A robust stepped-frequency ISAR imaging radar velocity estimation method

Abstract

The application discloses a kind of robust frequency-modulated stepped frequency ISAR imaging radar velocity estimation methods, belong to radar technical field, first construct multiple frames frequency-modulated stepped frequency data model;Again, the velocity estimation and compensation of rough accuracy are realized using the same frequency point offset estimation of multi-frequency point synthesis;Then, the high-resolution imaging of the coarse resolution distance unit where the special point is located is obtained by the IFFT of intra-frame sub-pulse dimension, and the high-precision velocity estimation and compensation are realized using the high-resolution image of multiple special point synthesis;Finally, ISAR imaging is realized through the FFT between frames.The application realizes the estimation and compensation of velocity gradually using two-step method, first realizes the estimation and compensation of large motion amount by rough estimation, then realizes the residual motion estimation and compensation of high precision by synthesis high resolution;Offset estimation is realized using envelope cross correlation, motion parameter estimation is realized using offset fitting, and the amount of calculation is small;The robustness of algorithm is improved using the characteristics that multiple frequency points are not easily disturbed simultaneously, and the robustness of high-resolution velocity estimation is improved using the motion identity of multiple special points.

Description

Technical Field

[0001] This invention relates to the field of radar technology, and specifically to a robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method. Background Technology

[0002] Compared to optical imaging, microwave imaging offers advantages such as all-weather, all-day, and long-range detection, playing an increasingly important role in military applications. ISAR imaging has the capability to image moving targets, obtaining a two-dimensional scattering intensity distribution image of the target of interest, making target identification easier. The range resolution of ISAR imaging is determined by the bandwidth of the radar system; to achieve high range resolution, a wide operating bandwidth is necessary. However, older radar equipment lacks the capability for instantaneous broadband operation; neither the data transmission bandwidth nor the bandwidth of the microwave components can meet the demands of large instantaneous bandwidth operation. On the other hand, stepped-frequency systems, by sequentially transmitting signals with a certain frequency distribution and synthesizing the frequencies during signal processing, achieve a large broadband signal, thereby obtaining a high-resolution one-dimensional range image of the target. This reduces the difficulty and cost of broadband radar system design and has gained widespread attention and application.

[0003] In engineering applications, chirp subpulse frequency-stepped signals combine the advantages of both chirp and frequency stepping. They can increase the time-bandwidth product of narrowband subpulses to achieve greater bandwidth and system power, while reducing the number of step cycles while maintaining the pulse width and total bandwidth of the frequency-stepped transmitted signal. The essence of this type of system is to distribute a wideband frequency across the time dimension using diversity. To ensure that the diversified signals can be reassembled into a wideband signal, coherence between the subpulses is required. However, for targets moving relative to the radar, the distance changes caused by the motion alter the phase of each subpulse, disrupting the coherence between them. Therefore, inter-subpulse compensation is necessary for moving targets. The estimation and compensation of motion characteristics become crucial for chirp radar imaging.

[0004] Current velocity estimation methods mainly include: estimating velocity using positional changes from multiple frames of imaging data, including range differentiation, frequency domain correlation, time domain cross-correlation, and positive / negative frequency modulation. Since target motion causes different imaging positions in different frames, these methods estimate velocity using the relationship between preceding and following positions; the accuracy varies depending on the method used; velocity search algorithms, which generally have high estimation accuracy but require significant processing time; using multiple radar systems for velocity estimation, which requires switching between different systems and demands high compatibility with the radar system; and composite velocity measurement methods, which effectively combine the advantages of the above methods, easily meeting system requirements. However, none of these methods consider interference issues and crosstalk between radars in the real environment, thus their robustness needs improvement. Therefore, a robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method is proposed. Summary of the Invention

[0005] The technical problem this invention aims to solve is how to address the shortcomings of existing velocity estimation methods, such as the failure to consider interference in the actual environment and the impact of inter-radar crosstalk, which hinders robustness. This invention provides a robust frequency-modulated step-frequency (FM-PM) ISAR imaging radar velocity estimation method. First, based on the coarse-resolution envelope similarity of pulses at the same frequency and the consistency of motion characteristics at multiple frequency points, velocity estimation and compensation are performed to align the coarse-resolution cells at each frequency point, resolving the problem of coarse-resolution cell crossing caused by motion. Then, using high-resolution range images obtained by IFFT of multiple coarse-resolution characteristic point cells, high-precision inter-frame residual velocity estimation is performed. This avoids the impact of high-resolution image distortion caused by fast-moving targets crossing range cells on velocity estimation, while improving the algorithm's stability through multiple estimations.

[0006] The present invention solves the above-mentioned technical problems through the following technical solution, and the present invention includes the following steps:

[0007] S1: Construct a multi-frame frequency modulation step frequency data model, and the distance sampling unit of a single frequency point becomes a coarse resolution distance unit;

[0008] S2: Coarse-precision velocity estimation and compensation are achieved by using the same frequency offset estimation of multi-frequency points;

[0009] S3: High-resolution imaging of the coarse-resolution range cell containing the prominent point is obtained by IFFT of the intra-frame sub-pulse dimension, and high-precision velocity estimation and compensation are achieved by using the high-resolution image of multi-point synthesis.

[0010] S4: ISAR imaging is achieved through inter-frame FFT.

[0011] Furthermore, in step S1, the specific processing procedure is as follows:

[0012] S11: A frame is composed of M sub-pulses with a step size of Δf. N frames of echoes are used for ISAR imaging. Each step-frequency sub-pulse is a linear frequency modulated signal. The baseband echo signal of the m-th linear frequency modulated sub-pulse in the n-th frame received by the radar system is:

[0013]

[0014] Among them, A m Let K be the echo amplitude of the m-th sub-pulse, rect[] be the rectangular window function, and K be the echo amplitude of the m-th sub-pulse. r T is the frequency modulation slope of the sub-pulse. r The time interval between sub-pulses is T, and the pulse width is T. The delay of the target echo is given by f0, where f0 is the starting frequency.

[0015] S12: Perform pulse compression on the echo of the m-th sub-pulse to obtain the compressed sub-pulse signal:

[0016]

[0017] Where B = K r T is the bandwidth of the linear frequency modulated signal. Envelope modulation representing pulse compression

[0018] S13: For a moving target, the distance R(mT) of the m-th sub-pulse is... r Performing a Taylor expansion, we obtain the distance change R(mT) caused by the velocity v. r )=R(0)+v*mT r +0(mT r If the distance corresponds to the phase term of the echo signal, then the phase term is represented as follows:

[0019]

[0020] Furthermore, in step S2, the specific processing procedure is as follows:

[0021] S21: Estimating the offset using the envelope of echoes at the same frequency:

[0022]

[0023] Where, x r,i,m =IFFT[FFT(|x r,m |)·FFT(|x i,m |) * ], FFT and IFFT represent velocity Fourier transform and inverse transform, * represents conjugate, and the echo x at the m-th frequency point of the r-th frame. r,m The envelope of the reference pulse, the echo x at the m-th frequency point of the i-th frame. i,m To process the pulse envelope, xr,i,m For cross-correlation envelopes, This indicates the cell number where the maximum value is obtained;

[0024] S22: Construct the offset vector sequence for a single frequency point:

[0025]

[0026] Where T represents transpose;

[0027] S23: Construct offset matrices for multiple frequency points:

[0028]

[0029] In this offset matrix, the columns correspond to the frequency point number, and the rows correspond to the frame number;

[0030] S24: Constructing offset estimates for multi-frequency synthesis:

[0031]

[0032] in, Med represents the median value;

[0033] S25: Velocity and acceleration estimates are obtained through fitting:

[0034]

[0035] Where t = nMT r The sampling time for each frame;

[0036] S26: Obtain the distance offset for each pulse and perform distance alignment. The distance offset for each pulse is as follows:

[0037] R(n,m)=R(0)+v c (nM+m)T r +0.5a c ((nM+m)T r ) 2 ;

[0038] S27: Constructing a compensated phase:

[0039]

[0040] Where, v(n) = v c +a c nMT r For the estimated velocity of the nth frame, R(n) = R(0) + v c nMT r +0.5a c (nMT r) 2 Let the initial distance be the distance in the nth frame, and follow the... Provide compensation;

[0041] Furthermore, in step S3, the specific processing procedure is as follows:

[0042] S31: Obtain the distance distribution of scattering points between two adjacent frames:

[0043]

[0044] Where t=kT fs T fs Where k is the distance sampling frequency, and k is a positive integer representing the index of the coarse-resolution distance cell;

[0045] S32: Construct the set of locations of the highlighted points:

[0046] Π={i1,i2,...,i p ,...i P}

[0047] Among them, i p Indicates the distance cell number of the prominent point;

[0048] S33: Obtain the offset estimate for each point in the set of highlighted points;

[0049] S34: For offset vector The elements are sorted, and the average of the middle set of values ​​is calculated to obtain the estimated inter-frame offset s(n).

[0050] S35: Process all frames using steps S31 to S34 to obtain the inter-frame offset estimate S = [s(1),...,s(N-1)];

[0051] S36: Obtain the overall offset through integration.

[0052] S37: High-precision velocity and acceleration parameters are obtained through fitting:

[0053] E(n)≈v p t+0.5a p t 2

[0054] Where t = nMT r The sampling time for each frame;

[0055] S38: Use step S26 to obtain the offset R(n,m) = v for each sub-pulse. p (nM+m)T r +0.5a p ((nM+m)Tr ) 2 Then the sub-pulse signal is transferred to the frequency domain, and in the frequency domain it passes through... After compensation, the process returns to its original state, achieving envelope compensation.

[0056] S39: Obtain the phase term of the offset using step S27. pass Compensation will be provided.

[0057] Furthermore, in step S33, the specific processing procedure is as follows:

[0058] S331: Obtain a high-resolution image of the coarse-resolution distance cell containing the prominent point through IFFT in the sub-pulse dimension.

[0059] S332: Obtain the offset between two frames through envelope similarity:

[0060]

[0061] in, That is, the offset is obtained by the position of the maximum value of the cross-correlation;

[0062] S333: By processing all points in the set of highlighted points using steps S331 and S332, construct the offset vector between the (n+1)th frame and the nth frame.

[0063] Furthermore, in step S4, the specific processing procedure is as follows:

[0064] S41: Stitch together the high-resolution images of multiple coarse-resolution range units within a frame to obtain a high-resolution range image of a single frame.

[0065] S42: ISAR imaging is achieved by performing inter-frame Fourier transforms on the high-resolution range images of all frames.

[0066] Furthermore, in step S41, the specific processing procedure is as follows:

[0067] S411: Perform IFFT on each coarse-resolution range within a frame in the sub-pulse dimension to obtain a high-resolution range image for each coarse-resolution unit. Where n represents the nth frame and i represents the i-th coarse resolution distance unit;

[0068] S412: Arrange the high-resolution range images according to the coordinate positions of the coarse-resolution cells to obtain the high-resolution one-dimensional range image of the frame.

[0069] Compared with the prior art, the present invention has the following advantages: This robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method uses a two-step approach to gradually estimate and compensate for velocity. First, a coarse estimation is used to estimate and compensate for large motion quantities, and then a high-precision residual motion estimation and compensation is achieved by synthesizing high resolution. The method uses envelope cross-correlation to estimate offsets and then uses offset fitting to estimate motion parameters, resulting in low computational load. The robustness of the algorithm is improved by utilizing the characteristic that multiple frequency points are not easily interfered with simultaneously. The robustness of high-resolution velocity estimation is improved by utilizing the motion identity of multiple display points. Attached Figure Description

[0070] Figure 1 This is a flowchart illustrating the robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method in this embodiment of the invention.

[0071] Figure 2 This is a schematic diagram of the coarse velocity estimation process in an embodiment of the present invention;

[0072] Figure 3 This is a target echo change diagram after pulse compression in an embodiment of the present invention;

[0073] Figure 4 This is a response diagram of the target in different frames of a pulse with the same frequency in an embodiment of the present invention;

[0074] Figure 5 This is a velocity map obtained from coarse velocity estimation in an embodiment of the present invention;

[0075] Figure 6 This is a flowchart illustrating the precise velocity estimation process in an embodiment of the present invention.

[0076] Figure 7 This is a velocity map obtained by precise velocity estimation in an embodiment of the present invention;

[0077] Figure 8 This is an ISAR imaging result of the target in an embodiment of the present invention. Detailed Implementation

[0078] The embodiments of the present invention are described in detail below. These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the following embodiments.

[0079] like Figure 1 As shown, this embodiment provides a technical solution: a robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method, including the following steps:

[0080] Step (1): Construct a multi-frame frequency modulation step frequency data model

[0081] In this embodiment, the specific processing procedure of step (1) is as follows:

[0082] Step (1a): A frame is composed of M sub-pulses with a step size of Δf. N frames of echoes are used for ISAR imaging. Each step-frequency sub-pulse is a linear frequency modulated signal. The baseband echo signal of the m-th linear frequency modulated sub-pulse in the n-th frame received by the radar system is:

[0083]

[0084] Among them, A m Let K be the echo amplitude of the m-th sub-pulse, rect[] be the rectangular window function, and K be the echo amplitude of the m-th sub-pulse. r T is the frequency modulation slope of the sub-pulse. r The time interval between sub-pulses is T, and the pulse width is T. The delay of the target echo is given by f0, where f0 is the starting frequency.

[0085] In this embodiment, Δf is 30MHz, M is 16 (i.e., 16 frequency points), N is 170, and there are a total of 2720 pulses, with 1024 sampling points per pulse.

[0086] Step (1b): Perform pulse compression on the echo of the m-th sub-pulse to obtain the compressed sub-pulse signal:

[0087]

[0088] Where B = K r T is the bandwidth of the linear frequency modulated signal. This represents envelope modulation for pulse compression. For example... Figure 3 The image shows the echo distribution of each pulse after the target pulse compression.

[0089] Step (1c): For a moving target, the distance R(mT) of the m-th sub-pulse is... r Performing a Taylor expansion, we obtain the distance change R(mT) caused by the velocity v. r )=R(0)+v*mT r +0(mT r If the distance corresponds to the phase term of the echo signal, then the phase term can be expressed as:

[0090]

[0091] Step (2): Use the multi-frequency integrated same-frequency offset estimation to achieve coarse-precision velocity estimation and compensation. The process is as follows: Figure 2 As shown.

[0092] In this embodiment, the specific processing procedure of step (2) is as follows:

[0093] Step (2a): As Figure 4 For target echoes at the same frequency but different frames, their envelopes are very similar. Therefore, offset estimation is performed using the envelope of the echo at the same frequency. Where, x r,i,m =IFFT[FFT(|x r,m |)·FFT(|x i,m |) * ], FFT and IFFT represent velocity Fourier transform and inverse transform, * represents conjugate, and the echo x at the m-th frequency point of the r-th frame. r,m The envelope of the reference pulse, the echo x at the m-th frequency point of the i-th frame. i,m To process the pulse envelope, x r,i,m For cross-correlation envelopes, This indicates the cell number where the maximum value is obtained. In this embodiment, the offset is x. r,i,m The offset corresponding to the maximum value.

[0094] Step (2b): Construct the offset vector sequence for a single frequency point: Where T represents transpose.

[0095] Step (2c): Construct offset matrices for multiple frequency points: The columns correspond to the frequency point numbers, and the rows correspond to the frame numbers.

[0096] Step (2d): Construct the offset estimate for multi-frequency point synthesis:

[0097]

[0098] in, Med represents the median value.

[0099] Step (2e): Obtain velocity and acceleration estimates through fitting:

[0100]

[0101] Where t = nMT r For the sampling time of each frame, such as Figure 5 To estimate the distribution of the obtained velocity values.

[0102] Step (2f): Obtain the distance offset for each pulse:

[0103] R(n,m)=R(0)+v c (nM+m)T r +0.5a c ((nM+m)T r ) 2

[0104] And perform distance alignment.

[0105] Step (2g): Constructing the compensated phase:

[0106]

[0107] Where, v(n) = v c +a c nMT r For the estimated velocity of the nth frame, R(n) = R(0) + v c nMT r +0.5a c (nMT r ) 2 Let the initial distance be the distance in the nth frame, and follow the... Compensation will be provided.

[0108] Step (3): High-precision velocity estimation and compensation are achieved using high-resolution images obtained from multi-point generalization. The process is as follows: Figure 6 As shown.

[0109] In this embodiment, the specific processing procedure of step (3) is as follows:

[0110] Step (3a): Obtain the distance distribution of scattering points between two adjacent frames:

[0111]

[0112] Where t=kT fs T fs is the distance sampling frequency, k is a positive integer, and represents the index of the coarse resolution distance cell.

[0113] Step (3b): Construct the set of locations of the highlighted points:

[0114] Π={i1,i2,...,i p ,...i P}

[0115] Among them, i p Indicates the distance unit number of the prominent point.

[0116] Step (3c): Obtain the offset estimate for each point in the set of highlighted points.

[0117] In this embodiment, the specific processing procedure of step (3c) is as follows:

[0118] Step (3c1): Obtain the high-resolution image of the coarse-resolution distance cell containing the prominent point through IFFT in the sub-pulse dimension.

[0119] Step (3c2): Obtain the offset between two frames through envelope similarity:

[0120]

[0121] in, That is, the offset is obtained by the position of the maximum value of cross-correlation.

[0122] Step (3c3): By processing all points in the set of highlighted points using steps (3c1) and (3c2), construct the offset vector between the (n+1)th frame and the nth frame.

[0123] Step (3d): For the offset vector The elements are sorted, and the average of a certain number of the middle values ​​is calculated to obtain the estimated inter-frame offset s(n).

[0124] Step (3e): Process steps (3a)-(3d) through all frames to obtain the inter-frame offset estimate S = [s(1),...,s(N-1)].

[0125] Step (3f): Obtain the overall offset through integration.

[0126] Step (3g): Obtain high-precision velocity and acceleration parameters through fitting:

[0127] E(n)≈v p t+0.5a p t 2

[0128] Where t = nMT r For the sampling time of each frame, such as Figure 7 The figure shows the results of the precise velocity estimation.

[0129] Step (3h): Use step (2f) to obtain the offset R(n,m) = v for each sub-pulse. p (nM+m)T r +0.5a p ((nM+m)T r ) 2 Then the sub-pulse signal is transferred to the frequency domain, and in the frequency domain it passes through... After compensation, the process returns to its original state, achieving envelope compensation.

[0130] Step (3i): Obtain the phase term of the offset using step (2g). pass Compensation will be provided.

[0131] Step (4): Obtain high-resolution imaging of each coarse-resolution range cell through intra-frame sub-pulse dimension IFFT, and then achieve ISAR imaging through inter-frame FFT. For example... Figure 8 As shown.

[0132] In this embodiment, the specific process of step (4) is as follows:

[0133] Step (4a): Stitch together the high-resolution images of multiple coarse-resolution range units within a frame to obtain a high-resolution range image of a frame.

[0134] Step (4b): Perform inter-frame Fourier transform using the high-resolution range images of all frames to achieve ISAR imaging.

[0135] In step (4a), each coarse-resolution range within a frame is subjected to IFFT in the sub-pulse dimension to obtain a high-resolution range image for each coarse-resolution unit. Where n represents the nth frame and i represents the i-th coarse resolution distance cell; then, the high-resolution distance image is arranged according to the coordinate position of the coarse resolution cell to obtain the high-resolution one-dimensional distance image of the frame.

[0136] In summary, the robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method described in the above embodiments utilizes a two-step approach to progressively estimate and compensate for velocity. First, a coarse estimation is used to estimate and compensate for large motion quantities, and then a high-precision residual motion estimation and compensation is achieved by synthesizing high resolution. The method employs envelope cross-correlation to estimate offsets, and then uses offset fitting to estimate motion parameters, resulting in low computational complexity. The robustness of the algorithm is enhanced by leveraging the characteristic that multiple frequency points are less susceptible to simultaneous interference. Furthermore, the robustness of high-resolution velocity estimation is improved by utilizing the motion uniformity of multiple display points.

[0137] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method, characterized in that, Includes the following steps: S1: Construct a multi-frame frequency modulation step frequency data model, and the distance sampling unit of a single frequency point becomes a coarse resolution distance unit; S2: Coarse-precision velocity estimation and compensation are achieved by using the same frequency offset estimation of multi-frequency points; S3: High-resolution imaging of the coarse-resolution range cell containing the prominent point is obtained by IFFT of the intra-frame sub-pulse dimension, and high-precision velocity estimation and compensation are achieved by using the high-resolution image of multi-point synthesis. S4: ISAR imaging is achieved through inter-frame FFT; In step S3, the specific processing procedure is as follows: S31: Obtain the distance distribution of scattering points between two adjacent frames: ； in, , For distance sampling frequency, A positive integer representing the index of the coarse-resolution distance cell; S32: Construct the set of locations of the highlighted points: ； in, Indicates the distance cell number of the prominent point; S33: Obtain the offset estimate for each point in the set of highlighted points; S34: For offset vector The elements are sorted, and the average of a predetermined number of the middle values ​​is calculated to obtain an estimate of the inter-frame offset. ; S35: Process all frames using steps S31 to S34 to obtain the inter-frame offset estimate. ; S36: Obtain the overall offset through integration. ; S37: High-precision velocity and acceleration parameters are obtained through fitting: ； in, The sampling time for each frame; S38: Obtain the offset of each sub-pulse using step S26. Then the sub-pulse signal is transferred to the frequency domain, and in the frequency domain it passes through... After compensation, the process returns to its original state, achieving envelope compensation. S39: Obtain the phase term of the offset using step S27. ,pass Compensation will be provided.

2. The robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method according to claim 1, characterized in that: In step S1, the specific processing procedure is as follows: S11: Using step mode of Each sub-pulse makes up a frame. ISAR imaging is performed using frame echoes, with each step frequency sub-pulse being a linear frequency modulated signal. The radar system receives the first frame echo. Frame number The baseband echo signal of each linear frequency modulated sub-pulse is: ； in, For the first The echo amplitude of each sub-pulse For rectangular window functions, The frequency modulation slope of the sub-pulse. The time interval of the sub-pulses. The pulse width. For the delay of the target echo, The starting frequency; S12: For the first The echoes of the sub-pulses are compressed to obtain the compressed sub-pulse signal: ； in, , where is the bandwidth of the linear frequency modulated signal. , representing envelope modulation of pulse compression S13: For a moving target, the first... Distance between subpulses Perform a Taylor expansion to obtain the velocity. The resulting change in distance The phase term of the echo signal corresponding to the distance is represented as follows: 。 3. The robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method according to claim 2, characterized in that: In step S2, the specific processing procedure is as follows: S21: Estimating the offset using the envelope of echoes at the same frequency: ； in, FFT and IFFT represent velocity Fourier transform and inverse transform, respectively, * denotes conjugate, and the echo at frequency m in the r-th frame. The echo at frequency m in the i-th frame is the envelope of the reference pulse. To process the pulse envelope, For cross-correlation envelopes, This indicates the cell number where the maximum value is obtained; S22: Construct the offset vector sequence for a single frequency point: ； Where T represents transpose; S23: Construct offset matrices for multiple frequency points: ； In this offset matrix, the columns correspond to the frequency point number, and the rows correspond to the frame number; S24: Constructing offset estimates for multi-frequency synthesis: ； in, , Indicates taking the median value; S25: Velocity and acceleration estimates are obtained through fitting: ； in, The sampling time for each frame; S26: Obtain the distance offset for each pulse and perform distance alignment. The distance offset for each pulse is as follows: ； S27: Constructing a compensated phase: ； in, The estimated velocity for the nth frame. Let the initial distance be the distance in the nth frame, and follow the... Compensation will be provided.

4. The robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method according to claim 3, characterized in that: In step S33, the specific processing procedure is as follows: S331: Obtain a high-resolution image of the coarse-resolution distance cell containing the prominent point through IFFT in the sub-pulse dimension. , ; S332: Obtain the offset between two frames through envelope similarity: ； in, That is, the offset is obtained by the position of the maximum value of the cross-correlation; S333: By processing all points in the set of highlighted points using steps S331 and S332, construct the offset vector between the (n+1)th frame and the nth frame. .

5. The robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method according to claim 4, characterized in that: In step S4, the specific processing procedure is as follows: S41: Stitch together the high-resolution images of multiple coarse-resolution range units within a frame to obtain a high-resolution range image of a single frame. S42: ISAR imaging is achieved by performing inter-frame Fourier transforms on the high-resolution range images of all frames.

6. The robust frequency-modulated step-frequency ISAR imaging radar velocity estimation method according to claim 5, characterized in that: In step S41, the specific processing procedure is as follows: S411: Perform IFFT on each coarse-resolution range within a frame in the sub-pulse dimension to obtain a high-resolution range image for each coarse-resolution unit. , where n represents the nth frame and i represents the i-th coarse resolution distance unit; S412: Arrange the high-resolution range images according to the coordinate positions of the coarse-resolution cells to obtain the high-resolution one-dimensional range image of the frame.

Images