ISAR imaging method based on time-varying resolution estimation

Abstract

The invention provides an ISAR (Inverse Synthetic Aperture Radar) imaging method based on time-varying resolution estimation, and aims to solve the problems of overlong data acquisition time, time resource waste and relatively low reconnaissance efficiency caused by the fact that fixed imaging data acquisition time is set in the conventional ISAR imaging and then an optimal data segment is selected from the fixed imaging data acquisition time for imaging. The transverse resolution is estimated in real time in a sliding window mode, the imaging accumulation time can be adaptively changed, so that the time utilization rate is improved, imaging reconnaissance of multiple targets on the wide-area sea surface in a short time can be achieved, scattering points are prevented from crossing distance units under a large rotation angle through distance down-sampling, and the imaging reconnaissance accuracy is improved. A short frame multiplexing stream processing structure is used for reducing the calculation amount during rotating speed estimation, the estimation precision under the non-uniform-speed rotating condition is improved through segmented accumulated rotating angles, the problem of efficient imaging of dense sea surface ship targets can be effectively solved, and the method has large application potential in the aspects of intelligence reconnaissance, wide-area monitoring and target recognition.

Description

Technical Field

[0001] The invention belongs to the technical field of ISAR imaging design, and in particular relates to an ISAR imaging method based on time-varying resolution estimation. Background Art

[0002] Inverse synthetic aperture radar (ISAR) is an active microwave remote sensing device. Compared with the defect of synthetic aperture radar (SAR) that it can only image stationary targets, ISAR can image non-cooperative moving targets such as ships, aircraft, missiles, satellites, etc. around the clock and all weather, obtain information on target structure, shape, size, etc., and realize target recognition, classification and feature extraction, etc. It has great application prospects in the fields of space target observation and sea surface target reconnaissance.

[0003] Airborne ISAR imaging of surface ships is currently the most commonly used ISAR application scenario for airborne radars. Since surface ships are widely distributed and numerous, and ISAR can only perform beam irradiation on a fixed small area, it is difficult to effectively obtain the target ISAR echo without prior information on the ship's position. Therefore, it is usually necessary to first use a wide-area sea search method to detect and track a large range of multiple batches of ship targets, and then select a target of interest to enter the ISAR imaging mode. In the ISAR mode, since the motion state of the target is unknown, in order to obtain a better imaging effect, a longer accumulation time is generally set in advance, and then a suitable time period is selected for imaging. The single-frame accumulation time of this ISAR imaging mode is fixed, and it cannot be adaptively adjusted according to the target motion state. The time utilization rate is low. When the number of surface ships is large, it will compete with the wide-area sea search for time resources, and it is impossible to take into account both high-precision fine imaging and wide-area situational awareness.

[0004] In addition, effective classification and recognition of ISAR imaging depends on high-precision target size estimation, namely lateral calibration. The core of lateral calibration is to obtain the "real" rotation angle of the target relative to the radar within the cumulative imaging time. When the ship's rotation parameters are unknown, there are certain difficulties. Existing lateral calibration methods include the following categories:

[0005] The first type is to use radar narrowband measurement data to fit the target motion trajectory, and then obtain the imaging angle through the relationship between the trajectory and the radar position. However, due to the low accuracy of radar angle measurement and the inability to calculate the imaging angle caused by the change of the ship target posture, the azimuth calibration accuracy is low.

[0006] The second type is the prominent point tracking method, which uses the scattering points in ISAR imaging to estimate the rotation angle. It does not require target trajectory and attitude data. The accuracy depends on the signal processing algorithm and radar imaging performance. It relies on prior characteristics such as uniform rotation of the target. When the target's attitude motion is more complex, it will seriously affect the lateral calibration accuracy.

[0007] The third category is the motion parameter estimation method based on the overall quality of ISAR imaging. Within all possible imaging angle ranges, the most likely angle is determined by image entropy, contrast, etc. It does not require the constraint of prior information on target motion, but requires a fine search granularity and a heavy computational burden.

[0008] In addition, the existing horizontal calibration method obtains the average value within a fixed time and does not have time variability.

[0009] This application is proposed in view of the above-mentioned technical defects. Summary of the invention

[0010] The purpose of this application is to provide an ISAR imaging method based on time-varying resolution estimation, in which the imaging accumulation time is fixed and cannot be adaptively adjusted according to the target motion state when airborne ISAR images surface ship targets. The time utilization rate is low, and there are problems such as long-term acquisition, selective imaging, and low efficiency of multi-target imaging. The method uses lateral resolution as a judgment criterion, adaptively estimates the image resolution that changes with time, stops data acquisition when the resolution requirement is met, shortens the ISAR accumulation time, and improves the ISAR imaging efficiency under massive ship targets. In addition, the method avoids the cross-unit migration of scattering points under large turning angles by downsampling in the range direction, improves the phase estimation accuracy, estimates the instantaneous angular velocity by a sliding window method in the azimuth direction, and accumulates the angle in segments to avoid the problem of low rotation speed estimation accuracy and easy defocusing of imaging due to non-uniform rotation and periodic rotation of the target.

[0011] The technical solution of this application is:

[0012] An ISAR imaging method based on time-varying resolution estimation, comprising:

[0013] Step 1: Parameter setting: set the maximum number of ISAR imaging points per frame to N and the azimuth resolution index to ρ a , the number of points of sliding window estimation speed is N sub , the number of step points of the sliding window is N step ;

[0014] Step 2: accumulating echo signals, and performing pulse compression, motion compensation, envelope alignment, and translational phase compensation on the echo signals;

[0015] Step 3: When the azimuth accumulation reaches the segmented sliding window condition, the sliding window is used to estimate the rotation speed in real time, and the accumulated rotation angle and resolution are calculated. When the resolution meets the azimuth resolution index ρ a Or when the azimuth accumulation reaches the maximum imaging accumulation point number N, the accumulation is stopped and ISAR imaging processing is carried out with the accumulated echo signals;

[0016] Among them, the segmented sliding window condition is n = Nsub +kN step , where n is the cumulative number of the orientation.

[0017] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, the azimuth resolution index ρ a ISAR is set to 0.3m~3m;

[0018] The maximum accumulated points N is set to 4096;

[0019] Sliding window width N sub Set to 1 / 16 to 1 / 8 of the maximum accumulated points N;

[0020] Sliding window step setting N step Set to the sliding window width N sub 1 / 8 to 1 / 4 of.

[0021] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, in step 2, the signal after pulse compression and motion compensation of the echo signal is s1(n,τ):

[0022]

[0023]

[0024]

[0025] in,

[0026] s0(n,τ) is the original echo signal, n=1,2,...,N is the azimuth accumulation number, τ is the fast time in the range direction; FFT and IFFT are fast Fourier transform and inverse fast Fourier transform, subscript f r represents frequency domain transformation, and the subscript τ represents time domain transformation;

[0027] H MF (f r ) is the pulse compression function in the range frequency domain; f r is the range frequency; B is the bandwidth of the transmitted signal; γ is the frequency modulation slope of the transmitted signal; j is the imaginary sign; rect(·) is the rectangular window function, which takes 1 between [-0.50.5] and 0 for the rest;

[0028] H LRWC (f r ) is the motion compensation function; v is the platform flight speed; θ is the azimuth angle; t a is the azimuth time; c is the speed of light.

[0029] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, in step 2, the signal after envelope alignment of the echo signal is s2(n,τ):

[0030]

[0031] H corr (n,f r )=exp{-j4πf r ·△R(n) / c};

[0032]

[0033] in,

[0034] H corr (n,f r ) is the envelope alignment function;

[0035] △R(n,τ) is the offset between the nth distance line and the reference datum; P max is the position corresponding to the maximum value of the distance in the cross-correlation function R(n,τ); N r is the number of sampling points in the distance direction;

[0036]

[0037] or L ref (n,τ)=a·L ref (n-1,τ)+s1(n-1,τ);

[0038] in,

[0039] L ref (n,τ) is the reference standard of the nth echo signal s1(n,τ); a is the weighting coefficient, which is 0.95.

[0040] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, in step 2, the signal after the translational phase compensation of the echo signal is s3(n,τ):

[0041]

[0042]

[0043] in,

[0044] The superscript * indicates the conjugation of a complex number; s diff (n.τ) is the intermediate calculation amount.

[0045] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, step three is specifically:

[0046] Determine whether n=N sub +kN step If not, return to step 2. If yes, start sliding the window and select the signal segment s3(n,τ),kN for estimating the speed. step +1≤n≤N sub +kN step , where k is the sequence number of the step window, and the starting value is 0;

[0047] The signal s3(n,τ) after translation compensation is downsampled by distance. Specifically, the signal s3(n,τ) after translation phase compensation is FFT-ed in the distance direction and the middle 1 / N is intercepted. d The spectrum of width is transformed into the range time domain by performing distance FFT, and the downsampled signal s4(n,τ) is obtained, where N d is the downsampling multiple;

[0048] Perform azimuth FFT on the downsampled signal s4(n,τ) and transform it to the range-Doppler domain to obtain the image s5(f n ,τ);

[0049] The image s5(f n ,τ) select the L distance unit data with the largest contrast o=1,2,…,L, estimate the rotation speed of all L distance units in pairs of combinations li and lj And according to the distance difference |l i -l j |Perform weighted summation to obtain the speed ω within a single sliding window k :

[0050]

[0051] Calculate the cumulative rotation angle △θ at the current sliding window moment k :

[0052]

[0053] in,

[0054] PRF is the pulse repetition frequency;

[0055] Calculate the lateral resolution ρ at the current sliding window moment k :

[0056]

[0057] in,

[0058] λ is the radar operating wavelength;

[0059] If k ≤ρ a , then stop echo accumulation and carry out ISAR imaging processing with the accumulated echo signals, otherwise return to step 2.

[0060] According to at least one embodiment of the present application, in the above-mentioned ISAR imaging method based on time-varying resolution estimation, in step 3, the rotation speeds of the two combinations of L distance units are The sub-aperture image phase difference method is used for estimation, specifically:

[0061] right Select the point with the largest azimuth amplitude, add windows on both sides for protection, set the window outside to zero, and slice the two azimuth images after adding windows. Do IFFT along the azimuth direction and get and Conjugate and multiply the two to calculate the phase after unwrapping

[0062]

[0063] Among them, unwrap(·) represents the phase unwrapping function; angle(·) is the phase taking function;

[0064] The unwrapped phase kN step +1≤n≤N sub +kN step Divide it into two segments of equal length along the azimuth direction, and use the least squares method to fit the slope of each segment

[0065] Calculate distance unit l i , l j Estimated speed between

[0066]

[0067]

[0068]

[0069] Among them, △y is l i , l j The distance between them; flag represents the direction of the rotation speed; sign(.) is the sign function, with a positive value of 1 and a negative value of -1; λ is the radar operating wavelength; F s is the sampling rate of echo range.

[0070] This application has at least the following beneficial technical effects:

[0071] The invention provides an ISAR imaging method based on time-varying resolution estimation. A fixed imaging data acquisition time is set for existing ISAR imaging, and then the optimal data segment is selected for imaging, which leads to the problems of long data acquisition time, waste of time resources and low reconnaissance efficiency. With the resolution reaching the standard as the criterion, the lateral resolution is estimated in real time by sliding window mode, and the imaging accumulation time can be adaptively changed to improve the time utilization rate. The imaging reconnaissance of multiple targets on a wide sea surface in a short time can be realized. The distance downsampling is used to avoid the scattering point crossing the distance unit under large turning angle, the short frame multiplexing stream processing structure is used to reduce the calculation amount when estimating the rotation speed, and the segmented accumulation angle is used to improve the estimation accuracy under non-uniform rotation conditions. The method can effectively solve the problem of efficient imaging of dense sea surface ship targets, and has great application potential in intelligence reconnaissance, wide-area surveillance and target recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 is a flow chart of an ISAR imaging method based on time-varying resolution estimation provided in an embodiment of the present application;

[0073] Figure 2 It is a schematic diagram of estimating time-varying speed by segmented sliding window provided in an embodiment of the present application;

[0074] Figure 3 It is a three-dimensional scatter plot of a simulated ship target provided in an embodiment of the present application;

[0075] Figure 4 This is a diagram of the simulation results of the time-varying turning angle estimation provided in the embodiment of the present application.

[0076] In order to better illustrate the present embodiment, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product. In addition, the drawings are only used for illustrative purposes and should not be construed as limitations on the present application. DETAILED DESCRIPTION

[0077] In order to make the technical solution and advantages of the present application clearer, the technical solution of the present application will be described in further detail in detail and in detail with reference to the accompanying drawings. It can be understood that the specific embodiments described here are only partial embodiments of the present application, which are only used to explain the present application, not to limit the present application. It should be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, and other related parts can refer to the general design.

[0078] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of this application shall have the usual meanings understood by those of ordinary skill in the art to which this application belongs. The words used in the description of this application to indicate orientation are only used to indicate relative directions or positional relationships. When the absolute position of the described object changes, its relative positional relationship may also change accordingly. The word "include" used in the description of this application indicates that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but does not exclude other elements or objects.

[0079] In addition, it should be noted that, unless otherwise clearly specified and limited, the words "installation", "connection" and similar terms used in the description of this application should be understood in a broad sense. For example, the connection can be a fixed connection or a detachable connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. Technical personnel in the field can understand its specific meaning in this application according to the specific circumstances.

[0080] The present application provides an ISAR imaging method based on time-varying resolution estimation. By estimating the rotation angle that varies with time, using the set resolution as the judgment criterion, the resolution at the current accumulation moment is estimated in real time, and the imaging accumulation time is adaptively selected, so as to solve the problems of low time utilization and low imaging efficiency of existing ISAR imaging and save time resources.

[0081] The ISAR imaging method based on time-varying resolution estimation provided by the present application first completes ISAR range processing, including pulse compression, motion compensation and envelope alignment, and then completes translational phase compensation, and then performs range downsampling, selects range resolution units and scattering points for speed estimation, and then estimates the instantaneous speed within the current sliding window in real time through an azimuth sliding window, calculates the rotation angle and resolution at the current moment by segmented accumulation, and stops accumulation and completes ISAR imaging when the resolution meets the requirements or exceeds the maximum accumulation time, such as Figure 1 shown.

[0082] Step 1: Parameter setting: set the maximum number of ISAR imaging points per frame to N and the azimuth resolution index to ρ a , the number of points of sliding window estimation speed is N sub , the number of step points of the sliding window is N step .

[0083] Azimuth resolution index ρ a It is the resolution index of ISAR imaging, usually between 0.3m and 3m, and is determined by external input.

[0084] The maximum number of accumulation points N is the fixed number of accumulation points per ISAR frame, and its typical value is 4096.

[0085] Sliding window width N sub Set to 1 / 16 to 1 / 8 of the maximum accumulated points N.

[0086] Sliding window step setting N step Set to the sliding window width N sub 1 / 8 to 1 / 4 of.

[0087] Step 2: accumulate echo signals, and perform pulse compression, motion compensation, envelope alignment, and translational phase compensation on the echo signals.

[0088] Starting from the first azimuth echo, pulse compression, motion compensation, envelope alignment, and translational phase compensation are performed on each collected echo.

[0089] Assume that the original echo signal is s0(n,τ), where n=1,2,...,N is the azimuth accumulation number; τ is the fast time in the range direction. Pulse compression, motion compensation, envelope alignment and translational phase compensation can be performed with reference to the existing processing methods. The specific related information can be referred to as follows.

[0090] Pulse compression function H in the range frequency domain MF (f r )for:

[0091]

[0092] Among them, f r is the range frequency; B is the bandwidth of the transmitted signal; γ is the frequency modulation slope of the transmitted signal; j is the imaginary sign; rect(·) is the rectangular window function, which takes the value of 1 between [-0.5 0.5] and 0 otherwise.

[0093] Motion compensation function H LRWC (f r )for:

[0094]

[0095] in,

[0096] v is the platform flight speed; θ is the azimuth angle; t a is the azimuth time; c is the speed of light.

[0097] The signal after pulse compression and motion compensation is s1(n,τ), where FFT and IFFT are fast Fourier transform and inverse fast Fourier transform, subscript f r represents frequency domain transformation, the subscript τ represents time domain transformation, and the subscript n represents azimuth transformation.

[0098]

[0099] Envelope alignment uses the weighted cumulative cross-correlation method. For the nth echo signal s1(n,τ), the reference reference L ref (n,τ) is the weighted sum of the previous n-1 lines, as shown in formula (6)

[0100]

[0101] Where a is the weighting coefficient, and its typical value is 0.95. Reference Benchmark L ref (n,τ) can be calculated recursively by formula (7) to reduce the amount of calculation:

[0102] L ref (n,τ)=a·L ref (n-1,τ)+s1(n-1,τ) (5)

[0103] The cross-correlation function R(n,τ) is calculated as:

[0104]

[0105] Calculate the offset △R(n,τ) between the nth distance line and the reference datum, where P max is the position corresponding to the maximum value of the distance in the cross-correlation function R(n,τ), N r is the number of sampling points in the distance direction.

[0106]

[0107] Envelope alignment function H corr (n,f r )for:

[0108] H corr (n,f r )=exp{-j4πf r ·△R(n,τ) / c} (8)

[0109] The signal s2(n,τ) after envelope alignment is:

[0110]

[0111] The Doppler center tracking method as shown in equation (12) is used to complete the translation compensation. The compensated signal is s3(n,τ), where the superscript * indicates the conjugate of the complex number.

[0112]

[0113]

[0114] Step 3: When the azimuth accumulation reaches the segmented sliding window condition, the sliding window is used to estimate the rotation speed in real time, and the accumulated rotation angle and resolution are calculated. When the resolution meets the azimuth resolution index ρ a Or when the azimuth accumulation reaches the maximum imaging accumulation point number N, the accumulation is stopped and the ISAR imaging processing is carried out with the echo signal accumulated in step 2.

[0115] This step is the core step of this application. Figure 2 This is a schematic diagram of segmented sliding window estimation, which mainly includes the following parts:

[0116] First, determine whether the current azimuth cumulative sequence number n satisfies the segmented sliding window condition n = N sub +kN step , where k is the number of the step window, the starting value is 0, if not satisfied, return to step 2, n = n + 1; if satisfied, start sliding the window and select the signal segment s3 (n, τ) for estimating the speed, kN step +1≤n≤N sub +kN step , k=k+1, and continue the subsequent operations.

[0117] The signal s3(n,τ) after translation compensation is downsampled by distance. The specific method is to perform distance FFT on s3(n,τ) and intercept the middle 1 / N d Width of spectrum, N d is the downsampling multiple. The higher the resolution, the higher the N d The larger the value, the larger the distance is. Then, perform FFT in the distance direction and transform it to the distance time domain to obtain the downsampled signal s4(n,τ).

[0118] Perform azimuth FFT on s4(n,τ) and transform it to the range-Doppler domain to obtain the image s5(f n ,τ):

[0119]

[0120] The image s5(f n ,τ) select the L distance unit data with the largest contrast The calculation formula of contrast IC(x) is:

[0121]

[0122] std(·) is the standard deviation function, and the superscript “-” represents the mean.

[0123] Choose any two distance units from L distance units. i , l j As a strong scatter point, estimate the speed within the sliding window

[0124] Speed The sub-aperture image phase difference method is used for estimation. Select the point with the largest azimuth amplitude, add windows on both sides for protection, and set the value outside the window to zero. Then slice the image of the two azimuths after adding windows. Do IFFT along the azimuth direction and get and The two are conjugate multiplied and the unwrapped phase is calculated according to formula (16): Where unwrap(·) represents the phase unwrapping function, and angle(·) is the phase taking function.

[0125]

[0126] Will kN step +1≤n≤N sub +kN step The azimuth direction is divided into two segments of equal length. According to formula (17), the slope of each segment is fitted using the least squares method.

[0127]

[0128] Where M = N sub / 2 is the half window width.

[0129] According to formula (18), the distance unit l is calculated i , l j Estimated speed between Where △y is l i , l j The distance between them, flag represents the direction of the rotation speed, sign(.) is the sign function, positive number is 1, negative number is -1.

[0130]

[0131]

[0132]

[0133] Where λ is the radar operating wavelength; PRF is the pulse repetition frequency; F s is the sampling rate of echo range.

[0134] Estimate the rotation speed of all L distance units in pairs And according to the distance difference |l i -l j |Perform weighted summation and calculate the speed ω within a single sliding window according to formula (20) k :

[0135]

[0136] According to formula (21), the cumulative rotation angle △θ at the current sliding window time k is calculated k :

[0137]

[0138] According to formula (22), the lateral resolution ρ at the current sliding window time k is calculated k , if k ≤ρ a , then stop echo accumulation, otherwise return to step 2, k=k+1, and continue the speed estimation and resolution calculation within the next sliding window width.

[0139]

[0140] In a specific embodiment, the three-dimensional scatter diagram of the ship target is as follows: Figure 3 As shown, there are 312 point targets in total, the ship is 150m long, the simulated radar works in the X-band, the wavelength is 3cm, the range is 50km, and the ISAR resolution index is ρ a =0.5m, signal bandwidth is 400MHz, sampling rate is 500MHz, PRF=1000, maximum accumulation points N=8192, sliding window width N sub =512, sliding window step N step =32, assuming that the ship performs one-dimensional pitch periodic motion, the rotation angle θ(t) = A·sin(2πt / Ω), where the amplitude A = 2.5° and the period Ω = 6.7s. The comparison between the estimated rotation angle at different accumulation moments and the true value is as follows Figure 4 As shown, the rotation angle that meets the resolution index is 1.72°. At this time, the estimated accumulation time is 0.71s, and the actual accumulation time is 0.79s, which is much smaller than the set fixed accumulation time of 8.2s.

[0141] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. Those skilled in the art should understand that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the scope of protection of the present application.

Claims

1. An ISAR imaging method based on time-varying resolution estimation, characterized in that: include: Step 1: Parameter setting: set the maximum number of ISAR imaging points per frame to N and the azimuth resolution index to ρ a , the number of points of sliding window estimation speed is N sub , the number of step points of the sliding window is N step ; Step 2: accumulating echo signals, and performing pulse compression, motion compensation, envelope alignment, and translational phase compensation on the echo signals; Step 3: When the azimuth accumulation reaches the segmented sliding window condition, the sliding window is used to estimate the rotation speed in real time, and the accumulated rotation angle and resolution are calculated. When the resolution meets the azimuth resolution index ρ a Or when the azimuth accumulation reaches the maximum imaging accumulation point number N, the accumulation is stopped and ISAR imaging processing is carried out with the accumulated echo signals; Among them, the segmented sliding window condition is n = N sub +kN step , where n is the cumulative number of the orientation.

2. The ISAR imaging method based on time-varying resolution estimation according to claim 1, characterized in that: Azimuth resolution index ρ a ISAR is set to 0.3m~3m; The maximum accumulated points N is set to 4096; Sliding window width N sub Set to 1 / 16 to 1 / 8 of the maximum accumulated points N; Sliding window step setting N step Set to the sliding window width N sub 1 / 8 to 1 / 4 of.

3. The ISAR imaging method based on time-varying resolution estimation according to claim 2, characterized in that: In step 2, the signal after pulse compression and motion compensation of the echo signal is s1(n,τ): in, s0(n,τ) is the original echo signal, n=1,2,...,N is the azimuth accumulation number, τ is the fast time in the range direction; FFT and IFFT are fast Fourier transform and inverse fast Fourier transform, subscript f r represents frequency domain transformation, and the subscript τ represents time domain transformation; H MF (f r ) is the pulse compression function in the range frequency domain; f r is the range frequency; B is the bandwidth of the transmitted signal; γ is the frequency modulation slope of the transmitted signal; j is the imaginary sign; rect(·) is the rectangular window function, which takes 1 between [-0.50.5] and 0 for the rest; H LRWC (f r ) is the motion compensation function; v is the platform flight speed; θ is the azimuth angle; t a is the azimuth time; c is the speed of light.

4. The ISAR imaging method based on time-varying resolution estimation according to claim 3, characterized in that: In step 2, the signal after envelope alignment of the echo signal is s2(n,τ): H corr (n,f r )=exp{-j4πf r ·△R(n) / c}; in, H corr (n,f r ) is the envelope alignment function; △R(n,τ) is the offset between the nth distance line and the reference datum; P max is the position corresponding to the maximum value of the distance in the cross-correlation function R(n,τ); N r is the number of sampling points in the distance direction; or L ref (n,τ)=a·L ref (n-1,τ)+s1(n-1,τ); in, L ref (n,τ) is the reference standard of the nth echo signal s1(n,τ); a is the weighting coefficient, which is 0.

95.

5. The ISAR imaging method based on time-varying resolution estimation according to claim 4, characterized in that: In step 2, the signal after the translational phase compensation of the echo signal is s3(n,τ): in, The superscript * indicates the conjugation of a complex number; s diff (n.τ) is the intermediate calculation amount.

6. The ISAR imaging method based on time-varying resolution estimation according to claim 5, characterized in that: Step three is as follows: Determine whether n=N sub +kN step If not, return to step 2. If yes, start sliding the window and select the signal segment s3(n,τ),kN for estimating the speed. step +1≤n≤N sub +kN step , where k is the sequence number of the step window, and the starting value is 0; The signal s3(n,τ) after translation compensation is downsampled by distance. Specifically, the signal s3(n,τ) after translation phase compensation is FFT-ed in the distance direction and the middle 1 / N is intercepted. d The spectrum of width is transformed into the range time domain by performing distance FFT, and the downsampled signal s4(n,τ) is obtained, where N d is the downsampling multiple; Perform azimuth FFT on the downsampled signal s4(n,τ) and transform it to the range-Doppler domain to obtain the image s5(f n ,τ); The image s5(f n ,τ) select the L distance unit data with the largest contrast Estimate the pairwise combination l of all L distance units i , l j Speed And according to the distance difference |l i -l j |Perform weighted summation to obtain the speed ω within a single sliding window k : Calculate the cumulative rotation angle △θ at the current sliding window moment k : in, PRF is the pulse repetition frequency; Calculate the lateral resolution ρ at the current sliding window moment k : in, λ is the radar operating wavelength; If k ≤ρ a , then stop echo accumulation and carry out ISAR imaging processing with the accumulated echo signals, otherwise return to step 2.

7. The ISAR imaging method based on time-varying resolution estimation according to claim 6, characterized in that: In step 3, the rotation speed of each pair of L distance units is The sub-aperture image phase difference method is used for estimation, specifically: right Select the point with the largest azimuth amplitude, add windows on both sides for protection, set the window outside to zero, and slice the two azimuth images after adding windows. Do IFFT along the azimuth direction and get and Conjugate and multiply the two to calculate the phase after unwrapping Among them, unwrap(·) represents the phase unwrapping function; angle(·) is the phase taking function; The unwrapped phase kN step +1≤n≤N sub +kN step Divide it into two segments of equal length along the azimuth direction, and use the least squares method to fit the slope of each segment Calculate distance unit l i , l j Estimated speed between Among them, △y is l i , l j The distance between them; flag represents the direction of the rotation speed; sign(.) is the sign function, with a positive value of 1 and a negative value of -1; λ is the radar operating wavelength; F s is the sampling rate of echo range.

Images