A forward-looking imaging method for azimuth multi-channel synthetic aperture radar

Abstract

This invention proposes a forward-looking imaging method for azimuth-oriented multi-channel synthetic aperture radar (SAR), solving the problems of channel mismatch correction and non-uniform signal reconstruction in forward-looking imaging mode. The method includes: calculating the extended spectrum band caused by the spatially varying characteristics of the Doppler center frequency; converting the original echo data from the forward-looking azimuth multi-channel to the range-Doppler domain; calculating the covariance matrix based on the range-Doppler domain of the original echo data and constructing cost functions corresponding to different spectrum bands; estimating the inter-channel amplitude and phase errors using classical optimization methods and correcting the original echo data; performing range-trip correction on the echo data of each received channel in the range frequency domain based on the corrected original echo data; reconstructing the non-uniform signal spectrum of the echo signal after range-trip correction; performing upsampling and zero-padding and inverse range-trip processing on the reconstructed signal; and finally obtaining a finely focused SAR image through two-dimensional focusing processing.

Description

Technical Field

[0001] This invention belongs to the field of signal processing technology, and particularly relates to a forward squint imaging method for azimuth multi-channel synthetic aperture radar. Background Technology

[0002] High-resolution, wide-coverage spaceborne SAR systems enable high-resolution detailed surveys of ultra-wide areas. This reduces the frequency of repeated observations, improving timeliness, and allows for the acquisition of crucial intelligence information through wide-area observation. It serves as a vital Earth observation tool to meet the needs of military strategic and tactical applications. The spaceborne multi-channel forward-looking SAR imaging mode, through flexible control of antenna beam pointing, enables high-resolution, wide-coverage, and repeated observations of targets in hotspot areas to enhance their three-dimensional structural features, thereby improving SAR image interpretation and target identification capabilities. It can also serve as a pre-reconnaissance mode, providing clear prior information for forward and side-looking imaging.

[0003] However, in the azimuth multi-channel forward-looking system, the spatially varying Doppler center frequency with range will lead to strong two-dimensional coupling of the echo signal in both azimuth and range, as well as azimuth spectral expansion. This poses new challenges to the inter-channel amplitude and phase consistency correction and the high-precision, high-efficiency reconstruction of the non-uniform azimuth signal. Existing classical reconstruction algorithms and inter-channel consistency correction algorithms do not consider the spatially varying characteristics of the Doppler center frequency and are difficult to apply directly.

[0004] Therefore, it is necessary to study the channel consistency correction method and the azimuth non-uniform signal reconstruction method in the forward-looking mode of azimuth multi-channel SAR, so as to lay the foundation for the high-precision two-dimensional focusing processing of echo signal spectrum reconstruction and two-dimensional decoupling under the spaceborne multi-channel forward-looking SAR system. Summary of the Invention

[0005] This invention proposes a forward-looking imaging method for azimuth multi-channel synthetic aperture radar, which solves the problems of channel mismatch correction and non-uniform signal reconstruction in azimuth multi-channel SAR systems under forward-looking imaging mode.

[0006] The present invention is achieved through the following technical solution.

[0007] A forward-looking imaging method for azimuth-oriented multi-channel synthetic aperture radar includes the following steps:

[0008] Step 1: Calculate the extended spectrum caused by the spatial variation characteristics of the Doppler center frequency;

[0009] Step 2: Convert the raw echo data of the forward-looking azimuth multi-channel to the range-Doppler domain;

[0010] Step 3: Calculate the covariance matrix based on the distance-Doppler domain of the original echo data, and construct the cost function corresponding to different spectral bands;

[0011] Step 4: For the cost functions corresponding to different spectral bands in Step 3, use classical optimization methods to estimate the inter-channel amplitude and phase errors and correct the original echo data;

[0012] Step 5: Perform distance-free travel correction on the echo data of each receiving channel in the distance-frequency domain based on the original echo data corrected in Step 4.

[0013] Step Six: Perform non-uniform signal spectrum reconstruction on the echo signal after distance travel correction in Step Five;

[0014] Step 7: Perform upsampling, zero-padding, and inverse distance traversal processing on the reconstructed signal from Step 6; then obtain the finely focused SAR image through two-dimensional focusing processing.

[0015] The beneficial effects of this invention are:

[0016] 1. Compared with the traditional forward-looking azimuth multi-channel SAR imaging processing method, the present invention solves the high-precision two-dimensional focusing processing in the azimuth multi-channel forward-looking SAR mode by constructing a complete signal processing flow;

[0017] 2. This invention considers the extended spectrum caused by the Doppler center frequency that varies with distance, and uses the decomposed covariance matrix to construct the cost function corresponding to different spectrum segments.

[0018] 3. This invention can be used in high-precision focusing processing of spaceborne azimuth multi-channel synthetic aperture radar. Attached Figure Description

[0019] Figure 1 This is a flowchart of the forward-looking imaging method for azimuth multi-channel synthetic aperture radar according to the present invention;

[0020] Figure 2 This is a schematic diagram of forward-looking imaging using the azimuth multi-channel synthetic aperture radar of the present invention;

[0021] Figure 3 This is the spectral overlay diagram of the azimuth multi-channel forward-looking SAR of the present invention;

[0022] Figure 4 This is a schematic diagram of the phase error estimation results in the forward squint imaging mode of the azimuth multi-channel synthetic aperture radar of the present invention;

[0023] Figure 5 This is a schematic diagram of the reconstructed two-dimensional spectrum and two-dimensional focusing processing results in the forward squint imaging mode of the azimuth multi-channel synthetic aperture radar of the present invention. Detailed Implementation

[0024] Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the embodiments shown and described in the drawings are merely exemplary and are intended to illustrate the principles and spirit of the present invention, and are not intended to limit the scope of the present invention.

[0025] like Figure 1 As shown, the present invention provides a forward-looking imaging method for azimuth-oriented multi-channel synthetic aperture radar, which specifically includes the following steps:

[0026] Step 1: Calculate the extended spectrum caused by the spatial variation of the Doppler center frequency; the specific formula is as follows:

[0027]

[0028] Among them, v s For platform speed, B r For system bandwidth, θ c The angle is forward, and c is the speed of light.

[0029] Step 2: Convert the raw echo data of the forward-looking azimuth multi-channel to the range-Doppler domain; in this embodiment, the specific steps are as follows:

[0030] The echo data from the M azimuth channels are s(η,τ)=[s1(η,τ),2(η,τ),…, M (η,τ)] T Perform a Fast Fourier Transform (FFT) along the azimuth direction to transform the echo signal from the time domain to the range Doppler domain:

[0031] S(f a ,τ)=FFT[s(η,τ)] (1)

[0032] Where η and τ are the azimuth time and distance time, respectively, and f a For the azimuth frequency, S(f a ,) represents the range Doppler domain signal of the original echo data.

[0033] Step 3: Calculate the covariance matrix based on the range-Doppler domain of the original echo data, and construct the cost function corresponding to different spectral bands; in this embodiment, the specific steps are as follows:

[0034] 3.1 Using the echo signal S(f) in the range-Doppler domain obtained in step two a The covariance matrix is ​​estimated for each distance Doppler cell, and then eigenvalue decomposition is performed on the covariance matrix.

[0035]

[0036] Where Σ and U are defined as the descending order of the eigenvalues ​​λ. i The corresponding diagonal matrix and eigenvectors; Us The eigenvectors corresponding to the N largest eigenvalues, arranged in descending order, constitute the signal subspace; U n These are the eigenvectors corresponding to the last MN smaller eigenvalues, arranged in descending order.

[0037] 3.2 Construct the cost function Ω={Ω1,Ω2,Ω3} corresponding to different spectral bands using the covariance matrix after eigenvalue decomposition; the specific formula is as follows:

[0038]

[0039] in, Let K be the steering vector corresponding to the echo signal, K be the Doppler ambiguity number, N be the number of spectral aliasings after PRF undersampling, and L = L1, L2, L3}. The ranges of different spectral segments are shown below:

[0040]

[0041] Where Δf=Frac[N]·PRF / 2 is the frequency offset of the fractional part corresponding to the number of aliasing N, B sq The extended spectrum band introduced by the spatially varying characteristics of the Doppler center frequency; PRF is the pulse repetition frequency, f bc The baseband Doppler center frequency;

[0042] Step 4: For the cost functions corresponding to different spectral bands in Step 3, use classical optimization methods to estimate the inter-channel amplitude and phase errors and correct the original echo data; in this embodiment, the specific steps are as follows:

[0043] Based on the cost functions for different spectral bands in step three, under the external constraint Γ H ·w=1,(w=[0,…,0,1,0,…,0] T The amplitude and phase errors are extracted and corrected using optimization methods respectively.

[0044]

[0045] in, δ i and These are the amplitude and phase errors of the i-th receiving channel, respectively;

[0046] The optimized parsing expression is then expressed as:

[0047]

[0048] Using the above optimization results Error correction is performed on the echo signal, i.e.

[0049] like Figure 4 As shown, in specific implementation, the extended spectrum band B is introduced by the forward oblique angle. sq The effect of this is that, after PRF undersampling, if the cost function within the L2 spectrum remains unchanged across the entire frequency range of [-PRF / 2, PRF / 2], the phase error estimation result will exhibit f. j,1 and f j,2 There are two transition points, and the following correspondence exists:

[0050]

[0051] Step 5: Perform range-trip correction on the echo data of each receiving channel in the range-frequency domain based on the original echo data corrected in Step 4; the specific formula is as follows:

[0052]

[0053] Among them, Δx=[Δx1,Δx2,…,Δx M [ ] represents the phase center spacing between each receiving channel, f0 and f r These are the carrier frequency and the range frequency, respectively, S′(η,f r ) represents the corrected original echo data in the range frequency domain for the echo data of each receiving channel, v s For platform speed, θ c η is the forward oblique angle, and η is the azimuth time.

[0054] Step 6: Perform non-uniform signal spectrum reconstruction on the echo signal after distance travel correction in Step 5 within the frequency range of [-PRF / 2, PRF / 2].

[0055] The principle behind this step is that after processing in step five, most of the energy of the echo signal is concentrated at the zero Doppler position. Therefore, within the range of [-PRF / 2, PRF / 2], the echo signals of each receiving channel can be reconstructed using classical filter methods to recover the uniformly sampled signal.

[0056] Step 7: Perform upsampling, zero-padding, and inverse range walk processing on the reconstructed signal from Step 6; then obtain the finely focused SAR image through two-dimensional focusing processing;

[0057] This step is because the azimuth bandwidth of the echo signals from each receiving channel, after range travel correction in step five and signal reconstruction in step six, is B. a =(0.886·2v) s cosθ c ) / L a This is not the true Doppler history of the signal, therefore it is necessary to recover the true Doppler history of the signal.

[0058] This embodiment specifically includes the following steps:

[0059] 7.1 Upsample the reconstructed signal and coordinate axes in the time domain, i.e., perform zero-padding in the frequency domain; the number of zeros padded is as follows:

[0060] ΔN a = (B a +B sq -N·PRF)·T a (10)

[0061] Among them, T a For the time to synthesize the pore size, B a For the azimuth spectral bandwidth, B sq The extended spectrum segment introduced by the spatial variation characteristics of the Doppler center frequency, where N is the azimuth spectral aliasing number.

[0062] 7.2 The upsampled signal is subjected to inverse range-trip correction processing in the range-direction frequency domain to recover the true Doppler history of the signal; the inverse range-trip correction function is as follows:

[0063]

[0064] Where η' is the azimuth time axis after upsampling;

[0065] At this point, the signal has been fully recovered to reflect the true Doppler trajectory;

[0066] 7.3 The signal after inverse range migration processing can be processed by two-dimensional focusing using the classic range migration algorithm (RMA) to obtain the focused SAR image.

[0067] In this embodiment, to verify the effectiveness of the method proposed in this patent, a simulation experiment was conducted according to the parameters in Table 1, such as... Figure 3 The following table shows the spectral aliasing after azimuth undersampling and de-skew processing corresponding to the simulation parameters in Table 1. Figure 4 for Figure 3 The phase error estimation results corresponding to the aliasing spectrum situation are shown in the figure. The phase error estimated in the L2 frequency band is the actual phase mismatch between channels in the system, which is consistent with the preset results in Table 1. Figure 5 The images are the reconstructed two-dimensional spectrum and the SAR image after two-dimensional focusing processing in the forward-looking mode of the azimuth multi-channel synthetic aperture radar after channel mismatch correction.

[0068] Table 1 Simulation Parameters

[0069]

[0070] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. An azimuthally multi-channel forward-looking synthetic aperture radar imaging method, characterized in that, The method comprises the following steps: Step one: calculate the spread spectrum caused by the Doppler center frequency space variation characteristics; the spread spectrum is calculated in the following way: wherein is the system bandwidth, is the front view angle, c is the speed of light; Step two: convert the forward glancing azimuth multi-channel original echo data to the range Doppler domain; the conversion is performed in the following way: azimuth direction echo data A fast Fourier transform (FFT) is performed in the azimuth direction to transform the echo signals from the time domain to the range Doppler domain: wherein, and are azimuth time and range time, respectively, is azimuth frequency, is the range Doppler domain signal of the raw echo data; Step three: calculate the covariance matrix according to the range Doppler domain of the original echo data, and construct the cost function corresponding to different spectrum segments; Step four: estimate the amplitude and phase errors between channels and correct the original echo data by using the classical optimization method for the cost function corresponding to different spectrum segments in step three; Step five: perform the range migration correction on the echo data of each receiving channel in the range frequency domain for the original echo data corrected in step four; Step six: perform the non-uniform signal spectrum reconstruction on the echo signal corrected in step five; Step seven: perform the up-sampling zero padding and inverse range migration processing on the reconstructed signal in step six; and then obtain the fine-focused SAR image by performing the two-dimensional focusing processing.

2. The azimuthally multi-passive synthetic aperture radar forward looking imaging method according to claim 1, characterized in that, Step three is performed in the following way: 3.1 estimate the covariance matrix by using the echo signal in the range Doppler domain obtained in step two, and perform the eigenvalue decomposition on the covariance matrix; 3.2 construct the cost function corresponding to different spectrum segments by using the eigenvalue-decomposed covariance matrix.

3. The azimuthally multi-passive synthetic aperture radar forward looking imaging method of claim 2, wherein, Step seven is performed in the following way: 7.1 perform the up-sampling in the time domain on the reconstructed signal and the coordinate axis, that is, the frequency domain zero padding operation; 7.2 perform the inverse range migration processing on the up-sampled signal in the range frequency domain, and restore the true Doppler history of the signal; 7.3 perform the two-dimensional focusing processing on the inverse range migration processed signal by using the classical range migration algorithm, and obtain the focused SAR image.

Images