A video-level dense false target interference suppression method based on repetition frequency clustering

Abstract

The present invention relates to a video-level dense false target interference suppression method based on repetition frequency clustering. Aiming at the shortcoming that traditional radars cannot distinguish strong threat real targets from a high-confidence dense false target group, a method for accurately suppressing active deceptive interference entering the radar system from the main lobe or side lobe of the radar antenna is proposed. The specific steps include: channel echo over-threshold envelope condensation, repetition frequency histogram construction, multi-order repetition frequency clustering, amplitude stability judgment, and dense false target elimination. The present invention effectively enhances the radar's ability to resist active deceptive interference and improves the radar's target detection probability and combat effectiveness in a complex electromagnetic environment.

Description

Technical Field

[0001] The invention relates to an interference elimination method, which is mainly applied to the field of radar interference signal suppression. Background Art

[0002] Many actual battles since the 21st century have shown that modern warfare is, without exception, a high-tech three-dimensional war led by electronic warfare and promoted by the integration of land, sea, air, space and electricity. The electromagnetic environment under actual combat conditions is very complex, and various active and passive electromagnetic interferences pose a serious threat to the operation of various radar sensors. It can be said that information confrontation in a complex electromagnetic environment is one of the most typical characteristics of modern warfare. Deceptive dense false target interference not only has an absolute advantage in energy, but also has a high degree of similarity with the real target echo signal in multiple dimensions such as airspace, time domain and frequency domain. This makes it difficult for radar to effectively extract the real target from the high-confidence false target group, and a large number of false targets will also greatly consume the available resources of phased array radars.

[0003] The patent "A missile-borne forwarding interference suppression method based on false target characteristics" (CN104931939A) controls the adjustable parameters in the three characteristic parameter thresholds of noise floor, false target spacing interval and false target spectrum peak broadening to mark the point data as true and false targets, and performs target association based on the marking results to form a real target track; the patent "A radar dense false target interference suppression method" (CN105044686A) determines the existence and interference style of interference by sensing the interference environment, and then uses frequency domain filtering technology to perform anti-interference suppression processing on dense false target interference data; the document "Dense false target interference suppression algorithm based on frequency agility combined with waveform entropy" Method" (System Engineering and Electronic Technology, 2021, Vol.43, No.6) uses waveform entropy to identify targets and interference, and then uses the local outlier factor detection algorithm to remove data with strong interference side lobes superimposed in the target echo; the literature "Using waveform entropy to identify targets and interference; then using the local outlier factor detection algorithm to remove data with strong interference side lobes superimposed in the target echo" (Journal of Xi'an Jiaotong University, 2020, Vol.54, No.12) estimates the interference frequency modulation slope of each pulse repetition period through the maximum high-order moment criterion, determines the optimal order of fractional Fourier transform, and then calculates the optimal order FRFT of the echo, and filters out part of the interference by peak masking.

[0004] Aiming at the characteristics of "equal amplitude and equal interval distribution" of typical deceptive dense false target interference, the present invention proposes a video-level dense false target interference suppression method based on repetition frequency clustering. The method has the advantages of small computational complexity, high correct rejection rate and convenient engineering implementation. Summary of the invention

[0005] In order to overcome the shortcoming that traditional radar cannot distinguish strong threat real targets from high confidence dense false target groups, the present invention proposes a video-level dense false target interference suppression method based on repetition frequency clustering through repetition frequency histogram construction, multi-order repetition frequency clustering and amplitude stability judgment.

[0006] The present invention proposes a video-level dense false target interference suppression method based on repetition frequency clustering, which includes the following steps. The implementation process and software flow are as follows: Figure 1 As shown:

[0007] S1: Channel echo exceeds the threshold envelope condensation.

[0008] S2: Construction of repetition frequency histogram. Solve the maximum amplitude of each condensation point as the amplitude A of the envelope mj , the distance unit corresponding to the maximum amplitude is taken as the distance unit I of the envelope mj . Set the distance range of repetition clustering to [I min ,I max ], the width of the frequency histogram grid is w, then the number of frequency histogram grids is:

[0009] N unit =floor((I max -I min ) / w)+1 (8)

[0010] Where floor(·) means rounding towards negative infinity. The center of each frequency histogram grid is:

[0011] I c =(c-0.5)×w+I min (9)

[0012] Where c is the histogram grid index, ranging from 1≤c≤N unit . Traverse all point trace pairs {I ma ,I mb}, initially set a∈[1,J], b=a+1, calculate its repetition frequency, that is, the distance difference I delta =|I ma -I mb |, if I min ≤I delta ≤I max , it is mapped to the histogram grid, and the corresponding histogram grid count is increased by one:

[0013] c=floor((I delta -I min ) / w)+1 (10)

[0014] S(c)=S(c)+1 (11)

[0015] Among them, S(c) represents the number of elements in the cth histogram grid.

[0016] S3: Multi-order repetition frequency clustering. Reset a∈[1,J], b=a+d, d is the distance difference order, traverse all point trace pairs again, repeat the distance difference calculation and histogram mapping in the above S2 step until the traversal is completed, S(c) is the distance difference histogram statistical result.

[0017] S4: Amplitude stability determination. Set the histogram grid interference count detection threshold T jam , take the histogram S where the count is greater than the threshold T jam The grid S(c) of the vector is used for the second amplitude discrimination. The amplitudes of all different envelopes falling into the grid S(c) are sorted, and the sorted vector is {A m1 ,A m2 ,...,A mH}, where H is the number of envelopes falling into the grid S(c). The amplitude stability criterion is:

[0018] A jam =A m(round(H / 2)) (12)

[0019] Where round(·) means rounding to the nearest integer. Set the amplitude jitter threshold to T A Decibel, recalculate the histogram bin counts:

[0020] S(c)=S(c)-1,|L×logK(A mj )-L×logK(A jam )|≥T A (13)

[0021] Wherein, L represents the amplitude coefficient, and logK(·) represents the logarithm with K as the base.

[0022] S5: Eliminate dense false targets. Still set the detection threshold T for the number of dense false targets jam , if S(c)≥T jam , it means that there is dense false target interference, and the video echo {A mj1 ,A mj2 ...A mjk} are all set to 0; otherwise, it means there is no dense false target interference and no operation is performed.

[0023] Furthermore, in S1: using the 2 / 3 criterion, the video echo data after the Doppler channel constant false alarm fixed threshold detection is traversed channel by channel to complete the connection of the envelope splitting points:

[0024]

[0025] Among them, S m (n) indicates that the nth range unit of the mth Doppler channel has exceeded the threshold, Z m (n) indicates the connection mark of the nth distance unit of the mth Doppler channel. m (n) Merge and aggregate, that is, count the number of data segments with consecutive values ​​of 1, remove the data segments with a value less than or equal to 1, and the remaining number of aggregate points is J. The envelope structure information of each aggregate point is {A mj1 ,A mj2 ...A mjk}, where A mjk It represents the amplitude of the kth range unit of the jth condensation point in the mth Doppler channel.

[0026] Furthermore, the step S2 of constructing the repetition frequency histogram and the step S3 of multi-order repetition frequency clustering map the envelope distance difference into an equal-width histogram, thereby merging and clustering envelopes with the same distance difference, and realizing the recognition of equal-distance difference distribution characteristics of dense false targets.

[0027] The beneficial effects of the present invention are as follows: by adopting the method described in the present invention, by adopting a video-level dense false target interference suppression method based on repetition frequency clustering, active deceptive interference entering the radar system from the main lobe or side lobe of the radar antenna can be accurately suppressed, the radar's ability to resist active deceptive interference can be effectively enhanced, and the radar's target detection probability and combat effectiveness in a complex electromagnetic environment can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The flowchart of a video-level dense false target interference suppression method based on repetition frequency clustering is shown in FIG. S1-S5 in the figure correspond to the S1-S5 processes described in the invention content.

[0029] Figure 2 and Figure 3 They are the echo video data before and after interference suppression by the video-level dense false target interference suppression method based on repetition frequency clustering under dense false target interference. DETAILED DESCRIPTION

[0030] The implementation process of the present invention is as follows Figure 1 As shown, the specific process is described as follows:

[0031] S1: Channel echo crosses the threshold envelope condensation. Dense false target interference usually has one or more dimensional feature differences with the real target in the Doppler domain, distribution domain and energy domain. Based on the data of the Doppler channel after the conventional radar processes the moving target and performs constant false alarm detection, the 2 / 3 criterion is used to traverse the video echo data after the Doppler channel constant false alarm fixed threshold detection channel by channel to complete the connection of the envelope splitting points:

[0032]

[0033] Among them, S m (n) indicates that the nth range unit of the mth Doppler channel has exceeded the threshold, Z m (n) indicates the connection mark of the nth distance unit of the mth Doppler channel. m (n) Merge and aggregate, that is, count the number of data segments with consecutive values ​​of 1, remove the data segments with a value less than or equal to 1, and the remaining number of aggregate points is J. The envelope structure information of each aggregate point is {A mj1 ,A mj2 ...A mjk}, where A mjk It represents the amplitude of the kth range unit of the jth condensation point in the mth Doppler channel.

[0034] S2: Construction of repetition frequency histogram. Typical deceptive dense false target interference presents equal amplitude in energy and equal interval distribution in distance. First, the distance distribution characteristics are statistically analyzed, and the maximum amplitude of each condensation point is solved as the amplitude A of the envelope. mj , the distance unit corresponding to the maximum amplitude is taken as the distance unit I of the envelope mj . Set the distance range of repetition clustering to [I min ,I max ], the width of the frequency histogram grid is w, then the number of frequency histogram grids is:

[0035] N unit =floor((I max -I min ) / w)+1 (16)

[0036] Where floor(·) means rounding towards negative infinity. The center of each frequency histogram grid is:

[0037] I c =(c-0.5)×w+I min (17)

[0038] Where c is the histogram grid index, ranging from 1≤c≤N unit . Traverse all point trace pairs {I ma ,I mb}, initially set a∈[1,J], b=a+1, calculate its repetition frequency, that is, the distance difference I delta =|I ma -I mb |, if I min ≤I delta ≤I max, it is mapped to the histogram grid, and the corresponding histogram grid count is increased by one:

[0039] c=floor((I delta -I min ) / w)+1 (18)

[0040] S(c)=S(c)+1 (19)

[0041] Among them, S(c) represents the number of elements in the cth histogram grid.

[0042] S3: Multi-order repetition frequency clustering. Due to the influence of real targets and clutter, dense false targets have the characteristics of discontinuous distribution, so multi-envelope interval statistical distance difference is performed, a∈[1,J] is reset, b=a+d, d is the order of distance difference, and all point trace pairs are traversed again. The distance difference calculation and histogram mapping in the above S2 step are repeated until the traversal is completed. S(c) is the statistical result of the distance difference histogram.

[0043] S4: Amplitude stability judgment. Also due to the influence of real targets and clutter, some real targets and clutter may have the same distance distribution as the interference, but usually there is a significant difference between the amplitude of clutter and target and the amplitude of interference, so the amplitude stability is used to perform secondary judgment of dense false target interference. Set the histogram grid interference count detection threshold T jam , take the histogram S where the count is greater than the threshold T jam The grid S(c) of the vector is used for the second amplitude discrimination. The amplitudes of all different envelopes falling into the grid S(c) are sorted, and the sorted vector is {A m1 ,A m2 ,...,A mH}, where H is the number of envelopes falling into the grid S(c). The amplitude stability criterion is:

[0044] A jam =A m(round(H / 2)) (20)

[0045] Where round(·) means rounding to the nearest integer. Set the amplitude jitter threshold to T A Decibel, recalculate the histogram bin counts:

[0046] S(c)=S(c)-1,|L×logK(A mj )-L×logK(A jam )|≥T A (twenty one)

[0047] Wherein, L represents the amplitude coefficient, and logK(·) represents the logarithm with K as the base.

[0048] S5: Dense false target elimination. Dense false target interference presents equal amplitude and equal interval distribution. Interference can be identified by counting the number of equal amplitude and equal interval envelopes. The dense false target number detection threshold T is still set. jam , if S(c)≥T jam , it means that there is dense false target interference, and the video echo {A mj1 ,A mj2 ...A mjk} are all set to 0; otherwise, it means there is no dense false target interference and no operation is performed.

Claims

1. A video-level dense false target interference suppression method based on repetition frequency clustering, characterized by: S1: channel echo exceeds threshold envelope condensation; S2: Construct the repetition frequency histogram; solve the maximum amplitude of each condensation point as the amplitude A of the envelope mj , the distance unit corresponding to the maximum amplitude is taken as the distance unit I of the envelope mj ; Set the distance range of repetition clustering to [I min ,I max ], the width of the frequency histogram grid is w, then the number of frequency histogram grids is: N unit =floor((I max -I min ) / w)+1 (1) Where floor(·) means rounding towards negative infinity; the center of each frequency histogram grid is: I c =(c-0.5)×w+I min (2) Where c is the histogram grid index, ranging from 1≤c≤N unit ; Traverse all point trace pairs {I ma ,I mb }, initially set a∈[1,J], J is the number of condensation points, b=a+1, calculate its repetition frequency, that is, the distance difference I delta =|I ma -I mb |, if I min ≤I delta ≤I max , it is mapped to the histogram grid, and the corresponding histogram grid count is increased by one: c=floor((I delta -I min ) / w)+1 (3) S(c)=S(c)+1 (4) Among them, S(c) represents the number of elements in the cth histogram grid; S3: Multi-order repetition frequency clustering; reset a∈[1,J], b=a+d, d is the distance difference order, traverse all point trace pairs again, repeat the distance difference calculation and histogram mapping in the above S2 step until the traversal is completed, S(c) is the statistical result of the distance difference histogram; S4: Amplitude stability determination; setting histogram grid interference count detection threshold T jam , take the histogram S where the count is greater than the threshold T jam Grid S c , perform a second amplitude determination; c The amplitudes of all different envelopes in are sorted, and the sorted vector is {A m1 ,A m2 ,...,A mH }, where H is the grid S c The number of medium envelopes; the amplitude stability criterion is: A jam =A m(round(H / 2)) (5) Where round(·) means rounding to the nearest integer; set the amplitude jitter threshold to T A Decibel, recalculate the histogram bin counts: S(c)=S(c)-1,|L×logK(A mj )-L×logK(A jam )|≥T A (6) Where L represents the amplitude coefficient, logK(·) represents the logarithm with K as the base; S5: Eliminate dense false targets; set the detection threshold T for the number of dense false targets jam1 , if S(c)≥T jam1 , it means that there is dense false target interference, and the video echo {A mj1 ,A mj2 …A mjk } are all set to 0, where A mj1 A represents the amplitude of the first distance unit of the jth condensation point in the mth Doppler channel. mj2 A represents the amplitude of the second distance unit of the jth condensation point in the mth Doppler channel. mjk It indicates the amplitude of the kth distance unit of the jth condensation point of the mth Doppler channel; otherwise, it indicates that there is no dense false target interference and no operation is performed.

2. The video-level dense false target interference suppression method based on repetition frequency clustering according to claim 1 is characterized in that: The channel echo over-threshold envelope condensation described in step S1 uses the 2 / 3 criterion to traverse the Doppler channel constant false alarm fixed threshold detection video echo data channel by channel to complete the connection of the envelope splitting points: Among them, S m (n) indicates that the nth range unit of the mth Doppler channel has exceeded the threshold, Z m (n) indicates the connection mark of the nth distance unit of the mth Doppler channel; for the connected data Z m (n) Merge and condense, that is, count the number of data segments with consecutive values ​​of 1, and remove the data segments with a number less than or equal to 1. The envelope structure information of each condensation point is {A mj1 ,A mj2 …A mjk }.

Images