; J、 K, end for; Finally, all distance units are traversed, the operations in steps AK are executed, and the imaging results are output to realize left / right ambiguity resolution and super-resolution imaging of dual-channel forward-looking SAR.
Citation Information
Patent Citations
Airborne bistatic SAR one-transmitting-antenna and two-receiving-antenna clutter suppression method
CN107728117A
Airborne radar oblique forward-looking super-resolution imaging method
CN111060909A
Space-time adaptive iterative super-resolution imaging method for airborne foresight array radar
CN114545401A
Airborne radar foresight three-dimensional imaging technology based on two-dimensional super-resolution algorithm
CN115480245A
Multi-channel radar foresight super-resolution imaging method
CN115685203A
