A shock resistance method based on period estimation in clutter background

Abstract

The invention discloses an anti-impact method under clutter background based on period estimation, comprising the following steps: S1: performing group sliding window processing on echo data to obtain a noise floor for detection; S2: comparing all data in the group with the detection threshold to obtain all pulse units where the detection threshold value is exceeded; S3: grouping the time sampling units exceeding the detection threshold according to the impact to obtain the number of impacts, the impact start time and the impact end time; S4: using the obtained impact number, impact start time and impact end time to find a clutter data window without impact, taking different intervals in the clutter data window to calculate the sequence peak variance, and determine the clutter repetition period; S5: using the obtained clutter repetition period, replacing it with the values ​​of the same position of adjacent periods to complete the impact interference suppression; S6: comparing with the method based on time domain pulse cancellation. The purpose of the invention is to suppress the lightning impact interference received by high-frequency ground wave radar during sea detection.

Description

Technical Field

[0001] The invention belongs to the field of radar anti-shock, and in particular relates to an anti-shock method under a clutter background based on period estimation. Background Art

[0002] The electromagnetic environment of the working frequency band of high-frequency ground wave radar is complex and harsh. There are often a lot of interference and external noise in the high-frequency band. According to the duration, the interference that is much shorter than the radar coherent accumulation period is defined as impact interference or impact noise. Take lightning interference as an example. Its amplitude is very strong and its energy is higher than the internal noise of the radar. At the same time, the spectrum is also very wide, affecting almost all Doppler units. Therefore, the existence of impact interference will increase the noise floor of the radar, generally 5 to 20 dB higher, which will reduce the available time of the high-frequency ground wave radar by 25%, submerge the useful signal and deteriorate the radar detection performance.

[0003] Lightning impulse is a natural discharge phenomenon and a typical impulse interference. It occurs about once every 1-5 seconds in the thunderstorm season, and each time lasts about 200-600ms. For over-the-horizon (OTH) radars for sea detection, lightning impulses can reduce their sensitivity by 10dB or even more. For radars, lightning impulse signals are external interference signals and have nothing to do with the transmitted signal. For radar systems, radar impulse interference has an impact in both distance and Doppler frequency dimensions. In the distance dimension, the impact range of lightning impulse interference is related to the radar duration. The longer the duration, the larger the impact range. In the Doppler frequency dimension, due to the instantaneous characteristics of lightning impulse interference, the noise floor of the Doppler frequency dimension is raised. Whether in the distance dimension or the Doppler frequency dimension, radar impulse interference has an adverse effect on radar detection performance. For areas with high incidence of radar, dense radar impulse interference further reduces radar detection performance, and the existing technology has limited suppression effect on dense impulse interference. Therefore, it is necessary to study radar anti-impact interference technology. Summary of the invention

[0004] The purpose of the present invention is to suppress the lightning impulse interference received by high-frequency ground wave radar during sea detection. After processing the echo signal in the time domain, the time domain sliding window processing is used to find the time sampling position of the impulse, and then the echo period estimation algorithm is performed. The period is used to find the reference unit and replace the interference unit to achieve the suppression of the impulse interference.

[0005] The specific technical solutions of the present invention are as follows:

[0006] A shock resistance method under clutter background based on period estimation comprises the following steps:

[0007] The beneficial effects of the anti-shock method under clutter background based on period estimation of the present invention are as follows:

[0008] S1: First, divide nR distance units into n groups, each group contains R distance units, and first perform d i1 , d i2 , …, d iR Find the mean d i =(d i1 +d i2 +…+d iR ) / R; then the echo data is processed by pulse dimension grouping sliding window, each window has L pulse units, and d 1i , d 2i , …, d Li Find the mean S i =(d 1i +d 2i +…+d Li ) / L; then calculate the mean of all windows again, S = (S1+S2+…+S N ) / N, and obtain the noise floor used for detection;

[0009] S2: Compare all the data in the group with the detection threshold T. If N i >T, then it is judged that there is an impact. 1，i+1 , d 2，i+1 , …, d L，i+1 Perform the same process to obtain all Doppler units where the detection threshold value is exceeded;

[0010] S3: grouping the time sampling units exceeding the detection threshold by impact, and obtaining the number of impacts, the impact start time and the impact end time;

[0011] S4: Use the obtained impact number, impact start time and impact end time to find the clutter data window d1, d2, ..., d without impact in the original data. n , take different intervals i∈I=1,2,…,n in the noise data window to calculate the sequence peak variance Where [·] represents taking the integer part to determine the clutter repetition period

[0012] S5: Use the obtained clutter repetition period to replace the value at the same position in the previous n periods to complete the impulse interference suppression;

[0013] Preferably, S1 includes the following sub-steps:

[0014] S11: First, the nR range units of the original echo data are divided into n groups, each group contains R range units, and d i1 , d i2 , …, d iR Find the mean d i=(d i1 +d i2 +…+d iR ) / R;

[0015] S12: Using the local extremum search method based on sliding window detection, there are L pulse units in each window, and the pulse dimension d of the echo data in each window is 1i , d 2i , …, d Li Take the mean S i =(d 1i +d 2i +…+d Li ) / L;

[0016] S13: Calculate the mean of all windows again S = (S1 + S2 + ... + S N ) / N, as the detection noise floor;

[0017] Preferably, in S2, the pulse unit where the impulse interference is located is obtained by setting the detection threshold T to p times the noise floor.

[0018] Preferably, S3 includes the following sub-steps:

[0019] S31: After finding all the time sampling units where the larger values ​​are located, group them by impact, and classify the adjacent larger values ​​as the same impact, until a non-adjacent larger value appears, which is counted as the next impact;

[0020] S32: Store the number of impacts, as well as the start and end times of these impacts.

[0021] Preferably, S4 includes the following sub-steps:

[0022] S41: Setting the value range of cycle i to i∈I=1, 2, ..., n;

[0023] S42: Find a data window d1, d2, ..., d that does not contain an impact n ;

[0024] S43: In the data window without shock, sample at intervals i∈I=1, 2, ..., n, and calculate the variance of these sampling points Then the period estimate N is the interval that minimizes the variance.

[0025] Preferably, in S5, the impact unit is replaced with the impact-free unit of the previous period using the estimated value of the period obtained in S4.

[0026] 1. The purpose of the present invention is to suppress the lightning impulse interference received by high-frequency ground wave radar during sea detection. After processing the echo signal in the time domain, the pulse dimension position of the impulse is found using the time domain sliding window processing, and then the echo period estimation algorithm is performed. The reference unit is found using the period, and the interference unit is replaced to achieve the suppression of the impulse interference.

[0027] 2. The present invention utilizes a method for estimating the sea clutter period to eliminate the impact without destroying the sea clutter spectrum. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Flowchart of the present invention

[0029] Figure 2 Shock test results

[0030] Figure 3 24th distance unit suppression result

[0031] Figure 4 48th distance unit suppression result

[0032] Figure 5 200th distance unit suppression result

[0033] Figure 6 Distance-Doppler frequency spectrum before and after shock suppression

[0034] Figure 7 Comparison of Doppler profiles of different distance units before and after shock suppression DETAILED DESCRIPTION

[0035] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0036] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0037] like Figure 1 As shown, a method for resisting shocks in a clutter background based on period estimation comprises the following steps:

[0038] Step 1: Impact detection. Group the echo data, perform sliding window processing on each group of data, obtain the decision threshold, compare all the data in the group with the decision threshold, and obtain all the pulse units with larger values. Then group the pulse units with larger values ​​by impact, and obtain the impact number, impact start position, and impact end position;

[0039] Step 2: Period estimation. To estimate the period of the background noise sampling sequence, first find a data window without shocks, calculate the variance at different intervals within the shock window, and determine the period value;

[0040] Step 3: Impact suppression. Use the obtained cycle to replace the value of the same position in the previous n cycles to complete the cycle suppression. The present invention is used for radar anti-transient interference in the sea clutter background.

[0041] Taking a set of measured data as an example, the measured data pulse dimension has 5120 pulse units (or slow time sampling units), and a single pulse includes 300 distance units. The 1st to 75th sampling points in the pulse are the sea clutter-dominated areas, and impact interference occurs at the 2933rd, 2949th, 2963rd and 3202nd pulse units. Figure 2 shown.

[0042] In step 1: the echo data detected each time is grouped as the original data, and the grouping method is: the original matrix is ​​grouped along the distance dimension, and a group of 60 distance units is input into the impact detection function.

[0043] In step 1, when detecting the impact, a method based on sliding window detection to find the local extreme value is used to find the average value d of the time sampling unit of the echo matrix. i =(d i1 +d i2 +…+d iR ) / R, and get a column vector. The sliding window length is 1000 points, and each sliding is 500 points. The mean value S in each window is calculated. i =(d 1i +d 2i +…+d Li ) / L, and then calculate the pulse dimension mean S = (S1+S2+…+S N ) / N, and the detection background is obtained. The detection threshold is set according to the detection performance requirements, which is set to 4 times the noise floor. All data in the group are compared with the detection threshold, and the time sampling unit position of data greater than the detection threshold is recorded.

[0044] In step 1, after finding all the time sampling units where the larger values ​​are located, they are grouped by impact, and the adjacent larger values ​​are classified as the same impact until a non-adjacent larger value appears, which is counted as the next impact. A total of 4 impact interferences are detected.

[0045] In step 2, the obtained number of shocks and shock positions are used to estimate the period. First, a data window without shocks is found. If the shock is not in the first 50 points, the first 50 points are taken as the data window; if the shock is not in the last 50 points, the last 50 points are taken as the data window; if the interval between two adjacent shocks is greater than 50 points, 50 points in the interval are taken as the data window. The estimated number of periods is 10.

[0046] In step 3, the estimated periodic interference position is used for replacement. Figure 3 , Figure 4 , Figure 5 The amplitude and phase of the 24th, 48th and 200th distance units are given respectively, where the blue line (circle mark) is the result of the impact suppression method based on period estimation, and the red line (triangle mark) is the original data before impact resistance; the left figure is the amplitude spectrum, and the right figure is the phase spectrum. The arrow points to the impact interference replacement point.

[0047] Depend on Figure 3 , Figure 4 , Figure 5 It can be seen that when replacing the impact unit, only the unit where the impact is located is processed, and the other units remain unchanged. This can eliminate the impact while maintaining the integrity of the original sea clutter spectrum, which is beneficial to subsequent clutter suppression, target detection and other processing.

[0048] The result after replacing the period estimation unit is Fourier transformed along the pulse dimension. At the same time, the data is processed by the time domain pulse cancellation method, and the processed result is also Fourier transformed along the pulse dimension. The impact resistance results of the two methods are as follows Figure 6 As shown; and take the Doppler profiles of three distances in the near zone and far zone, and give the Doppler dimension amplitude before and after the two methods are processed, as shown Figure 7 shown.

[0049] It can be clearly seen from the above figure that the impulse suppression method based on period estimation can effectively suppress impulse interference, and compared with the time domain pulse cancellation method, the impulse suppression method based on period estimation has better suppression performance.

[0050] In summary, the beneficial effect of the present invention is that the impulse suppression method based on period estimation with low computational complexity can suppress the impulse interference under the sea clutter background, and can effectively improve the impulse suppression performance under the sea clutter background. And by using the method of estimating the period of sea clutter, the sea clutter spectrum is not destroyed while the impulse is eliminated.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A shock resistance method based on period estimation in a clutter background, characterized in that: The following steps are involved: S1: First, divide nR distance units into n groups, each group contains R distance units, and first perform d i1 , d i2 , …, d iR Find the mean d i =(d i1 +d i2 +…+d iR ) / R; then the echo data is processed by pulse dimension grouping sliding window, each window has L pulse units, and d 1i , d 2i , …, d Li Find the mean S i =(d 1i +d 2i +…+d Li ) / L; then calculate the mean of all windows again, S = (S1+S2+…+S N ) / N, and obtain the noise floor used for detection; S2: Compare all the data in the group with the detection threshold T. If N i >T, it is judged that there is an impact, and the next set of data d is then 1，i+1 , d 2，i+1 , …, d L，i+1 Perform the same process to obtain all pulse units that exceed the detection threshold value; S3: grouping the time sampling units exceeding the detection threshold by impact, and obtaining the number of impacts, the impact start time and the impact end time; S4: Use the obtained impact number, impact start time and impact end time to find the clutter data window d1, d2, ..., d without impact in the original data. n , take different intervals i∈I=1,2,…,n in the noise data window to calculate the sequence peak variance Where [·] represents taking the integer part to determine the clutter repetition period S5: Use the obtained clutter repetition period to replace it with the value at the same position in the previous n periods to complete the impulse interference suppression.

2. The anti-shock method based on period estimation under clutter background according to claim 1 is characterized in that: The S1 includes the following sub-steps: S11: First, the nR range units of the original echo data are divided into n groups, each group contains R range units, and d i1 , d i2 , …, d iR Find the mean d i =(d i1 +d i2 +…+d iR ) / R; S12: Using the local extremum search method based on sliding window detection, the pulse dimension d of the echo data is 1i , d 2i , …, d Li Take the mean S i =(d 1i +d 2i +…+d Li ) / L; S13: Calculate the mean of all windows again S = (S1 + S2 + ... + S N ) / N, as the detection noise floor.

3. The anti-shock method based on period estimation under clutter background according to claim 1 is characterized in that: In S2, the pulse unit where the impulse interference is located is obtained by setting the detection threshold T to p times the noise floor.

4. The anti-shock method based on period estimation under clutter background according to claim 1 is characterized in that: The S3 includes the following sub-steps: S31: after finding all the time sampling units where the values ​​exceeding the detection threshold are located, group them by impact, and classify the adjacent values ​​exceeding the detection threshold as the same impact, until a non-adjacent value exceeding the detection threshold appears, which is counted as the next impact; S32: Store the number of impacts, as well as the start and end times of these impacts.

5. The anti-shock method based on period estimation under clutter background according to claim 1 is characterized in that: The S4 includes the following sub-steps: S41: Setting the value range of cycle i to i∈I=1, 2, ..., n; S42: Find a data window d1, d2, ..., d without impact n ; S43: In the data window without shock, sample at intervals i∈I=1, 2, ..., n, and calculate the variance of these sampling points Then the period estimate N is the interval that minimizes the variance.

6. The anti-shock method based on period estimation under clutter background according to claim 1 is characterized in that: In S5, the period estimation value N obtained in S4 is used to replace the impacted time sampling units with the impact-free time sampling units of the adjacent period.

Images