A Subband Decomposition and Long-Time Coherent Integration Method for Weak High-Dynamic Signals

Abstract

The present invention provides a method for long-time coherent integration of sub-band decomposition of weak high-dynamic signals. While using a broadband signal to perform high-precision tracking on a target whose radar information has been intercepted, narrowband signals are constructed by using frequency-domain sub-band decomposition. After coherent synthesis, low-computation high-efficiency search can be performed on targets that have not been intercepted, reducing the long-time coherent integration loss caused by range migration and Doppler spread.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar signal processing, and particularly relates to a method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals. Background Art

[0002] Due to its characteristics of all-weather, all-day, and long-distance operation, radar is the main means of target detection and tracking. Radar undertakes important tasks such as target detection, positioning, surveillance, imaging, recognition, and precise tracking. However, with the breakthrough improvement of the maneuverability and stealth performance of new aerospace attack and defense equipment, the radar detection field is facing technical challenges brought by highly maneuverable, high-speed, and low-observable targets, which makes it of great significance to study the detection technology of weak maneuvering targets.

[0003] Increasing the observation time and adopting long-time accumulation technology can effectively improve the signal-to-noise ratio before detection, and thus improve the detection performance of weak targets. Aiming at the problems of range migration correction and Doppler frequency expansion in the long-time accumulation of weak maneuvering targets, scholars at home and abroad have conducted a large number of studies and achieved relatively rich research results. The existing research results can be mainly divided into two categories. The first category is the cascaded decoupling method, that is, first perform range migration correction on the target echo, and then perform coherent accumulation and parameter estimation in the slow-time dimension. Typical methods for range migration correction include: Keystone transform, Hough transform, Radon transform, coordinate system rotation, etc. Common methods for coherent accumulation and parameter estimation in the slow-time dimension include: de-chirping method (Dechirp), discrete Chirp-Fourier transform, polynomial phase transform, fractional Fourier transform (FRFT), Wigner-Ville distribution, Lv distribution (LVD), etc.

[0004] Another type is the combined coherent accumulation algorithm, which conducts multi-dimensional searches in the target motion parameter space based on the motion model, coherently accumulates the echo signals along the motion trajectory, and simultaneously realizes range migration correction and Doppler frequency migration correction. Typical methods include Radon-FrFT, Radon-LVD, Radon-Fourier Transform, Radon linear canonical transform, etc. The GRFT algorithm proposed by Xu Jia in the paper "Radon-Fourier transform for radar target detection (I): Generalized Doppler Filter Bank" published in IEEE Transactions on Aerospace and Electronic Systems in April 2011 projects the received echo signals into the parameter space composed of the three elements of range, velocity, and acceleration to form a "multi-dimensional focused image" of the target when dealing with uniformly accelerating moving targets, and then jointly compensates for the range migration and Doppler spread of the target envelope to obtain accurate target parameter values.

[0005] In addition, to optimize the computational complexity, Tian Jing et al. proposed a long-time accumulation method based on conjugate of dual-frequency subbands and LVD in the paper "A New Motion Parameter Estimation Algorithm Based on SDFC-LVT" published in IEEE Transactions on Aerospace and Electronic Systems in October 2016. The method reduces the Doppler frequency spread caused by target motion parameters by conjugately multiplying the dual-frequency subband signals, and then conducts LVD processing in the slow time dimension, thereby reducing the computational amount. Li Xiaolong proposed "A High-Speed Target Accumulation Detection Method Based on Frequency-Domain Segmented Processing" in 2022. This method applies frequency-domain segmented processing and the frequency-domain FBRFT algorithm to correct the range walk caused by the target velocity within the segment after pulse compression, realizes the effective detection of high-speed targets, and greatly reduces the computational complexity of echo accumulation.

[0006] The signal processing algorithm for long-time coherent accumulation of weak maneuvering targets is one of the hot research topics in the current radar signal processing field. However, the above methods generally have problems such as high algorithm complexity and difficult engineering implementation. In summary, the range migration correction type of methods mainly correct the range migration caused by velocity and cannot correct the range migration caused by higher-order motion parameters; the methods that achieve low computational complexity through subband decomposition introduce nonlinear operations, resulting in a large signal-to-noise ratio loss; the combined coherent accumulation method requires multi-dimensional joint searches in the target motion parameter space, with a large number of search times and high computational complexity, which is not conducive to engineering applications. Summary of the Invention

[0007] To solve the above problems, the present invention provides a method for long-time coherent integration of weak high-dynamic signal sub-band decomposition, which can effectively reduce the computational complexity, has no signal-to-noise ratio loss, and can achieve coherent integration of weak high-dynamic signals.

[0008] A method for long-time coherent integration of weak high-dynamic signal sub-band decomposition includes the following steps:

[0009] S1: Perform quadrature down-conversion processing on the target echo signal received by the radar to obtain a baseband signal;

[0010] S2: Perform frequency-domain pulse compression on the baseband signal using a frequency-domain matched filter to obtain a frequency-domain matched filtering signal;

[0011] S3: Decompose the frequency-domain matched filtering signal into multiple frequency-domain sub-band signals, and perform inverse Fourier transform on each frequency-domain sub-band signal to obtain multiple time-domain sub-band signals;

[0012] S4: Use an acceleration phase compensation function to compensate the Doppler frequency migration generated by each time-domain sub-band signal in the slow-time dimension to obtain a first compensation signal corresponding to each time-domain sub-band signal;

[0013] S5: Use a velocity phase compensation function to compensate the error generated by each first compensation signal due to the change in Doppler frequency during the inter-segment coherent integration process to obtain a second compensation signal corresponding to each first compensation signal;

[0014] S6: Perform slow-time dimension Fourier transform on each second compensation signal respectively to obtain a coherent integration result corresponding to each time-domain sub-band signal;

[0015] S7: Perform inter-segment coherent synthesis on the coherent integration results corresponding to each time-domain sub-band signal to obtain a coherent integration result corresponding to the target echo signal.

[0016] Further, the frequency-domain matched filtering signal in step S2 is:

[0017]

[0018] where is the total number of maneuvering targets existing within the radar observation range, represents the th backscattering coefficient of the th maneuvering target, is the slow time corresponding to the th pulse echo signal, is the carrier frequency, is the The instantaneous radial distance of a maneuvering target , is the total number of pulse echo signals received by the radar during the observation period is related to the fast time The corresponding fast-time dimension frequency is the radar transmit signal bandwidth is a rectangular pulse with a width of is the speed of light represents the imaginary part

[0019] Furthermore, any one of the time-domain subband signals in step S3 is:

[0020]

[0021] Among them, represents the amplitude accumulation gain of each time-domain subband signal is the pulse width of the radar transmit signal represents the time-domain subband signal bandwidth is the number of frequency-domain subband signals after decomposition, and , represents the th carrier frequency offset of the frequency-domain subband signal

[0022] Furthermore, if it is necessary to eliminate the error generated when the target range migration performs coherent accumulation on each time-domain subband signal, the time-domain subband signal bandwidth and the number of frequency-domain subband signals after decomposition satisfy:

[0023]

[0024]

[0025] Among them, is the upper limit of the radial velocity between the set maneuvering target and the radar is the upper limit of the radial acceleration between the set maneuvering target and the radar is the coherent accumulation time

[0026] Furthermore, the acceleration phase compensation function corresponding to any one of the time-domain subband signals is:

[0027]

[0028] Among them, represents the acceleration search interval represents the acceleration search factor, and satisfies exist Round between Indicates The initial radial acceleration from the maneuvering target to the radar, is the rounding function;

[0029] Multiply the acceleration phase compensation function with the time domain subband signal to obtain the first compensation signal as follows:

[0030]

[0031] in, , Respectively represent The initial radial distance and initial radial velocity between the maneuvering target and the radar, Represents the residual acceleration.

[0032] Furthermore, when the acceleration search factor satisfy When the residual acceleration satisfies , the acceleration search interval satisfies , The wavelength of the radar signal.

[0033] Furthermore, the velocity phase compensation function corresponding to any time domain subband signal is for:

[0034]

[0035] in, is the speed search factor, Search interval for speed;

[0036] Multiply the velocity phase compensation function by the first compensation signal to obtain the second compensation signal as follows:

[0037]

[0038] in, Indicates the residual speed.

[0039] Furthermore, when the speed search factor satisfy When the residual velocity satisfies , the speed search interval satisfies .

[0040] Furthermore, a slow-time Fourier transform is performed on any second compensation signal to obtain the corresponding coherent accumulation result as follows:

[0041]

[0042] Among them, is the slow-time dimension frequency domain corresponding to the relative one.

[0043] Furthermore, the coherent accumulation results corresponding to each time-domain sub-band signal are coherently synthesized between segments, and the coherent accumulation result corresponding to the target echo signal is obtained as follows:

[0044]

[0045] Among them, is the Fourier transform frequency domain corresponding to the relative one.

[0046] Beneficial effects:

[0047] The present invention provides a method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals. While using a broadband signal to perform high-precision tracking on a target whose radar information has been intercepted, narrowband signals are constructed by using frequency-domain sub-band decomposition. After coherent synthesis, low-computation high-efficiency search can be performed on un-intercepted targets, reducing the long-time coherent accumulation loss caused by range migration and Doppler spread. Description of the drawings

[0048] Figure 1 is the flowchart of the present invention.

[0049] Figure 2 is the schematic diagram of sub-band decomposition of the present invention.

[0050] Figure 3 is the result after pulse compression corresponding to different echo pulses of the present invention.

[0051] Figure 4 is the schematic diagram of two-dimensional time-domain comparison before and after sub-band decomposition of the present invention.

[0052] Figure 5 is the schematic diagram of the result after inter-segment coherent accumulation of the present invention.

[0053] Figure 6 is the schematic diagram of the operation complexity result of the present invention.

[0054] Figure 7 is the schematic diagram of the detection performance result of the present invention. Detailed implementation manners

[0055] In order to enable those skilled in the art of the present technology to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.

[0056] The present invention proposes a method for long-time coherent integration of sub-band decomposition of weak high-dynamic signals. While using wideband signals to perform high-precision tracking on targets whose radar information has been intercepted, narrowband signals are constructed by using frequency-domain sub-band decomposition. After coherent synthesis, low-computing-power and efficient search can be carried out on un-intercepted targets, reducing the long-time coherent integration loss caused by range migration and Doppler spread, as Figure 1 shown. The specific steps are as follows:

[0057] Step 1: Perform quadrature down-conversion processing on the target echo signal received by the radar to obtain a baseband signal;

[0058] Specifically, the radar uses a linear frequency modulation signal as the transmitted signal. During the observation period, the radar has received a total of N pulse echo signals. The nth pulse signal transmitted by the radar is , assuming that there are maneuvering targets within the observation range, ignoring the influence of third-order and higher-order motions, the echo signal of the th maneuvering target in the nth pulse received by the radar becomes a baseband signal after quadrature down-conversion, which can be expressed as:

[0059] (1)

[0060] (2)

[0061] In the formula, is the fast time, is the slow time, represents a rectangular pulse with as the pulse width, is the chirp rate of the linear frequency modulation signal, is the carrier frequency, the pulse repetition period is , and there is , represents the total number of coherent integration pulses (this can reflect long time); is the instantaneous radial distance of the th maneuvering target in the nth pulse, , , and respectively represent the initial radial distance, initial radial velocity, and initial radial acceleration between the th target and the radar; represents the backscattering coefficient of the th target, is the speed of light.

[0062] Step 2: Frequency-domain matched filtering of the baseband echo signal, that is, performing frequency-domain pulse compression on the baseband signal using a frequency-domain matched filter to obtain a frequency-domain matched filtering signal;

[0063] Specifically, a frequency-domain matched filter is constructed according to the received signal, and frequency-domain pulse compression is performed on the baseband echo signal. The frequency-domain echo expression of the baseband signal , the frequency-domain expression of the matched filter and the frequency-domain expression of the echo signal after pulse compression can be respectively expressed as:

[0064] (3)

[0065] (4)

[0066] (5)

[0067] In the formula, is the fast-time dimension frequency corresponding to the fast time , represents the bandwidth of the radar transmitted signal, represents the Doppler frequency. Since , its influence can be ignored; is the total number of maneuvering targets existing within the radar observation range; is a rectangular pulse with a width of , is the speed of light, represents the imaginary part.

[0068] Step 3: Decompose the frequency-domain matched filtering signal into multiple frequency-domain sub-band signals, and perform inverse Fourier transform on each frequency-domain sub-band signal to obtain multiple time-domain sub-band signals;

[0069] Specifically, decomposing the frequency-domain signal of the frequency-domain matched filtering into multiple sub-band signals can achieve an equivalent range resolution of the extended sub-band signals, thereby effectively reducing the range migration caused by long-time coherent integration and reducing the impact on the coherent integration of maneuvering targets. The frequency-domain expression of the th sub-band signal is:

[0070] (6)

[0071] In the formula, represents the sub-band signal bandwidth, is the number of sub-bands after frequency-domain decomposition, and , represents the carrier frequency offset of the th sub-band signal.

[0072] Inverse Fourier transform is performed on the sub-band signals after sub-band decomposition, and the time-domain expression of the th sub-band signal can be obtained.

[0073] (7)

[0074] Among them, represents the amplitude accumulation gain of each sub-band signal.

[0075] In order to eliminate the influence of target range migration on the coherent accumulation performance of sub-band signals, the maximum range migration amount of the target within the coherent accumulation interval should not be greater than the equivalent range resolution of the sub-band signals, and it is necessary to satisfy:

[0076] (8)

[0077] After transforming the above formula, the limit condition of the sub-band bandwidth and the number of sub-band decomposition segments in the frequency domain can be expressed as:

[0078] (9)

[0079] (10)

[0080] Among them, is the upper limit of the radial velocity between the set maneuvering target and the radar, is the upper limit of the radial acceleration between the set maneuvering target and the radar, is the coherent accumulation time.

[0081] Step 4: Construct the slow-time acceleration phase compensation function of the sub-band signal, and use the acceleration phase compensation function to compensate the Doppler frequency migration generated by each time-domain sub-band signal in the slow-time dimension to obtain the first compensation signal corresponding to each time-domain sub-band signal;

[0082] After sub-band decomposition of the received signal, the target echo energy of the sub-band signal is concentrated within the same range gate, and the influence of range migration on the inter-pulse coherent accumulation of the sub-band signal can be ignored.

[0083] Construct the acceleration phase compensation function , to compensate the Doppler frequency migration generated by each sub-band signal in the slow-time dimension. Then, in the th sub-band signal, the slow-time dimension signal of the th target in the corresponding range gate can be expressed as:

[0084] (11)

[0085] (12)

[0086] In the formula, represents the acceleration search interval, represents the acceleration search factor, and satisfies Taking the integer between ; represents the residual acceleration.

[0087] When the acceleration search factor satisfies , the residual acceleration satisfies . At this time, when the Doppler frequency migration caused by the residual acceleration is less than one Doppler frequency unit, coherent accumulation of the echo energy of the sub-band signal can be achieved, expressed as:

[0088] (13)

[0089] After simplification, the acceleration search interval constraint condition and the first compensation signal after acceleration phase compensation can be respectively expressed as:

[0090] (14)

[0091] (15)

[0092] Among them, , respectively represent the initial radial distance and the initial radial velocity between the th maneuvering target and the radar, is the wavelength of the radar transmitted signal.

[0093] Step 5: Construct the slow-time velocity phase compensation function of the sub-band signal, and use the velocity phase compensation function to compensate the errors generated by the Doppler frequency change during the inter-segment coherent accumulation process of each first compensation signal to obtain the second compensation signal corresponding to each first compensation signal;

[0094] Construct the velocity phase compensation function to compensate for the influence caused by the Doppler frequency change during the inter-segment coherent accumulation process of different sub-band signals. The velocity phase compensation function and the compensated second compensation signal can be respectively expressed as:

[0095] (16)

[0096] (17)

[0097] In the formula, represents the residual velocity, is the velocity search factor, is the velocity search interval.

[0098] When the velocity search factor Meet When, the residual velocity meets . At this time, when the Doppler frequency change of different sub - band signals does not exceed one Doppler frequency unit, coherent accumulation between sub - band signal segments can be achieved, expressed as:

[0099] (18)

[0100] After simplification, the velocity search interval limit condition can be expressed as:

[0101] (19)

[0102] Step 6: Coherent accumulation of sub - band signals in slow time, that is, performing Fourier transform in the slow - time dimension on each second compensation signal respectively to obtain the coherent accumulation result corresponding to each time - domain sub - band signal;

[0103] First, perform Fourier transform in the slow - time dimension on each second compensation signal, that is, perform coherent accumulation on the processing results of all sub - band signals, which can be expressed as:

[0104] (20)

[0105] In the formula, is the slow - time dimension frequency domain corresponding to .

[0106] Step 7: Perform coherent synthesis between segments on the coherent accumulation results corresponding to each time - domain sub - band signal to obtain the coherent accumulation result corresponding to the target echo signal;

[0107] From the compensated signal expression, the Doppler frequency of the th sub - band signal is expressed as:

[0108] (21)

[0109] After compensation, the Doppler frequencies between sub - band signals are fixed. Therefore, the coherent accumulation result between segments of all sub - band signals can be obtained through Fourier transform as follows:

[0110] (22)

[0111] Among them, is the Fourier transform frequency domain corresponding to .

[0112] Next, the performance of a weak high - dynamic signal sub - band decomposition long - time coherent accumulation method provided by the present invention is evaluated.

[0113] After inter - frame coherent integration, the echo energy of the target is accumulated into a peak. According to the position of the peak, the target's radial distance and motion parameters can be obtained, that is:

[0114] (23)

[0115] After analysis, the number of complex multiplications required for the method proposed in the present invention can be expressed as:

[0116] (24)

[0117] In the formula, represents the number of range gates, and respectively represent the number of velocity and acceleration search times.

[0118] For the common maximum - likelihood estimation method, a three - dimensional joint search for the target's radial distance, velocity, and acceleration is required. The number of complex multiplications required can be expressed as:

[0119] (25)

[0120] In the formula, and respectively represent the number of velocity and acceleration search times for the maximum - likelihood estimation method. In addition, the velocity search interval and the acceleration search interval should satisfy , , so the number of complex multiplications of the maximum - likelihood estimation method can also be expressed as:

[0121] (26)

[0122] Analyze the output signal - to - noise ratio of the method proposed in the present invention:

[0123] After spectrum sub - band decomposition, the signal - to - noise ratio of the th target in each sub - band signal can be expressed as:

[0124] (27)

[0125] In the formula, is the signal - to - noise ratio of the th target echo signal before frequency - domain matched filtering, is the sub - band decomposition gain, and there is , is the time - bandwidth product of the linear frequency - modulated signal.

[0126] Therefore, the signal - to - noise ratio after completing inter - frame coherent integration can be expressed as:

[0127] (28)

[0128] Therefore, the output signal-to-noise ratio of the low computational complexity long-time coherent integration method based on sub-band decomposition mentioned in the present invention is not affected by sub-band decomposition, and there is no signal-to-noise ratio loss compared with the conventional maximum likelihood estimation method.

[0129] As shown in the appendix Figure 1 The present invention proposes a low-complexity long-time coherent integration method based on sub-band decomposition. By performing frequency-domain sub-band decomposition on the received signal, the method realizes high-precision tracking of the intercepted target using broadband signals, and constructs narrowband signals to achieve low-complexity long-time coherent integration and search of the target.

[0130] In this embodiment, the radar transmit signal is set as a linear frequency modulation signal with a carrier frequency in the S band and a bandwidth . The radar system parameters and target motion parameters are shown in Table 1:

[0131] Table 1

[0132]

[0133] Figure 2 The sub-band decomposition schematic diagram is given. The results after pulse compression corresponding to different echo pulses are shown in the appendix Figure 3 .

[0134] To eliminate the influence of range migration on signal accumulation, the signal is subjected to sub-band decompositions in the simulation. The two-dimensional time-domain diagrams of the signal before and after sub-band decomposition are shown in Figure 4 . It can be seen that the echo energy of the target after sub-band decomposition is concentrated within the same range gate;

[0135] After completing the inter-segment coherent integration of all sub-band signals, the energy of all sub-band signals of the target is effectively accumulated into a sharp peak, as shown in Figure 5 . On this basis, the estimation results of the target's radial distance and motion parameters can be obtained respectively, as shown in Table 2:

[0136] Table 2

[0137]

[0138] As shown in Figure 6 , comparing the computational complexity of the method proposed in the present invention with the conventional maximum likelihood estimation method, in this embodiment, the method proposed in the present invention can reduce the computational complexity by about 20 times (more than one order of magnitude).

[0139] Sub-band decomposition gain is , the coherent integration gain in the slow time dimension is , the sub-band coherent synthesis gain is , so the theoretical signal-to-noise ratio should be , the signal-to-noise ratio of the result in the embodiment is calculated as , which is basically consistent with the theoretical result.

[0140] In summary, the present invention proposes a method for coherent integration of weak radar echoes of maneuvering targets with low computational complexity and long time based on frequency-domain sub-band decomposition, which does not depend on the maneuvering characteristics of the targets. While obtaining a signal-to-noise ratio gain comparable to that of the maximum likelihood estimation method, the computational complexity is greatly reduced.

[0141] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can certainly make various corresponding changes and deformations according to the present invention, but these corresponding changes and deformations should all fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for long-time coherent integration of sub-band decomposition of weak high-dynamic signals, characterized in that Including the following steps: S1: Perform quadrature down-conversion processing on the target echo signal received by the radar to obtain a baseband signal; S2: Use a frequency-domain matched filter to perform frequency-domain pulse compression on the baseband signal to obtain a frequency-domain matched filtering signal; S3: Decompose the frequency-domain matched filtering signal into multiple frequency-domain sub-band signals, and perform inverse Fourier transform on each frequency-domain sub-band signal to obtain multiple time-domain sub-band signals; S4: Use an acceleration phase compensation function to compensate the Doppler frequency migration generated by each time-domain sub-band signal in the slow-time dimension to obtain a first compensation signal corresponding to each time-domain sub-band signal; S5: Use a velocity phase compensation function to compensate the error generated by each first compensation signal due to the change in Doppler frequency during the inter-segment coherent integration process to obtain a second compensation signal corresponding to each first compensation signal; S6: Perform slow-time dimension Fourier transform on each second compensation signal respectively to obtain a coherent integration result corresponding to each time-domain sub-band signal; S7: Perform inter-segment coherent synthesis on the coherent integration results corresponding to each time-domain sub-band signal to obtain a coherent integration result corresponding to the target echo signal.

2. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 1, characterized in that, The frequency-domain matched filtering signal in step S2 is as follows: Among them, is the total number of maneuvering targets existing within the radar observation range, represents the backscattering coefficient of the th maneuvering target, is the slow time corresponding to the th pulse echo signal, is the carrier frequency, is the th instantaneous radial distance of the th maneuvering target in the th pulse echo signal, is the total number of pulse echo signals received by the radar during the observation period, is the fast-time dimension frequency corresponding to the fast time is the radar transmit signal bandwidth, is a rectangular pulse with a width of , is the speed of light, represents the imaginary part.

3. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 2, characterized in that Any one of the time-domain subband signals in step S3 is as follows: Among them, represents the amplitude accumulation gain of each time-domain sub-band signal, is the pulse width of the radar transmitted signal, represents the time-domain sub-band signal bandwidth, is the number of frequency-domain sub-band signals after decomposition, and , represents the carrier frequency offset of the th frequency-domain sub-band signal.

4. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 3, characterized in that If it is necessary to eliminate the error generated during the coherent accumulation of the time-domain sub-band signals due to the target range migration, the bandwidth of the time-domain sub-band signals and the number of the frequency-domain sub-band signals after decomposition shall satisfy: Among them, is the upper limit of the radial velocity between the set maneuvering target and the radar, is the upper limit of the radial acceleration between the set maneuvering target and the radar, is the coherent integration time.

5. A method for long-time coherent integration of weak high-dynamic signal sub-band decomposition according to claim 3, characterized in that, The acceleration phase compensation function corresponding to any time-domain subband signal is as follows: Among them, represents the acceleration search interval, represents the acceleration search factor, and satisfies is rounded within and represents the th initial radial acceleration between the maneuvering target and the radar, is the rounding function; Multiply the acceleration phase compensation function by the time-domain subband signal to obtain a first compensation signal as follows: Among them, and respectively represent the initial radial distance and the initial radial velocity between the th maneuvering target and the radar, and represents the residual acceleration.

6. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 5, characterized in that When the acceleration search factor satisfies the residual acceleration satisfies and the acceleration search interval satisfies , being the radar emission signal wavelength.

7. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 5, characterized in that Velocity phase compensation function corresponding to any time-domain subband signal is as follows: Among them, is the speed search factor, is the speed search interval; Multiply the speed phase compensation function by the first compensation signal to obtain a second compensation signal as follows: Among them, represents the residual speed.

8. A method for long-time coherent accumulation of sub-band decomposition of weak high-dynamic signals according to claim 7, characterized in that When the speed search factor satisfies the residual speed satisfies and the speed search interval satisfies .

9. A method for long-time coherent integration of weak high-dynamic signal sub-band decomposition according to claim 7, characterized in that Perform slow-time dimension Fourier transform on any second compensation signal to obtain the corresponding coherent integration result as follows: Among them, is the slow-time dimension frequency domain corresponding to the slow-time dimension frequency domain corresponding to 10. A method for long-time coherent integration of sub-band decomposition of weak high-dynamic signals according to claim 9, characterized in that, Performing inter-segment coherent synthesis on the coherent integration results corresponding to each time-domain sub-band signal to obtain a coherent integration result corresponding to the target echo signal is as follows: Among them, is the Fourier transform frequency domain corresponding to the corresponding one.

Images