Bipolar ultra-wide spectrum ultra-narrow pulse moving target coherent accumulation processing method

Abstract

The invention discloses a bipolar ultra-wide spectrum ultra-narrow pulse moving target coherent accumulation processing method based on direct wave estimation and frequency domain GRFT, and relates to the field of target detection, and the method comprises the steps: 1, obtaining a plurality of pulse echo signals in the same wave position; 2, performing direct wave inter-pulse jitter estimation on the received multiple pulse data; step 3, carrying out direct wave suppression by using the recorded pure background data; 4, converting a real signal into a complex signal through Hilbert transform, and transforming the complex signal to a frequency domain through FFT to complete narrow pulse rejection; 5, motion parameter search is completed in a frequency domain based on the guiding information; step 6, performing phase-coherent accumulation on the output signal, then performing CFAR detection, and recording a trace point exceeding a threshold; step 7, outputting a target trace point detected after phase-coherent accumulation; according to the invention, the pulse jitter is reduced from dozens of picoseconds to picoseconds, the multi-pulse accumulation loss does not exceed 2dB, and the multi-pulse accumulation performance of the moving target is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of target detection, and particularly relates to a coherent accumulation processing method for bipolar ultra-wideband extremely narrow pulse moving targets based on direct wave estimation and frequency domain Generalized Radon-Fourier Transform (GRFT). Background Art

[0002] Ultra-wideband technology has characteristics such as being insensitive to channel fading, having a low transmit signal power spectral density, and a low system complexity, and has received increasing attention in the field of radar detection. Among them, impulse ultra-wideband radar has a relatively large spectral width (in the order of GHz), has the characteristics of low intercept and high resolution, and can avoid interference from other systems, playing an important role in the field of target detection.

[0003] Impulse ultra-wideband radar mainly uses carrier-free impulse signals as the transmit signals of the radar. The radiation process of the signal by the planar radiation antenna is equivalent to differentiating the input time-domain signal waveform. That is, the input zero-order Gaussian signal becomes a first-order Gaussian signal after passing through the antenna. Therefore, the ultra-wideband extremely narrow pulse signal appears as a bipolar impulse signal after passing through the antenna. However, due to its short pulse width and large spectral width span, the target detection range is limited.

[0004] In view of the detection power limitation of impulse radar, current impulse radar needs to adopt a multi-pulse accumulation method for detecting long-distance targets, but there are the following two technical problems: 1) There is pulse-to-pulse jitter in the transmission of impulse signals, and the randomness of the jitter reduces the accumulation performance of the target; 2) The radar has high resolution and is extremely sensitive to the moving speed of the target during multi-pulse accumulation; conventional guidance accuracy is difficult to meet the alignment in the range-Doppler dimension, resulting in a serious decline in the accumulation performance.

[0005] In view of the problem of insufficient detection power of impulse radar for long-distance targets, there is an urgent need to design a moving target accumulation detection technology applicable to impulse radar to improve its detection ability for medium- and long-range moving targets. Summary of the Invention

[0006] In view of the above problems, the object of the present invention is: based on the echo characteristics of impulse signals, a coherent accumulation method for bipolar ultra-wideband extremely narrow pulse moving targets based on direct wave estimation and frequency domain Generalized Radon-Fourier Transform (GRFT) is proposed to solve the problems of serious decline in accumulation performance caused by pulse-to-pulse jitter of impulse signals and sensitivity to the moving speed of targets during multi-pulse accumulation detection of medium- and long-range ultra-wideband impulse radar.

[0007] To achieve the above object, the technical solution adopted by the present invention is: a coherent accumulation processing method for bipolar ultra-wideband extremely narrow pulse moving targets based on direct wave estimation and frequency domain GRFT, including the following steps: Step 1, Echo Signal Acquisition: Obtain multiple pulse echo signals in the same wave position; Step 2, Pulse-to-Pulse Jitter Compensation: Estimate the direct wave pulse-to-pulse jitter for the received multiple pulse data; Step 3, Direct Wave Suppression: Use the recorded pure background data for direct wave suppression; Step 4, Narrow Pulse Rejection in Frequency Domain Based on Hilbert Transform: Convert the real signal to a complex signal through Hilbert transform and complete narrow pulse rejection after transforming to the frequency domain by FFT; Step 5, Parameter Search Based on Frequency Domain GRFT: Complete the motion parameter search in the frequency domain based on the guiding information; Step 6, Coherent Integration: Perform coherent integration on the signal output from Step 5 through slow-time dimension FFT, then perform CFAR detection on the integrated signal, and record the points crossing the threshold; Step 7, Point Output: Output the target points detected after coherent integration.

[0008] The pulse-to-pulse jitter estimation in Step 2 includes: For the m-th pulse echo signal The discretized form is expressed as The fast-time frequency domain can be written as:

[0009] Where Indicates performing FFT operation on the fast-time dimension of the two-dimensional data, and l is the discretized representation in the frequency domain; by processing with the first pulse, the maximum value of the correlation peak is extracted, and then the pulse-to-pulse jitter amount of the m-th pulse is estimated Is:

[0010] Where Indicates finding the correlation, and then performing pulse-to-pulse jitter compensation on the signal after direct wave suppression to obtain:

[0011] The direct wave suppression in Step 3 includes: After completing the pulse-to-pulse jitter estimation, process with the recorded direct wave echo background without targets Extract the maximum value of the correlation peak, and then estimate the pulse-to-pulse jitter amount of the m-th pulse Is:

[0012] Use To align the background with the received signal to obtain the aligned background signal , and use And Subtract to obtain the signal after direct wave rejection: .

[0013] The fourth step of narrow pulse rejection in the frequency domain based on Hilbert transform includes: Discriminate the pulse width and intensity. When the pulse time width is less than the set value, replace the data at that place. The amplitude is the mean value of the data sampled over the threshold, and the phase is retained:

[0014] Where is the frequency domain signal of the m-th pulse, A is a certain set value, and K is a certain set width.

[0015] The fifth step of parameter search based on frequency domain GRFT includes: For each search compensation, a frequency domain implementation method is adopted. The coherent accumulation result based on GRFT is expressed as:

[0016] Where represents performing an IFFT operation on the slow time dimension of two-dimensional data, , where the search speed is expressed as:

[0017] Where , .

[0018] Compared with the prior art, the technical solution adopted by the present invention has the following beneficial effects: Aiming at the problem of multi-pulse accumulation of moving targets in medium and long-range ultra-wideband impulse radars, the present invention adopts a coherent accumulation method for bipolar ultra-wideband extremely narrow pulse moving targets based on direct wave estimation and frequency domain generalized Radon-Fourier transform (GRFT), reducing the pulse jitter from the order of dozens of picoseconds to the picosecond order, achieving a multi-pulse accumulation loss of no more than 2 dB, and effectively improving the multi-pulse accumulation performance of moving targets. Description of the Drawings

[0019] Figure 1 is a flow chart of the bipolar ultra-wideband extremely narrow pulse target accumulation processing method based on direct wave estimation and frequency domain GRFT in this embodiment.

[0020] Figure 2 is a schematic diagram of direct wave pulse-to-pulse jitter estimation in this embodiment.

[0021] Figure 3 is a schematic diagram of direct wave cancellation and suppression in this embodiment.

[0022] Figure 4 is a partially enlarged schematic diagram of direct wave cancellation and suppression in this embodiment.

[0023] Figure 5 It is a schematic diagram for comparison before and after suppression of unintentional interference in the frequency domain of this embodiment.

[0024] Figure 6 It is a schematic diagram for comparison of the effects before and after accumulation based on frequency-domain GRFT in this embodiment. Detailed implementation manners

[0025] The technical solution of the present invention will be further explained below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0026] Based on the requirement of multi-pulse accumulation detection of moving targets by medium- and long-range ultra-wideband impulse radars, this solution first conducts inter-pulse jitter estimation based on direct wave echoes and direct wave suppression, then completes the elimination of unintentional interference in the frequency domain on the basis of Hilbert transform processing of bipolar ultra-narrow pulse signals, and finally uses frequency-domain GRFT to achieve multi-pulse coherent accumulation.

[0027] Since there is trigger jitter in the impulse signal source, and the single-pulse echo of the target is usually relatively weak, it is difficult to detect the jitter amount of the signal source through the echo of the target. Therefore, the direct wave of the impulse signal is used to measure the inter-pulse jitter amount.

[0028] On the basis of completing the direct wave estimation, in order to reduce the influence of the direct wave on target detection, the direct wave background without a target is collected, and the direct wave cancellation in the target scene is completed through the method of correlation alignment.

[0029] The ultra-wideband ultra-narrow pulse signal mainly uses a carrier-free impulse signal as the radar transmission signal. The ideal impulse signal can be analytically represented by Gaussian functions of different orders, and it satisfies

[0030] The radiation process of the planar radiation antenna on the signal is equivalent to differentiating the input time-domain signal waveform That is, the input zero-order Gaussian signal becomes a first-order Gaussian signal after passing through the antenna , resulting in the ultra-wideband ultra-narrow pulse signal being a bipolar impulse signal after passing through the antenna, and its frequency spectrum is:

[0031] Taking its Hilbert transform pair The Fourier transform is equivalent to and performing frequency-domain multiplication to obtain

[0032] The finally obtained complex analytic signal Its frequency spectrum is expressed as:

[0033] It only has information in the positive frequency, realizing the spectrum fusion function of the ultra-wideband extremely narrow pulse signal.

[0034] Due to the very wide frequency spectrum of the ultra-wideband extremely narrow pulse, there are many unintentional interferences in the echo. It is necessary to perform narrow pulse detection on the frequency spectrum of the target echo of the frequency spectrum, and adopt the method of directly removing and filling noise to remove the strong narrow pulses with pulse width less than a certain set value in the original signal.

[0035] The impulse radar multi-pulse accumulation experiment is carried out by separating the transmitting and receiving antennas. Each packet of data contains 20 pulse signals, relying on Figure 1 the bipolar ultra-wideband extremely narrow pulse target accumulation processing flow based on direct wave estimation and frequency domain GRFT in Step 1, perform direct wave inter-pulse jitter estimation on the received 20 pulse data. For the m-th pulse echo signal The discretized form is expressed as The fast time frequency domain can be written as:

[0036] where represents the FFT operation on the fast time dimension of the two-dimensional data, and l is the discretized representation in the frequency domain. By performing correlation processing with the first pulse, the maximum value of the correlation peak is extracted, and then the inter-pulse jitter amount of the m-th pulse is estimated, as Figure 2 shown.

[0037]

[0038] where represents finding the correlation.

[0039] Step 2, according to the inter-pulse jitter amount estimated in Step 1, perform inter-pulse jitter compensation on the signal.

[0040]

[0041] Step 3, after completing the inter-pulse jitter estimation, perform correlation processing with the received direct wave echo background without a target, extract the maximum value of the correlation peak, and then estimate the inter-pulse jitter amount of the m-th pulse.

[0042]

[0043] Using Align the background with the received signal to obtain the aligned background signal , subtract from to obtain the signal after direct wave removal: ,

[0044] The comparison before and after direct wave suppression is as Figure 3 and 4 shown. It can be seen from the local enlarged view that there is a target.

[0045] Step 4: Perform Hilbert transform on the original signal in Step 3 to convert the real signal into a complex signal. Since the spectrum of the ultra-wideband extremely narrow pulse is very wide and there are many unintentional interferences in the echo, it is necessary to perform narrow pulse detection on the spectrum of the target echo in Step 3, discriminate the pulse width and intensity. When the pulse width is less than the set value, replace the data at that place, with the amplitude being the mean of the sampled data exceeding the threshold and the phase being retained.

[0046] Discriminate the pulse width and intensity. When the pulse width is less than the set value, replace the data at that place, with the amplitude being the mean of the sampled data exceeding the threshold and the phase being retained.

[0047]

[0048] where is the frequency domain signal of the m-th pulse, A is a certain set value, and K is a certain set width.

[0049] Adopt the method of directly removing and filling noise to remove the strong narrow pulses with pulse width less than a certain set value in the original signal. The effect after removal is as Figure 5 shown.

[0050] Step 5: Utilize information such as the preset velocity and acceleration of the target in the observation range, etc., to obtain the search ranges within the velocity and acceleration ranges at the specified target position, which are successively , , which can reduce the computational amount of the search by several orders of magnitude.

[0051] In order to further improve the accuracy of target alignment, for each search compensation, a frequency domain implementation method is adopted. The coherent accumulation result based on GRFT can be expressed as

[0052] where represents performing IFFT operation on the slow time dimension of the two-dimensional data, , where the search velocity is expressed as:

[0053] Among them , .

[0054] The above-mentioned coherent accumulation process can be completed in the fast-time frequency-pulse domain. On the basis of narrowing the search range, the computational complexity of coherent accumulation detection is further reduced.

[0055] Step 6: Accumulate the signal through frequency-domain GRFT, and obtain an accumulation gain of 11 dB for 20 pulses, which is 2 dB less than the theoretical value; then perform CFAR detection on the accumulated signal and record the traces exceeding the threshold. The comparison of the effects before and after frequency-domain GRFT accumulation is as Figure 6 shown

[0056] Step 7: Output the target traces detected after coherent accumulation.

[0057] Although the present invention has been disclosed above with preferred embodiments, the embodiments and the drawings are not intended to limit the present invention. Any person skilled in the art can make various changes or modifications without departing from the spirit and scope of the present invention, but still within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be defined by the protection scope of the claims of this application.

Claims

1. A coherent integration processing method for bipolar ultra-wideband extremely narrow pulse moving targets, based on direct wave estimation and frequency domain GRFT, is characterized in that It includes the following steps: Step 1, echo signal acquisition: Acquire multiple pulse echo signals in the same wave position; Step 2, inter-pulse jitter compensation: Estimate the direct wave inter-pulse jitter for the received multiple pulse data; Step 3, direct wave suppression: Use the acquired pure background data for direct wave suppression; Step 4, narrow pulse rejection in the frequency domain based on Hilbert transform: Convert the real signal to a complex signal through Hilbert transform, and complete narrow pulse rejection after transforming to the frequency domain through FFT; Step 5, parameter search based on frequency domain GRFT: Complete motion parameter search in the frequency domain based on the guiding information; Step 6, coherent integration: Coherently integrate the signal output in Step 5 through slow-time dimension FFT, then perform CFAR detection on the integrated signal, and record the points crossing the threshold; Step 7, point output: Output the target points detected after coherent integration.

2. A coherent integration processing method for bipolar ultra-wideband extremely narrow pulse moving targets according to claim 1, characterized in that The inter-pulse jitter estimation in Step 2 includes: For the m-th pulse echo signal which is represented in a discretized form as The fast-time frequency domain can be written as: ， Among them represents performing an FFT operation on the fast time dimension of two-dimensional data, and l is the discretized representation in the frequency domain; by processing with the first pulse, the maximum value of the correlation peak is extracted, and then the inter-pulse jitter of the m-th pulse is estimated is as follows: , where represents seeking correlation, and then obtaining the signal after direct wave suppression for inter-pulse jitter compensation to get:

3. A coherent integration processing method for bipolar ultra-wideband extremely narrow pulse moving targets according to claim 1, characterized in that The direct wave suppression in Step 3 includes: After completing the inter-pulse jitter estimation, it is processed with the received direct wave echo background without targets to extract the maximum value of the relevant peak, and then estimate the inter-pulse jitter amount of the m-th pulse as follows: ， Utilize Align the background with the received signal to obtain the aligned background signal , subtract from to obtain the signal after direct wave removal: 。 4. A coherent integration processing method for bipolar ultra-wideband extremely narrow pulse moving targets according to claim 1, characterized in that The narrow pulse rejection in the frequency domain based on Hilbert transform in Step 4 includes: Discriminate the pulse width and intensity. When the pulse width is less than the set value, replace the data at that place, with the amplitude being the mean of the sampled data crossing the threshold and the phase being retained: ， Among them is the frequency-domain signal of the m-th pulse, A is a certain set value, and K is a certain set width.

5. A coherent integration processing method for bipolar ultra-wideband extremely narrow pulse moving targets according to claim 1, characterized in that The parameter search based on frequency domain GRFT in Step 5 includes: For each search compensation, use the frequency domain implementation method. The coherent integration result based on GRFT is expressed as: ， Among them represents performing an IFFT operation on the slow time dimension of two-dimensional data, , where the search speed is expressed as: ， Among them , .

Images