Radar signal sorting method based on multi-parameter clustering joint correction PRI transformation

Abstract

The invention discloses a radar signal sorting method based on multi-parameter clustering joint correction PRI transformation, and belongs to the technical field of information. The invention designs a new radar signal sorting algorithm, which abandons the previous sorting which only depends on a single pulse parameter, fully utilizes the radar signal information in the pre-sorting stage, dilutes the high-intensity pulse flow through a clustering mode, and then performs main sorting on the diluted pulse flow through a corrected PRI transformation method. And the defects of poor sorting effect and long time consumption caused by directly using PRI transformation are avoided. The method effectively improves the sorting capability of the radar signal sorting algorithm on the types of the jitter signals, gives consideration to the real-time performance while guaranteeing the accuracy of the sorting algorithm, and provides a guarantee for the real-time sorting of the radar signals.

Description

Technical Field

[0001] The invention belongs to the field of information technology, and in particular relates to a radar signal sorting method based on multi-parameter clustering joint modified PRI transformation. Background Art

[0002] With the rapid development of radar technology, in modern electromagnetic environments, the density of radar pulses can reach hundreds of thousands per second. High-density radar pulses make different radar signals have similar carrier frequencies and pulse widths, interlaced and overlapped in the time domain and frequency domain, and the parameter values ​​measured by the pulse stream are seriously distorted, thus affecting the accuracy of signal sorting. Jitter radar signals are a common type of radar signal, and common algorithms are more difficult to extract them effectively, which increases the difficulty of signal sorting. In order to improve the performance of radar reconnaissance systems and realize electromagnetic environment spectrum analysis, how to quickly and effectively sort jittered aliased radar signals is an urgent problem to be solved.

[0003] The patent "A radar signal sorting method and system based on multi-parameter clustering" (202111170713.9) provides a sorting method based on multi-parameter clustering. The method divides radar pulses into different bands; hierarchical clustering is performed based on the carrier frequency and pulse width of the pulse signal to divide the pulse signal into multiple clusters; clusters composed of messy pulses are screened out based on the arrival time difference of adjacent pulse signals in the same cluster; all retained pulse signals are numbered according to the order of arrival, and messy pulse signals are screened out based on the order difference of pulse signals in the cluster; the distance between two clusters is measured according to the start and end time, carrier frequency, pulse width and other information of all pulse signals in each cluster, and an inter-cluster distance matrix is ​​constructed; the number of cluster categories is calculated based on the inter-cluster distance matrix, and cluster sorting is completed according to the spectral information of the inter-cluster distance matrix. The invention makes full use of the available information in the radar signal, combines multiple parameters such as the carrier frequency, pulse width and arrival time of the pulse signal, and quickly maps the pulse stream to different cluster spaces through clustering, thereby improving the real-time and accuracy of radar signal sorting.

[0004] The patent "A method for sorting PRI jitter radar signals in the case of pulse loss and aliasing" (202011072664.0) provides a method for sorting PRI jitter radar signals in the case of pulse loss and aliasing. This method improves the PRI overlapping box structure and uses a multi-stage box structure to improve the jitter signal detection capability in the case of pulse loss and aliasing; by analyzing the PRI transformation results of the multi-stage box and the number curve of pulse pairs in the box, the jitter rate of the jitter signal is judged to improve the signal analysis capability; combining the PRI estimate and the jitter rate, the confidence is judged by correlation, and the pulse sequence is extracted according to the situation, effectively reducing the complexity of the electromagnetic space. It can be proved that the sorting algorithm of this method has good performance and is easy to implement in engineering.

[0005] The existing technical shortcomings are mainly reflected in the following aspects: First, most radar signals exist in the form of pulse signals in the electromagnetic environment. The pulse signals are densely distributed in the time domain and there is an interleaving phenomenon. Most sorting algorithms rely only on the TOA single parameter for signal sorting. The sample input is single, and it is difficult to distinguish signals with the same PRI period, so the reliability is low. Second, the radar waveform is complex, and the types of PRI signal changes include slip, jitter, sine, and jitter. The problems of missing and adding batches in the radar signal sorting process will become more and more serious, and the traditional PRI transformation algorithm is difficult to effectively sort them. Third, due to the highly intensive characteristics of the electromagnetic environment, the amount of calculation during the algorithm operation is greatly increased, which makes the sorting system saturated, and ultimately leads to sorting failure and information loss. Summary of the invention

[0006] In view of the above technical problems existing in the prior art, the present invention proposes a radar signal sorting method based on multi-parameter clustering and joint modified PRI (pulse repetition interval) transformation, which has a reasonable design, overcomes the shortcomings of the prior art and has good effects.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation includes the following steps:

[0009] Step 1: Obtain the aliased radar signal pulse signal data to be sorted;

[0010] Step 2: Feature parameter extraction and data preprocessing;

[0011] Step 3: pre-sorting signal pulse flow dilution;

[0012] Step 4: Use the modified PRI transformation method for primary sorting;

[0013] Step 5: Pulse sequence search and extraction.

[0014] Preferably, in step 1, radar pulse description word data with different TOA, PW, PA, CF, and DOA types are generated by simulation through the MATLAB platform, the signal carrier frequency range is 5 to 10 GHz, the pulse width range is 0.4 to 150 us, the arrival angle range is 60 to 120°, and the PRI type includes two types: fixed and jitter.

[0015] Preferably, in step 2, in the data preprocessing stage, the data set is analyzed and feature parameters are extracted, and normalization is used; the original data is normalized after removing abnormal points, so that the preprocessed data is limited to [-1, 1], and each indicator is at the same level. The normalized conversion function is:

[0016]

[0017] Among them, x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x' is the normalized data.

[0018] Preferably, in step 3, the sorting of unknown radar signals lacks prior information, and the DBSCAN density clustering algorithm is used to cluster and pre-sort the three parameters of CF, DOA, and PW to achieve dilution of the dense pulse flow of multiple aliased radars.

[0019] Preferably, in step 4, the main sorting is performed on the basis of the pre-sorting, mainly to complete the extraction process of the pulse repetition period; using the modified PRI transformation method, the TOA parameters after pre-sorting are subjected to PRI transformation by utilizing overlapping PRI boxes and setting a variable time starting point, so as to improve the detection capability of the jitter signal.

[0020] Preferably, the steps of performing primary separation using the modified PRI transformation method are as follows:

[0021] Step 4.1: Read the diluted radar pulse stream data, initialize the discrete form PRI transformation method, establish PRI boxes at different overlap rates, and determine the PRI box center and the PRI box width at each overlap rate;

[0022] Step 4.2: Let m = n-1, where m, n = 0, 1, ..., N-1, n> m; N is the total number of pulses;

[0023] Step 4.3: Let τ = t n -t m , if τ<(1-ε)τ min , then go to step 4.6, if τ ≥ (1 + ε) τ max , then go to step 4.12;

[0024] Step 4.4: Calculate the range of k,

[0025] Step 4.5: Repeat steps 4.6 to 4.10 for all k in [k1,k2];

[0026] Step 4.6: Initialize the starting point of time. If the value of k appears for the first time, let O k =t n ;

[0027] Step 4.7: Calculate the initial phase for decomposition, η0 = (t n -O k )÷τ k,v=(η0+0.499999...),ξ=η0 / v-1;

[0028] Step 4.8: If any of the following two conditions are met, set O k =t n ;

[0029] (a) v = 1 and t m =O k

[0030] (b) v ≥ 2 and ξ ≤ ξ 0

[0031] Step 4.9: Calculate the phase η = (t n -O k )÷τ k ;

[0032] Step 4.10: Update PRI transformation D k ,D k =D k +e jπiη ;

[0033] Step 4.11: Let m=m-1. If m<1, go to step 4.12. Otherwise, go to step 4.3.

[0034] Step 4.12: Let n=n+1. If n>N, end. Otherwise, go to step 4.2.

[0035] Step 4.13: According to the observation time principle, sub-harmonic elimination principle and noise elimination principle, the threshold of the modified PRI transformation method is set to Among them, α, β, and γ are artificially set threshold setting parameters, T is the observation time, and the threshold setting parameters change dynamically with the algorithm operation status;

[0036] Step 4.14: Search for the maximum value of the part exceeding the threshold in the PRI transformation result. If there are multiple maximum values ​​within the set interval, take the largest one.

[0037] Preferably, in step 5, after the primary sorting algorithm estimates the PRI value, the pulse sequence is searched according to the PRI value, and the pulse sequence belonging to the PRI is extracted.

[0038] Beneficial technical effects brought by the present invention:

[0039] In order to overcome the problems of low accuracy and low real-time performance of the existing technology in sorting aliased radar signals in complex electromagnetic environments, the present invention designs a new radar signal sorting algorithm. The present invention abandons the previous sorting that only relies on a single pulse parameter, fully utilizes radar signal information in the pre-sorting stage, dilutes the high-density pulse stream by clustering, and then performs main sorting on the diluted pulse stream by modifying the PRI transformation method, avoiding the shortcomings of poor sorting effect and long time consumption of directly using the PRI transformation. The present invention effectively improves the radar signal sorting algorithm's ability to sort jitter signal types, takes into account real-time performance while ensuring the accuracy of the sorting algorithm, and provides a guarantee for real-time sorting of radar signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The flowchart of radar signal sorting method using multi-parameter clustering and joint correction of PRI algorithm;

[0041] Figure 2 It is a flowchart of multi-parameter clustering pre-sorting;

[0042] Figure 3 This is the flow chart of the main sorting algorithm for the modified PRI transformation method. DETAILED DESCRIPTION

[0043] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0044] A radar signal sorting method based on multi-parameter clustering and joint modified PRI transform, the process is as follows Figure 1 As shown, the following steps are included:

[0045] Step 1: Obtain the aliased radar signal pulse signal data to be sorted;

[0046] Radar pulse description word data with different TOA, PW, PA, CF, and DOA types are generated through MATLAB platform simulation. The signal carrier frequency range is 5 to 10 GHz, the pulse width range is 0.4 to 150 us, the arrival angle range is 60 to 120°, and the PRI type includes fixed and jitter types.

[0047] Step 2: Feature parameter extraction and data preprocessing;

[0048] In the data preprocessing stage, the data set is analyzed and feature parameters are extracted, and normalization is used; the original data is normalized after outliers are removed, so that the preprocessed data is limited to [-1,1] and each indicator is at the same level. The normalized conversion function is:

[0049]

[0050] Among them, x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x' is the normalized data.

[0051] Step 3: pre-sorting signal pulse flow dilution;

[0052] The sorting of unknown radar signals lacks prior information, so the DBSCAN density clustering algorithm is used to cluster and pre-sort the three parameters of CF, DOA, and PW to achieve the dilution of the dense pulse flow of multiple aliased radars. Figure 2 shown.

[0053] Step 4: Use the modified PRI transformation method for primary sorting;

[0054] The main sorting is performed on the basis of the pre-sorting, and mainly completes the extraction process of the pulse repetition period; the modified PRI transformation method is used to perform PRI transformation on the TOA parameters after pre-sorting by using overlapping PRI boxes and setting a variable time starting point, thereby improving the detection capability of the jitter signal. Figure 3 shown.

[0055] The steps for primary sorting using the modified PRI transformation method are as follows:

[0056] Step 4.1: Read the diluted radar pulse stream data, initialize the discrete form PRI transformation method, establish PRI boxes at different overlap rates, and determine the PRI box center and the PRI box width at each overlap rate;

[0057] Step 4.2: Let m = n-1, where m, n = 0, 1, ..., N-1, n> m; N is the total number of pulses;

[0058] Step 4.3: Let τ = t n -t m , if τ<(1-ε)τ min , then go to step 4.6, if τ ≥ (1 + ε) τ max , then go to step 4.12;

[0059] Step 4.4: Calculate the range of k,

[0060] Step 4.5: Repeat steps 4.6 to 4.10 for all k in [k1,k2];

[0061] Step 4.6: Initialize the starting point of time. If the value of k appears for the first time, let O k =t n ;

[0062] Step 4.7: Calculate the initial phase for decomposition, η0 = (t n-O k )÷τ k ,v=(η0+0.499999...),ξ=η0 / v-1;

[0063] Step 4.8: If any of the following two conditions are met, set O k =t n ;

[0064] (a) v = 1 and t m =O k

[0065] (b) v ≥ 2 and ξ ≤ ξ 0

[0066] Step 4.9: Calculate the phase η = (t n -O k )÷τ k ;

[0067] Step 4.10: Update PRI transformation D k ,D k =D k +e jπiη ;

[0068] Step 4.11: Let m=m-1. If m<1, go to step 4.12. Otherwise, go to step 4.3.

[0069] Step 4.12: Let n=n+1. If n>N, end. Otherwise, go to step 4.2.

[0070] Step 4.13: According to the observation time principle, sub-harmonic elimination principle and noise elimination principle, the threshold of the modified PRI transformation method is set to Among them, α, β, and γ are artificially set threshold setting parameters, T is the observation time, and the threshold setting parameters change dynamically with the algorithm operation status;

[0071] Step 4.14: Search for the maximum value of the part exceeding the threshold in the PRI transformation result. If there are multiple maximum values ​​within the set interval, take the largest one.

[0072] Step 5: Pulse sequence search and extraction.

[0073] After the primary sorting algorithm estimates the PRI value, the pulse sequence is searched according to the PRI value, and the pulse sequence belonging to the PRI is extracted.

[0074] The aliasing radar sorting data of the present invention comes from jitter and fixed type radar signal samples.

[0075] The present invention combines the carrier frequency, pulse width and arrival angle of the pulse signal to make full use of the available information in the radar signal, and uses the DBSACAN algorithm to quickly divide the aliased pulse stream into different clusters, thereby improving the real-time performance of radar signal sorting.

[0076] The present invention uses a modified PRI change method to perform PRI transformation on a pulse stream that has been pre-sorted and diluted, and improves the signal sorting capability of a jitter type by changing the PRI box width.

[0077] The present invention protects a signal analysis unit in a signal analysis instrument for sorting, analyzing and identifying aliased radar signals.

[0078] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation, characterized in that: The steps include: Step 1: Obtain the aliased radar signal pulse signal data to be sorted; Step 2: Feature parameter extraction and data preprocessing; Step 3: pre-sorting signal pulse flow dilution; Step 4: Use the modified PRI transformation method for primary sorting; Step 5: Pulse sequence search and extraction.

2. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 1 is characterized in that: In step 1, radar pulse description word data with different TOA, PW, PA, CF, and DOA types are generated through MATLAB platform simulation. The signal carrier frequency range is 5 to 10 GHz, the pulse width range is 0.4 to 150 us, the arrival angle range is 60 to 120°, and the PRI type includes fixed and jitter types.

3. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 1 is characterized in that: In step 2, in the data preprocessing stage, the data set is analyzed and feature parameters are extracted, and normalization is used; the original data is normalized after outliers are removed, so that the preprocessed data is limited to [-1,1], and each indicator is at the same level. The normalized conversion function is: Among them, x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x' is the normalized data.

4. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 1 is characterized in that: In step 3, the sorting of unknown radar signals lacks prior information, and the DBSCAN density clustering algorithm is used to cluster and pre-sort the three parameters of CF, DOA, and PW to dilute the dense pulse flow of multiple aliased radars.

5. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 1 is characterized in that: In step 4, the main sorting is performed on the basis of the pre-sorting, mainly to complete the extraction process of the pulse repetition period; the modified PRI transformation method is used to perform PRI transformation on the TOA parameters after pre-sorting by using overlapping PRI boxes and setting a variable time starting point, thereby improving the detection capability of the jitter signal.

6. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 5 is characterized in that: The steps for primary sorting using the modified PRI transformation method are as follows: Step 4.1: Read the diluted radar pulse stream data, initialize the discrete form PRI transformation method, establish PRI boxes at different overlap rates, and determine the PRI box center and the PRI box width at each overlap rate; Step 4.2: Let m = n-1, where m, n = 0, 1, ..., N-1, n> m; N is the total number of pulses; Step 4.3: Let τ = t n -t m , if τ<(1-ε)τ min , then go to step 4.6, if τ ≥ (1 + ε) τ max , then go to step 4.12; Step 4.4: Calculate the range of k, Step 4.5: Repeat steps 4.6 to 4.10 for all k in [k1,k2]; Step 4.6: Initialize the starting point of time. If the value of k appears for the first time, let O k =t n ; Step 4.7: Calculate the initial phase for decomposition, η0 = (t n -O k )÷τ k ,v=(η0+0.499999...),ξ=η0 / v-1; Step 4.8: If any of the following two conditions are met, set O k =t n ; (a) v = 1 and t m =O k (b) v ≥ 2 and ξ ≤ ξ 0 Step 4.9: Calculate the phase η = (t n -O k )÷τ k ; Step 4.10: Update PRI transformation D k ,D k =D k +e jπiη ; Step 4.11: Let m=m-1. If m<1, go to step 4.

12. Otherwise, go to step 4.

3. Step 4.12: Let n=n+1. If n>N, end. Otherwise, go to step 4.

2. Step 4.13: According to the observation time principle, sub-harmonic elimination principle and noise elimination principle, the threshold of the modified PRI transformation method is set to Among them, α, β, and γ are artificially set threshold setting parameters, T is the observation time, and the threshold setting parameters change dynamically with the algorithm operation status; Step 4.14: Search for the maximum value of the part exceeding the threshold in the PRI transformation result. If there are multiple maximum values ​​within the set interval, take the largest one.

7. The radar signal sorting method based on multi-parameter clustering and joint modified PRI transformation according to claim 5 is characterized in that: In step 5, after the primary sorting algorithm estimates the PRI value, the pulse sequence is searched according to the PRI value, and the pulse sequence belonging to the PRI is extracted.

Images