A clutter removal method based on staggered MTI pattern folding under strong clutter

Abstract

For the staggered MTI mode, a single distant target will form multiple false points when it enters the observation area after folding, which seriously affects the radar false alarm suppression and radar detection. In response to the anti-folding clutter requirements of the staggered MTI mode in a strong clutter environment, the present invention proposes a folding asynchrony judgment and elimination method based on the combination of over-detection point positions after MTI cancellation. First, the position of the folded echo when entering the observation area is pre-calculated according to the pulse interval of the staggered MTI waveform, and described as a 0-1 reference sequence. Afterwards, the output after MTI cancellation is directly subjected to low-threshold CFAR detection screening, and the over-detection point position is characterized as a 0-1 over-detection sequence. Finally, the 0-1 over-detection sequence is convolved with the reference sequence to obtain a folding confidence sequence. According to the folding confidence sequence, it is judged whether there is a folding echo and the position of the folding echo, and finally the folding clutter is eliminated. The present invention judges whether there is folding asynchrony based on the position relationship of the over-detection point after MTI cancellation, so it is basically not affected by strong clutter in the near area.

Description

Technical Field

[0001] The present invention belongs to the field of signal anti-interference, and in particular relates to a method for eliminating clutter in a staggered MTI mode under strong clutter. Background Art

[0002] Radars measure target space and motion information by transmitting and receiving electromagnetic waves. In addition to the target echo, radar echoes often contain clutter energy. To detect targets of interest amidst clutter, radars require clutter cancellation capabilities, such as MTI and MTD / PD. MTI and MTD / PD modes effectively suppress clutter from unambiguous detection areas. However, due to factors such as atmospheric ducting, radars may also receive folded clutter from outside the unambiguous detection area. (For example, shipborne radars operating on the sea surface are often affected by strong folded clutter from out-of-line-of-sight oil wells and anchor fields.) Current radars address clutter folding with two main approaches: First, for MTD / PD (and even-periodic MTI), filler pulses are added to maintain the slow-time echo structure of the folded clutter, preventing it from affecting the Doppler filter response and ensuring cancellation effectiveness. Second, folding asynchrony is determined by comparing the amplitude differences of echoes from multiple receive gates (in the case of folding, there is no echo within the first receive gate), thereby eliminating folded clutter.

[0003] Current radars operating in MTI mode typically use staggered pulse spacing to expand the maximum velocity coverage of the MTI filter's response passband. In this case, the aforementioned anti-aliasing clutter solution based on filler pulses is no longer applicable (it only applies to waveforms with equal pulse spacing). In strong clutter environments, the energy of near-field clutter is often far greater than that of aliased clutter, making it impossible for existing asynchronous aliasing detection logic to identify the amplitude differences of aliased clutter within multiple receive gates. Furthermore, for radars operating in certain frequency bands, targets such as incoming civil aircraft will naturally exhibit fluctuations in multi-pulse echoes due to factors such as engine modulation. Existing asynchronous aliasing detection logic, which relies on amplitude differences across multiple gates, may mistakenly identify these real targets as aliased targets, resulting in target loss (or, to avoid loss, requiring an increased aliasing detection threshold at the expense of aliased clutter rejection capability). In summary, current radars operating in strong clutter environments using staggered MTI mode lack effective anti-aliasing clutter solutions. In particular, in staggered MTI mode, a single distant target aliased into the observation area can create multiple false points, severely impacting false alarm suppression and radar detection. Summary of the Invention

[0004] In response to the need for anti-folding clutter of the staggered MTI mode in a strong clutter environment, the present invention proposes a method for determining and eliminating folding asynchrony based on the combination of over-detection point positions after MTI cancellation. First, the position of the folded echo when entering the observation area is pre-calculated based on the pulse interval of the staggered MTI waveform, and described as a 0-1 reference sequence. Afterwards, the output after MTI cancellation is directly subjected to low-threshold CFAR detection screening, and the over-detection point position is characterized as a 0-1 over-detection sequence. Finally, the 0-1 over-detection sequence is convolved with the reference sequence to obtain a folding confidence sequence. The presence of a folded echo and the position of the folded echo are determined based on the folding confidence sequence, and the folding clutter is finally eliminated. The present invention determines whether folding asynchrony exists based on the position relationship of the over-detection points after MTI cancellation, and is therefore basically unaffected by strong clutter in the near area. It comprises the following steps:

[0005] 10) Folding position reference sequence calculation: Based on the staggered MTI waveform pulse interval, the position of the folded echo when it enters the observation area is calculated in advance and described as a 0-1 reference sequence Where x(m,n)=0 or 1, M is the number of staggered MTI sliding windows, N is the number of range units in the receiving wave gate, m = 1...M, n = 1...N;

[0006] 20) MTI cancellation: Perform MTI cancellation on the pulse pressure echo to obtain M groups of sliding window cancellation output results

[0007] 30) Multi-channel CFAR screening: Perform low-threshold CFAR detection on the output results of M groups of sliding window cancellation, and add a target protection mechanism to screen out potential folding point locations, which are described as over-detection sequences. Where y(m,n)=0 or 1,

[0008] 40) Confidence sequence calculation: Perform normalized convolution operation on the passed sequence y and the reference sequence x to obtain the confidence sequence used to determine whether there is a fold and the fold position

[0009]

[0010] Among them, x m represents the mth row of the reference sequence x, y m represents the mth row of the reference sequence y, k m is x m represents the number of independent targets, Represents a sequential convolution operation;

[0011] 50) Folding judgment and elimination: Compare the confidence sequence r with the predetermined confidence threshold △, 0<△≤1, and determine the folded echo position based on the threshold crossing point of the confidence sequence to achieve elimination.

[0012] Furthermore, the reference sequence calculation in step (10) includes:

[0013] 11) Folding position traversal: gradually increase the delay of the echo until the m-th pulse echo falls into the m+q-th receiving wave gate, where q is the number of echo folding times;

[0014] 12) Reference sequence calculation: Based on the selected folded echo situation and the MTI sliding window length, the reference sequence of the folded position of each sliding window MTI output is calculated, which is recorded as Where x(m,n)=0 or 1, M is the number of staggered MTI sliding windows, and N is the number of range units within the receiving gate;

[0015] 13) Reference sequence extension: Considering that the folded echo may come from a high-speed target, its echo position may move across the range gate between pulses. To address this situation, the folded position reference sequence x needs to be extended, namely:

[0016]

[0017] Furthermore, the step (30) specifically includes:

[0018] 31) Detection threshold selection: In order to ensure that potential folding points are released, a lower threshold is used. Assuming that the threshold used for subsequent target detection is thd, the threshold Γ for multi-channel CFAR screening is:

[0019]

[0020] Among them, max(A, B) means taking the larger value between A and B;

[0021] 32) Over-detection peak point screening: retain the peak points that exceed the threshold to form M 0-1 over-detection sequences

[0022] 33) Target protection mechanism: shielding the points where the pulse echoes of the same distance unit are detected at the same time, i.e.

[0023]

[0024] In the above formula, ∏(·) represents a multiplication operation (i.e., logical AND).

[0025] The beneficial effects of the present invention are:

[0026] Compared with existing technologies:

[0027] (1) In a strong clutter environment, the staggered MTI mode may cause the folded echo to fall within the strong clutter area and fail to show sufficient amplitude differences between the receiving gates, rendering the existing folded asynchrony judgment method based on the echo amplitude of multiple receiving gates ineffective. The present invention directly performs folded asynchrony judgment based on the results of MTI cancellation, significantly reducing the impact of strong clutter background and achieving more thorough folded clutter elimination.

[0028] (2) Existing methods for determining folded asynchrony may mistakenly identify targets with interpulse echo amplitude modulation, such as civil aviation, as folded targets (or, to avoid misjudgment, it is necessary to increase the folding determination threshold and reduce the sensitivity of folded clutter determination). The present invention performs folded asynchrony determination based on the position relationship of the MTI checkpoints, achieving folded clutter suppression while having little impact on such targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a technical flow chart of the present invention.

[0030] Figure 2 is an example of collapsing a 0-1 reference sequence.

[0031] Figure 3 It is an example of multi-channel CFAR-like detection screening.

[0032] Figure 4 is an example of the calculation results of the fold confidence sequence.

[0033] Figure 5 This is an example of the folded clutter location determination process.

[0034] Figure 6 It is a comparison chart of the detection results before and after the folded clutter removal. DETAILED DESCRIPTION

[0035] The present invention will be further described in detail below with reference to the accompanying drawings. Figure 1 As shown in FIG, a method for removing clutter in a staggered MTI pattern folding under strong clutter includes the following steps:

[0036] 10) Calculation of folded position reference sequence: Based on the pulse interval of the staggered MTI waveform, the position of the folded echo when it enters the observation area is calculated in advance, and the folded position is represented as a 0-1 position reference sequence and stored.

[0037] Figure 2The calculation results of a 7-slide-5 (i.e., 3-pulse cancellation, 5 sliding windows, totaling 7 pulses) staggered MTI are given. Within one staggered MTI processing cycle, 7 pulses are emitted, with pulse intervals of [1209, 1140, 1100, 1358, 1100, 1120] μs, a sampling frequency of 5 MHz, and N = 2990 sampling points within the gate. The folded position reference sequence calculation process includes the following steps:

[0038] 11) Folding position traversal: gradually increase the delay of the echo until the m-th pulse echo falls into the m+q-th receiving wave gate, where q is the number of echo folding times. Figure 2 a shows the position of the echo falling into each receiving gate after q=1 folding.

[0039] (Note: 1. The absolute position of the folded echo will vary depending on the specific delay, but for a certain number of folds, the relative position relationship of the folding points remains unchanged. Therefore, for a given number of folds q, only one case that meets the above conditions needs to be selected. Figure 2 The result is a special case where all echoes just enter the next receiving wave gate; 2. In practical applications, the amplitude of high-order folded echoes is usually weak, so generally only the case of q = 1 folding is considered, that is, for a given staggered MTI parameter, only a set of reference sequences needs to be pre-calculated and stored).

[0040] 12) Folding reference sequence calculation: Calculate the reference sequence of the folding position of each sliding window MTI output according to the selected folding echo situation and the MTI sliding window length. Figure 2 In the case where the MTI sliding window length is 3, the first group of MTI folding position reference sequences is the logical AND of the folding position sequences in the receiving gates m=1 to 3, the second group of MTI folding position reference sequences is the logical AND of the folding position sequences in the receiving gates m=2 to 4, and so on. A total of 5 groups of MTI folding position sequences are obtained, which are recorded as Where x(m,n)=0 or 1, M is the number of staggered MTI sliding windows, and N is the number of distance units within the receiving wave gate, such as Figure 2 b~f are shown. The number of effective folding points in each reference sequence is Corresponding to Figure 2 b~f results The order is 2, 3, 3, 2, 3.

[0041] In addition, according to the logical AND of each sliding window MTI folding reference sequence, the folding position reference sequence of the complete MTI can be obtained, which is recorded as Where z(n) = 0 or 1, like Figure 2 g.

[0042] 13) Folding reference sequence expansion: Considering that the folded echo may come from a high-speed target, its echo position may move across the range gate between pulses. In this case, the folded position reference sequences x and z need to be expanded, that is,

[0043]

[0044] 20) MTI cancellation: Perform MTI cancellation on the pulse pressure echo to obtain M groups of sliding window cancellation output results like Figure 3 The blue solid line results in each sub-figure.

[0045] 30) Multi-channel CFAR screening: Perform low-threshold CFAR detection on the output results of M groups of sliding window cancellation, and add a target protection mechanism to screen out potential folding points. Figure 3 As shown, the multi-channel CFAR screening process includes the following steps:

[0046] 31) Detection threshold selection: In order to release as many potential folding points as possible, a low threshold should be used in this step. Assuming that the threshold used for subsequent target detection is thd (unit: decibel), the threshold Γ (unit: decibel) for multi-channel CFAR screening in this step can be calculated as follows:

[0047]

[0048] Here, max(A, B) means taking the larger value between A and B.

[0049] 32) Over-detection peak point screening: retain the peak points that exceed the threshold, such as Figure 3 As shown in the middle circle, M over-detection 0-1 sequences are formed

[0050]

[0051] Where thd(m,n) is the threshold for superimposing the CFAR background.

[0052] 33) Target protection mechanism: shield the points where the pulse echoes of the same distance unit are detected at the same time (for staggered MTI, Figure 2 Each group of 0-1 reference sequences will not have folding points at the same distance gate at the same time), and we get Figure 3 The purple vertical line in the middle corresponds to the checkpoint result (the checkpoint of the 1242th range gate is blocked), that is,

[0053]

[0054] In the above formula, ∏(·) represents a multiplication operation (i.e., logical AND).

[0055] 40) Confidence sequence calculation: Convolution operation is performed on the passed sequence y and the reference sequence x to obtain the confidence sequence used to determine whether there is a fold and the fold position

[0056]

[0057] where x m represents the mth row of matrix x, y m Represents the m-th row of matrix y. Figure 4 Given Figure 2 Reference sequence and Figure 3 The confidence sequence calculated from the inspection sequence.

[0058] (Note: In the above formula, the normalized convolution output of each group The maximum value is no greater than 1, and the confidence sequence r after integration also ranges from 0 to 1, which can be understood as the confidence level of the corresponding delay in the occurrence of folding noise.

[0059] 50) Folding judgment and elimination: Compare the confidence sequence r with the predetermined confidence threshold △ (0<△≤1), and determine the position of the folded echo based on the threshold-crossing point of the confidence sequence to achieve elimination.

[0060] like Figure 4 , using the preset confidence threshold △ = 0.6, there are two threshold points, corresponding to offsets δ = [2411, 3191]. According to the convolution characteristics, it can be known that the folding position reference sequence z( Figure 2 g) Translation δ-N = [-582, 201] to determine the folding point position, such as Figure 5 shown. Figure 5 The complete MTI result in medium blue is Figure 3 The non-coherent accumulation results of each sliding window MTI result. Figure 5 The echoes that are judged as folded echoes are removed and replaced with background noise, and the MTI folded clutter removal results are output, such as Figure 6 As shown in d.

[0061] Figure 6 Figures a through d show, in order, the pulse echo pulse pressure results, the MTI cancellation results, the results of processing using a traditional folded clutter removal method, and the results of processing using the present invention (detection threshold thd = 13dB). As can be seen, due to the strong clutter background in the near zone, the traditional method, which relies on the amplitude differences of the echoes at each gate, is unable to identify the folded clutter at locations 208, 514, 614, 948, and 1398, resulting in a large number of false alarms. Using the present invention's method, the folded clutter hidden within the strong clutter background in the near zone is successfully identified and removed.

[0062] The present invention is not limited to the above specific embodiments, and various modifications and variations are possible. Any modification, equivalent replacement, improvement, etc. made to the above embodiments based on the technical essence of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for eliminating clutter using staggered MTI pattern folding under strong clutter, characterized by: The steps include: 10) Folding position reference sequence calculation: Based on the staggered MTI waveform pulse interval, the position of the folded echo when it enters the observation area is calculated in advance and described as a 0-1 reference sequence Where x(m,n)=0 or 1, M is the number of staggered MTI sliding windows, N is the number of range units in the receiving wave gate, m = 1...M, n = 1...N; 20) MTI cancellation: Perform MTI cancellation on the pulse pressure echo to obtain M groups of sliding window cancellation output results 30) Multi-channel CFAR screening: Perform low-threshold CFAR detection on the output results of M groups of sliding window cancellation, and add a target protection mechanism to screen out potential folding point locations, which are described as over-detection sequences. Where y(m,n)=0 or 1, 40) Confidence sequence calculation: Perform normalized convolution operation on the passed sequence y and the reference sequence x to obtain the confidence sequence used to determine whether there is a fold and the fold position Among them, x m represents the mth row of the reference sequence x, y m represents the mth row of the reference sequence y, k m is x m represents the number of independent targets, Represents a sequential convolution operation; 50) Folding judgment and elimination: Compare the confidence sequence r with the predetermined confidence threshold △, 0<△≤1, and determine the folded echo position based on the threshold-crossing point of the confidence sequence to achieve elimination.

2. The method for eliminating clutter by using staggered MTI mode folding under strong clutter according to claim 1, characterized in that: The reference sequence calculation in step (10) includes: 11) Folding position traversal: gradually increase the delay of the echo until the m-th pulse echo falls into the m+q-th receiving wave gate, where q is the number of echo folding times; 12) Reference sequence calculation: Based on the selected folded echo situation and the MTI sliding window length, the reference sequence of the folded position of each sliding window MTI output is calculated, which is recorded as Where x(m,n)=0 or 1, M is the number of staggered MTI sliding windows, and N is the number of range units within the receiving gate; 13) Reference sequence extension: Considering that the folded echo may come from a high-speed target, its echo position may move across the range gate between pulses. To address this situation, the folded position reference sequence x needs to be extended, namely:

3. The method for eliminating clutter by using staggered MTI pattern folding under strong clutter according to claim 1, characterized in that: The step (30) specifically includes: 31) Detection threshold selection: In order to ensure that potential folding points are released, a lower threshold is used. Assuming that the threshold used for subsequent target detection is thd, the threshold Γ for multi-channel CFAR screening is: Among them, max(A, B) means taking the larger value between A and B; 32) Over-detection peak point screening: retain the peak points that exceed the threshold to form M 0-1 over-detection sequences 33) Target protection mechanism: shielding the points where the pulse echoes of the same distance unit are detected at the same time, i.e. In the above formula, ∏(·) represents a multiplication operation.

Images