A GNSS external radiation source radar moving target imaging method based on piecewise quadratic accumulation

Abstract

The application is a GNSS external source radar moving target imaging method based on segmented quadratic accumulation, which solves the problem of weak GNSS external source radar echo signal and long time accumulation. The method comprises the following steps: performing distance frequency domain matching filtering on the echo signal; segmenting the signal in the azimuth direction according to the criterion that the distance migration and Doppler shift of the target can be ignored in the sub-section; performing azimuth FFT on each sub-section signal to complete the first focusing in the section; performing parameter search, completing the second coherent accumulation through the phase compensation and accumulation between the sub-section signals; performing IFFT on the obtained three-dimensional parameter matrix along the distance direction; performing threshold processing in the three-dimensional parameter space to estimate the Doppler parameter of the moving target; and performing motion compensation and moving target imaging by using the estimated optimal Doppler parameter. The method has high implementation efficiency, small performance loss compared with the prior art, good parameter estimation performance and good final moving target imaging effect.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of radar signal processing, and particularly relates to a GNSS external source radar moving target imaging method based on segmented quadratic accumulation. BACKGROUND

[0002] Traditional active radar refers to illuminating a target by radiating electromagnetic waves to detect, locate and track the target. External source radar refers to a bistatic or multistatic radar using an opportunity source as a radar transmitter. The transmission waveform of the external source radar is not controlled by the receiver, and the receiver does not need frequency allocation, and has the advantages of simple equipment, low cost, low power consumption, strong anti-interference ability, good concealment, etc., and is suitable for air monitoring of non-cooperative targets, long-range border monitoring, etc., and has obvious advantages in military and civilian application fields.

[0003] Among various opportunity sources, the global navigation satellite system (GNSS) has unique advantages as a radar source. GNSS satellites provide global coverage and use pseudo-random code spread spectrum modulation, and multiple GNSS satellites (such as GPS, GLONASS, Galileo or Beidou) can be used simultaneously at any location on the earth's surface. In addition, synchronization can be easily achieved with the aid of GNSS timing services, and GNSS external source radar also has the advantages of low power consumption, lightweight and concealed operation. At present, GNSS external source radar has been applied in the field of remote sensing, and mainly divided into two categories, one is GNSS-R, which can be used to invert ocean wind speed, soil moisture, etc., and the other is GNSS-based bistatic radar application, such as sea-air target detection and imaging.

[0004] The difficulty of signal processing of GNSS-based passive radar mainly lies in the long distance from the transmitter, which causes low power density of ground signals. Generally, a large receiving antenna and a relatively long coherent accumulation time are needed to obtain sufficient signal-to-noise ratio (SNR) for effective signal detection. However, the movement of the target will cause the range migration and Doppler frequency shift of the echo signal within the accumulation time, thereby limiting the accumulation time. The long-time coherent accumulation of weak targets can be realized by the pre-focusing method under the condition of unknown target motion parameters, and the most commonly used method is Radon Fourier Transform (RFT). The RFT realizes long-time accumulation of target echoes with significant range migration through joint search of target range and velocity, but it cannot handle the Doppler shift problem. A related scholar proposes a generalized RFT (GRFT), which provides a general method for long-time coherent integration of maneuvering targets. A related scholar proposes a hybrid accumulation scheme for intra-frame coherent accumulation and inter-frame non-coherent accumulation to obtain the final range-Doppler domain (RD domain) image, but the SNR gain obtained by this method will decrease when the accumulation time becomes longer. An improved RFT (MRFT) method is also proposed by a related scholar, which increases the Doppler rate search step compared with RFT, and can not only handle the range shift, but also handle the Doppler shift problem, significantly increasing the accumulation time. Compared with GRFT, MRFT avoids the sample update step of searching the Doppler rate each time. However, the computational complexity of this method is still high, and the processing efficiency will decrease as the total signal accumulation time becomes longer and the parameter search range expands. A related scholar proposes a segmented processing idea, which searches the target velocity within the segment and searches the target velocity and acceleration between the segments, thereby reducing the computational load. However, this method involves two parameter searches, which affects the implementation efficiency of the moving target imaging algorithm. SUMMARY

[0005] The present application is aimed at the problem of weak GNSS-based passive radar echo signals and long-time accumulation, and proposes a GNSS-based passive radar moving target imaging method based on segmented secondary accumulation. Compared with the optimal MRFT accumulation scheme, the method ensures that the SNR loss in the parameter space is acceptable while improving the efficiency of the moving target imaging algorithm. Compared with the scheme of searching parameters within and between segments, the present application only searches parameters once, thereby improving the processing efficiency of the moving target imaging algorithm. Finally, motion parameter compensation can be used to realize range-Doppler domain imaging of moving targets based on GNSS signals.

[0006] The GNSS-based passive radar moving target imaging method based on segmented secondary accumulation of the present application comprises the following steps:

[0007] Step one, read in the echo signal and system-related parameters, and perform distance-frequency domain matched filtering on the echo signal;

[0008] Step 2: Based on the criterion that the target's range migration and Doppler frequency shift within the sub-segment can be ignored, the echo signal processed in Step 1 is segmented in the azimuth direction.

[0009] Step 3: Perform azimuth FFT (Fast Fourier Transform) on each sub-segment signal to complete one focusing within the segment. Here, it is assumed that the range migration and Doppler frequency shift within the sub-segment are negligible.

[0010] Step 4: Perform parameter search, and complete the secondary coherent accumulation through inter-segment phase compensation and accumulation of sub-segment signals;

[0011] The search parameters include the detection range R0 and the target Doppler frequency f. d And the target Doppler modulation frequency f r Suppose that in a certain search, the target Doppler frequency and the Doppler modulation frequency are f0 and f1 respectively. d_temp f r_temp Extract f from each sub-segment signal d_temp The corresponding rows are concatenated into a matrix Sc, with dimensions N×Nr, where N is the number of sub-segments and Nr is the number of range vectors. Phase compensation is performed on the concatenated matrix Sc, and the results are accumulated to obtain the three-dimensional parameter matrix Spara(f τ ,f d ,f r );

[0012] Step 5: For the three-dimensional parameter matrix Spra(f) τ ,f d ,f r Perform an inverse fast Fourier transform (IFFT) along the range direction to transform the data into the three-dimensional parameter domain Spara(R0, f). d ,f r );

[0013] Step 6: Perform threshold processing in the three-dimensional parameter space to estimate the Doppler parameters of the moving target;

[0014] Step 7: Use the estimated optimal Doppler parameters to perform motion compensation and moving target imaging.

[0015] In step two, the segmentation criterion is to ignore the target's range migration and Doppler frequency shift within the sub-segment. Let the duration of the segment after segmentation be t. sub Then the following conditions are satisfied:

[0016]

[0017] Where λ represents the signal wavelength, f d f represents the Doppler frequency of the target. r The value represents the Doppler modulation frequency of the target, and c represents the speed of light.

[0018] The step three is to perform azimuth FFT on the sub-segment signal to obtain a signal s(f η ,f τ ) as follows:

[0019]

[0020] At this time, the intra-segment signal completes a focusing, and the signal is in the range frequency domain f τ , the azimuth Doppler domain f η ; A m is the signal amplitude, f c is the carrier frequency of the radar system, P(·) is the Fourier transform of the autocorrelation function of the navigation signal pseudo-random code, R0 represents the detection distance, and φ is the residual phase. When f η is equivalent to the target Doppler frequency f d , the signal s reaches the maximum value.

[0021] The step four is to perform parameter search, and the parameter search range is as follows:

[0022]

[0023] wherein R min represents the minimum detection distance, R max represents the maximum detection distance, v min represents the minimum target speed, and v max represents the maximum target speed.

[0024] The step four is to perform phase compensation on the splicing matrix Sc, and the phase compensation factor Hc of each row is as follows:

[0025]

[0026] wherein f c is the signal carrier frequency of the radar system; T is the azimuth time of the splicing matrix, and the time interval is the sub-segment duration t sub .

[0027] The signal s nsub extracted from the nth sub-segment signal is as follows:

[0028]

[0029] The three-dimensional parameter matrix is as follows:

[0030] The step seven includes:

[0031] 7.1) completing the range migration correction in the echo range frequency domain, and the compensation factor is as follows:

[0032]

[0033] where H rcm is the Doppler compensation factor at the range frequency, rcmis the range migration, f d _opis the optimal target Doppler frequency; t a is the azimuth time, the interval is the pulse repetition time.

[0034] 7.2) After the range IFFT, the azimuth compensation is performed, and finally the azimuth FFT is performed to realize the moving target imaging. The azimuth compensation factor H a is:

[0035] H a = exp(-jπf r _op·t a 2 )

[0036] The final imaging result Srd(r,fd) is:

[0037] Srd(r,fd) = aFFT(rIFFT(s(r,f τ )·H rcm )·H a )

[0038] where f r _opis the optimal target Doppler frequency, aFFT is the azimuth fast Fourier transform, and rIFFT is the range inverse fast Fourier transform.

[0039] Compared with the prior art, the present application has the advantages of:

[0040] (1) The performance loss of the method of the present application is small. Through appropriate segmentation criteria, the range migration and Doppler frequency shift are ignored within the segment, one accumulation is completed, and the second phase accumulation is realized by phase compensation between segments. Compared with MRFT, the performance loss is very low, the parameter estimation performance is good, and the final moving target imaging effect is good.

[0041] (2) The method of the present application has high implementation efficiency. On the one hand, all operations of the present application are performed in the range frequency domain, which can be quickly realized by using FFT operation. On the other hand, the present application realizes the second accumulation within and between segments, and only needs to search for the range parameter, the Doppler frequency parameter and the Doppler frequency parameter between segments, and only needs to search for the parameter once, which greatly reduces the calculation amount of the moving target Doppler parameter estimation and improves the implementation efficiency of the radar moving target imaging. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is the flowchart of the GNSS external radiation source radar moving target imaging method based on the segmented second accumulation of the embodiment of the present application;

[0043] Figure 2 is a result diagram of echo complete distance direction matched filtering of the embodiment of the application;

[0044] Figure 3 is a three-dimensional parameter search result diagram based on the MRFT method;

[0045] Figure 4 is a three-dimensional parameter search result diagram based on the segmented secondary accumulation method of the application;

[0046] Figure 5 is a direct azimuth FFT distance Doppler domain result diagram;

[0047] Figure 6 is a comparison diagram of imaging results of the method of the application and the MRFT method in the distance dimension;

[0048] Figure 7 is a comparison diagram of imaging results of the method of the application and the MRFT method in the Doppler dimension. DETAILED DESCRIPTION

[0049] The application will be further described in detail below in combination with the drawings and embodiments.

[0050] In view of the characteristics that the GNSS external illuminating source radar target signal is weak and needs long time accumulation, the application is based on a segmented secondary accumulation scheme, one focusing is completed in a subsegment through azimuth FFT (fast Fourier transform), secondary accumulation is completed through phase compensation and accumulation between segments, a three-dimensional parameter space is constituted, the optimal Doppler parameter of a moving target is estimated, and finally, the echo signal is subjected to motion compensation and imaging. The effectiveness of the method of the application is illustrated by processing the collected real data by using the method of the application through MATLAB software through an embodiment. The corresponding measured data parameters of the embodiment are shown in Table 1.

[0051] Table 1 Measured data parameters

[0052] Parameter Value Satellite GPS PRN1 L5 signal Total integration time 3s Signal carrier 1176.45 MHz Sampling rate 62 MHz Equivalent pulse repetition frequency 1000 Hz Signal bandwidth 10.23 MHz

[0053] The GNSS external illuminating source radar moving target imaging method based on segmented secondary accumulation of the embodiment of the application specifically includes the following steps as shown in Figure 1

[0054] Step one, read in original echo data and system related parameters, and complete distance frequency domain matched filtering.

[0055] 1) After the echo signal is demodulated to an intermediate frequency, pulse compression is completed through self-correlation processing with a navigation ranging code.

[0056] ​2) Data demodulation is performed to remove the random phase jumps in the echo. The echo data is then rearranged into a two-dimensional matrix, which is denoted as the range and azimuth directions, respectively.

[0057] 3) Precise satellite position information can be obtained from the satellite ephemeris information. The satellite motion compensation is completed for the echo signal based on the range migration and Doppler phase shift of the direct channel. Figure 2 The processing result of this step is shown, and the echo signal s(η,τ) is obtained as follows:

[0058]

[0059] wherein η represents the azimuth time, τ represents the range time, A m represents the signal amplitude, λ represents the signal wavelength, p(·) represents the autocorrelation function of the navigation signal pseudo-random code, c represents the speed of light, T a =N a *T0, represents the length of the coherent accumulation, T0 represents the length of the pseudo-random code, N a represents the number of pseudo-random codes of the coherent accumulation, j is the imaginary unit, rect represents the rectangular function, R(η) represents the sum of the distance from the GNSS satellite to the target and the distance from the target to the receiver, which is denoted as the total distance, R ref (η) represents the distance between the GNSS satellite and the receiver, which is denoted as the reference distance. The difference between the total distance and the reference distance R * (η) is:

[0060] R * (η) = R(η) - R ref (η) ≈ R0- λf d η- 0.5λf r η 2 (2)

[0061] wherein f d and f r represent the Doppler frequency and the Doppler frequency rate of the target, respectively, and R0 represents the detection distance.

[0062] 4) The echo signal shown in equation (1) is transformed into the range frequency domain, and equation (2) is substituted into equation (1) to prepare for subsequent processing, and the following is obtained:

[0063]

[0064] wherein f τ is the range frequency domain variable, and P(·) is the Fourier transform of p(·) in equation (1). f c is the signal carrier frequency of the radar system, which determines the working wavelength.

[0065] Step two, segment the echo signal in azimuth dimension according to the rule that the target's range migration and Doppler shift in the sub-segment can be ignored.

[0066] That is, the segment criterion is R * (η) is considered to be constant in the segment, and the duration of the sub-segment after segmentation is t sub , which satisfies:

[0067]

[0068] Wherein The limit value of the target to be measured is taken respectively, and here it is set according to experience. v represents the speed of the target to be measured, and R represents the distance of the target to be measured. In the embodiment, the target to be measured is a car, and its speed range is considered to be 5 m / s to 15 m / s, and accordingly the limit values of the Doppler frequency f d and the Doppler frequency f r can be calculated.

[0069] Step three, azimuth FFT is performed on each sub-segment signal to complete one focusing in the segment.

[0070] Azimuth FFT is actually an azimuth Doppler filter bank, and after processing, the signal:

[0071]

[0072] At this time, the signal in the segment has completed one focusing, and the signal is in the range frequency domain f τ , the azimuth Doppler domain f η , and φ is the residual phase. When f η takes a value corresponding to the Doppler frequency f d of the target, the signal s reaches the maximum value. Different rows are taken in the azimuth direction, and the focusing results of different Doppler frequency f d values are taken.

[0073] Step four, parameter searching is performed, and through inter-segment phase compensation and accumulation of the sub-segment signal, two times of coherent accumulation is completed. Step four includes the following three sub-steps.

[0074] Step 4.1) The search parameters include the detection distance R0, the Doppler frequency f d of the target, and the Doppler frequency f r of the target. The parameter search range satisfies:

[0075]

[0076] Wherein R min , R max represent the minimum and maximum detection distances respectively, v min , v maxrepresent the minimum and maximum target speed.

[0077] Step 4.2) for a certain parameter value in the parameter traversal, denoted as f d_temp , f r_temp Because the distance dimension traverses the entire distance direction point number N r , which is not listed separately here. Extract f d_temp (n) corresponding to the row in sub-section n and splice it into a row in matrix Sc, with dimensions N x Nr, N representing the total number of N sub-sections. The extracted signal s nsub in sub-section n is:

[0078]

[0079] Where R(f d_temp ) is the detection distance corresponding to the search Doppler frequency value f d_temp , f d is the true Doppler frequency of the target.

[0080] Step 4.3) phase compensation and accumulation are performed on the spliced matrix Sc, and the phase compensation factor Hc for each row is:

[0081]

[0082] Where T is the azimuth time of the spliced matrix, and the time interval is the duration t sub of the sub-section.

[0083] At this time, the three-dimensional parameter matrix Spara(f τ , f d , f r ) is obtained as:

[0084]

[0085] The three dimensions of the above three-dimensional parameter matrix are the distance frequency domain, the Doppler frequency dimension, and the Doppler frequency modulation dimension, respectively.

[0086] Step five, for the above three-dimensional parameter matrix Spara(f τ , f d , f r ), perform IFFT (inverse fast Fourier transform) along the distance dimension to convert to the three-dimensional parameter domain. At this time, the three dimensions of the three-dimensional parameter matrix correspond to the distance time domain, the Doppler frequency dimension, and the Doppler frequency modulation dimension, respectively. At this time, the three-dimensional parameter matrix is:

[0087] Spara(R0, f d , f r ) = rIFFT(Spara(f τ , f d , fr )) (10)

[0088] wherein rIFFT denotes a range vector inverse fast Fourier transform.

[0089] Step six, threshold processing is performed in the three-dimensional parameter space to complete the Doppler parameter estimation of the moving target.

[0090] For the single automobile target in the embodiment, when the taken parameter corresponds to the target R_op,f d _op,f r _op, the three-dimensional parameter space matrix takes a peak value, and the optimal Doppler parameter combination of the target is extracted at this time. That is

[0091] Spara(R0,f d ,f r ) max =Spara(R_op,f d _op,f r _op) (11)

[0092] In the embodiment, f d _op=77Hz, f r _op=-10.47Hz / s can be obtained. Figure 3 and Figure 4 respectively show the three-dimensional matrix diagrams of the optimal MRFT and the segmented quadratic accumulation method of the application, and the diagram is drawn when f d =f d _op. It can be seen that the segmented quadratic accumulation method of the application ignores the range migration and Doppler frequency shift in the segment, and thus an error is introduced, so that the final parameter space image is slightly divergent, but the peak value judgment is still accurate, and the final imaging result is basically not affected.

[0093] Step seven, motion compensation and moving target imaging are completed by using the estimated optimal Doppler parameter. The step includes the following two sub-steps.

[0094] Step 7.1) range migration correction is completed in the echo range frequency domain, and the compensation factor is:

[0095]

[0096] wherein H rcm is a Doppler compensation factor in the range frequency; rcm is a range migration amount, which changes with the change of the azimuth time t a ; t a is the azimuth time, and the interval is the equivalent pulse repetition time 1ms of the GPS signal.

[0097] Step 7.2) After the range IFFT, the azimuth phase compensation is performed, and finally the azimuth FFT is performed to complete the moving target imaging. The azimuth compensation factor is:

[0098] H a = exp(-jπf r _op·t a 2 ) (13)

[0099] The final imaging result Srd(r,f d ) is:

[0100] Srd(r,f d ) = aFFT(rIFFT(s(η,f τ )·H rcm )·H a ) (14)

[0101] Wherein, aFFT is the azimuth fast Fourier transform, rIFFT is the inverse fast Fourier transform in the range, s(η,f τ ) is the distance domain signal output after step one.

[0102] At this point, the GNSS external source radar moving target imaging whole process is completed, Figure 5 The moving target imaging result is given without distance migration and Doppler shift correction, and the azimuth FFT is directly performed, and it can be seen that the target defocus is serious. Figure 6 And Figure 7 The imaging effect comparison of the MRFT method and the segmented secondary accumulation method of the present application is given, Figure 6 is the distance dimension comparison, Figure 7 is the azimuth Doppler dimension comparison result. It can be seen that the parameter accuracy estimated by the secondary accumulation scheme of the present application is quite high, and the imaging effect is close compared with the optimal MRFT method, but the implementation complexity is greatly reduced, thereby verifying the effectiveness of the present application.

[0103] In addition to the technical features described in the specification, they are known to those skilled in the art. The present application omits the description of the known technology. The embodiments described in the above examples also do not represent all embodiments consistent with the present application. Various modifications or variations made by those skilled in the art without creative labor on the basis of the technical solutions of the present application are still within the protection scope of the present application.

Claims

1. A GNSS-based E-SMR moving target imaging method based on piecewise quadratic accumulation, characterized in that, The method comprises the following steps: Step one, the original echo signal is distance frequency domain matched filter, get the conversion to distance frequency domain echo signal s(η,f τ ), wherein η is azimuth time, f τ is distance frequency domain variable; Step two, segmenting the echo signal outputted in step one in the azimuth direction; Wherein, the distance migration and Doppler frequency shift of the target in the sub-section can be ignored as the segmentation criterion, and the duration of the segmented sub-section is t sub Then, the following conditions are met: where λ denotes the signal wavelength, f d denotes the target Doppler frequency, f r denotes the target Doppler frequency, c denotes the speed of light; Step three, azimuth FFT is performed on each sub-segment signal to complete in-segment focusing; the echo signal obtained after in-segment focusing is s(f η ,f τ ), f η is the azimuth Doppler domain, f τ is the distance frequency domain, and FFT is fast Fourier transform; Step four, searching parameters, completing secondary coherent accumulation through inter-segment phase compensation and accumulation of sub-segment signals; Wherein, the search parameters include a detection distance R0, a target Doppler frequency f d , and a target Doppler frequency modulation f r ; assuming that in a certain search, the target Doppler frequency and the Doppler frequency modulation are f d_temp , f r_temp , respectively, the rows corresponding to f d_temp in each sub-segment signal are extracted and spliced into a matrix Sc, the dimension of Sc is N x Nr, N is the number of sub-segment signals, and Nr is the number of distance direction points; phase compensation is performed on the spliced matrix Sc and accumulated to obtain a three-dimensional parameter matrix Spara(f τ , f d , f r ). Step five, inverse fast Fourier transform IFFT is performed on the three-dimensional parameter matrix Spara(f τ ,f d ,f r ) along the distance direction, and conversion is made to the three-dimensional parameter domain Spara(R0,f d ,f r ). Step six, performing threshold processing in three-dimensional parameter space to estimate Doppler parameters of the moving target; Step seven, using the estimated optimal Doppler parameters to perform motion compensation and moving target imaging.

2. The method of claim 1, wherein, The step one comprises: 1.1) after the echo signal is demodulated to intermediate frequency, pulse compression is completed through self-correlation processing with a navigation ranging code; 1.2) data code demodulation is performed to remove random phase jumps in the echo; then the echo signal is rearranged into a two-dimensional matrix, which is respectively denoted as the range direction and the azimuth direction; 1.3) satellite position is obtained through satellite ephemeris information, and satellite motion compensation is performed on the echo signal based on the distance migration and Doppler phase offset of the direct channel; 1.4) the echo signal is converted to the range frequency domain, as follows: Wherein, s(η,f τ A is the echo signal transformed to the range frequency domain. m It is the signal amplitude, T a It is the coherent accumulation time, P(·) is the Fourier transform of the autocorrelation function of the navigation signal pseudo-random code, and f c It is the signal carrier frequency of the radar system, R * (η) represents R(η) and R ref The distance difference (η) is represented by R(η), where R(η) represents the sum of the distance from the GNSS satellite to the target and the distance from the target to the receiver. ref (η) represents the distance between the GNSS satellite and the receiver.

3. The method of claim 1, wherein, In the second step, f is calculated based on the speed limit value of the target d and the limit value of f r wherein v denotes the speed of the target and R denotes the distance of the target.

4. The method according to claim 1 or 2, characterized in that, The step three is to get the signal s(f) after azimuth FFT of the sub-segment signal as follows: η ,f τ ) as follows: At this time, the intra-pulse signal completes a focusing, and the signal is in the range frequency domain f τ , azimuth Doppler domain f η ; A m is the signal amplitude, f c is the carrier frequency of the radar system, P(·) is the Fourier transform of the autocorrelation function of the navigation signal pseudo-random code, R0 represents the detection range, and φ is the residual phase; when f η is equivalent to the target Doppler frequency f d , the signal s reaches the maximum value.

5. The method of claim 1, wherein, In the step four, the parameter search range is as follows: where R min represents the minimum detection distance, R max represents the maximum detection distance, v min represents the minimum target speed, v max represents the maximum target speed.

6. The method according to claim 1 or 5, characterized in that, In the step four, phase compensation is performed on the splicing matrix Sc, and the phase compensation factor Hc of each row is as follows: wherein f c is the signal carrier frequency of the radar system; T is the azimuth time of the stitching matrix, the time interval is the subsegment duration time t sub ; Three-dimensional parameter matrix Signal s extracted from the nth sub-segment signal nsub As follows: where R(f d_temp ) is the corresponding detection range when searching for Doppler frequency f d_temp .

7. The method of claim 1, wherein, The step six, for single target, when the three-dimensional parameter domain matrix Spara(R0,f d ,f r ) takes the peak value, the optimal Doppler parameter R_op,f d _op,f r _op of the target is extracted.

8. The method of claim 1, wherein, The step seven comprises: 7.1) distance migration correction is performed in the range frequency domain, and the compensation factor is as follows: where H rcm is the Doppler compensation factor at range, rcmis the range migration, f d _opis the optimal target Doppler frequency, t a is the azimuth time; 7.2) After the range IFFT, azimuth phase compensation is performed, and finally the azimuth FFT is performed to complete the moving target imaging; wherein the azimuth compensation factor H a is: H a = exp(-jπf r _op·t a 2 ) where f r _op is the optimal target Doppler frequency; Finally, the target imaging result Srd(r, fd) is as follows: Srd(r,fd) = aFFT(r IFFT(s(r,f τ ) · H rcm ) · H a ) Wherein, aFFT is the azimuth direction fast Fourier transform, and rIFFT is the inverse fast Fourier transform in the range direction.

Images