A radar signal sorting method and system based on PRI interval information

Abstract

The present invention relates to a radar signal sorting method and system based on PRI interval information, belonging to the field of radar signal processing. First, a multi-level difference calculation is performed on the arrival time series of mixed overlapping radar pulse streams to generate a difference sequence; according to the distribution characteristics of the difference sequence, the neighborhood radius and density threshold of the density clustering algorithm are determined; according to the neighborhood radius and density threshold, the density clustering algorithm is used to perform density clustering on the difference sequence to obtain multiple clusters after clustering; the extraction priority of each cluster is calculated, and the PRI interval information to be extracted is generated for the cluster with the highest priority; an improved pulse sequence search method is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted; the improved pulse sequence search method is a pulse sequence search method with tolerance control. The method can improve the PRI valuation accuracy and information dimension, thereby improving the sorting success rate of radar pulse signals.

Description

Technical Field

[0001] The present invention relates to the field of radar signal processing, and in particular to a radar signal sorting method and system based on PRI interval information. Background Art

[0002] Radar signal sorting is one of the key technologies of electronic warfare equipment in modern battlefields. The sorting results directly affect the subsequent identification, precise positioning, threat judgment and confrontation strategy of the target. Common signal sorting is mainly based on two types of parameters. One type is non-time parameters such as pulse width, carrier frequency, azimuth, pulse amplitude and modulation type. The pulse signals from different radars are separated by classification algorithms; the other type is to extract the secondary parameter pulse repetition interval using the pulse arrival time (TOA), which contains the repetition law characteristics of the same radar signal in the time domain, and extract the pulse signal from the same radar from the aliased pulse in the form of a pulse train. With the increasing complexity of the modern electromagnetic environment, there may be a large number of radar radiation sources in the same time and space. The pulse width, carrier frequency and other parameters of the same radar will also have jitter or even jump modulation, and there will be serious overlap in the parameter space, which makes it difficult for the sorting method based on this type of parameters to adapt to the modern battlefield environment. Only by sorting out the pulse sequence of each radar signal from the randomly overlapping signal stream can the radar parameters be measured, analyzed, identified and the target intercepted.

[0003] The basic principle of radar signal sorting is to analyze various parameters of intercepted signals and automatically separate the randomly overlapping pulse signal streams into separate pulse sequences of each radar. The main parameters used for radar signal sorting are pulse direction of arrival (DOA), carrier frequency (RF), pulse width (PW), pulse amplitude (PA), etc. The existing signal sorting methods are based on pre-sorting according to these radar parameters, and then use the pulse arrival time to perform the main sorting of the pulse sequence, that is, pulse repetition interval (PRI) sorting, so as to complete the identification and interception of the target. Among them, the commonly used sorting methods based on pulse repetition interval PRI mainly include dynamic extended correlation method, cumulative difference histogram method, sequence difference histogram method, PRI transformation method, plane transformation method, etc., but the PRI values ​​estimated by these methods are not accurate, and there is a common problem of low PRI estimation accuracy. The estimation accuracy of PRI value directly affects the accuracy of pulse train extraction, and ultimately affects the sorting success rate. When the PRI estimation error is large, it will lead to a large number of pulse signal sorting errors or even sorting failures. Therefore, how to improve the PRI estimation accuracy in order to improve the sorting success rate of the existing PRI information sorting method is an urgent problem to be solved in radar signal sorting under complex environments. Summary of the invention

[0004] The purpose of the present invention is to provide a radar signal sorting method and system based on PRI interval information to improve the PRI estimation accuracy of mixed overlapping pulses in complex environments, thereby achieving the purpose of improving the radar pulse signal sorting success rate and solving the problems of low PRI estimation accuracy and low sorting success rate in existing radar signal sorting methods.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] On the one hand, the present invention proposes a radar signal sorting method based on PRI interval information, comprising:

[0007] Perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence;

[0008] Determining the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence;

[0009] According to the neighborhood radius and the density threshold, the difference sequence is density clustered using a density clustering algorithm to obtain a plurality of clusters after clustering;

[0010] Calculate the extraction priority of each cluster and generate the PRI interval information to be extracted for the cluster with the highest priority;

[0011] An improved pulse sequence search method is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching the pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as the tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted to extract the corresponding radar pulse signal.

[0012] Optionally, before the step of "performing multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence", the method further includes:

[0013] deriving the arrival time series of mixed overlapping pulses from the radar reconnaissance data to be sorted;

[0014] judging whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence;

[0015] When the number of elements in the arrival time series is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and the step of "performing multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence" is executed;

[0016] When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated.

[0017] Optionally, after the step of “using a pulse sequence search method with tolerance control to extract a radar pulse signal corresponding to the PRI interval information to be extracted”, the method further includes:

[0018] With the remaining radar pulse stream as input, return to the step of "determining whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence" to extract the remaining radar pulse signals until it is determined that there is no radar target.

[0019] Optionally, performing multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream to generate a difference sequence specifically includes:

[0020] The arrival time series TOA1, TOA2, ..., TOA of the mixed overlapping radar pulse stream n Perform multi-level difference calculation, where TOA1 represents the arrival time of the first pulse, TOA n represents the arrival time of the nth pulse, then the multi-level difference of the arrival time series is expressed as:

[0021] diff1=[TOA2-TOA1,TOA3-TOA2,…,TOA n -TOA n-1 ]

[0022] diff2=[TOA3-TOA1,TOA4-TOA2,…,TOA n -TOA n-2 ]

[0023] diff3=[TOA4-TOA1,TOA5-TOA2,…,TOA n -TOA n-3 ]

[0024]

[0025] Among them, diff1, diff2, and diff3 represent the first-level difference sequence, the second-level difference sequence, and the third-level difference sequence of the pulse arrival time, respectively. The generated difference sequence D is expressed as:

[0026] D=[diff1,diff2,diff3,…].

[0027] Optionally, determining the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence specifically includes:

[0028] The distance distribution matrix DIST is generated according to the difference sequence D, which is expressed as:

[0029] DIST=[dist(p,q)|p,q∈D]

[0030] Where p and q are any elements in the difference sequence D, and dist represents the Euclidean distance function;

[0031] Rearrange the distance values ​​of each row in the distance distribution matrix DIST from small to large. The kth distance value of each row conforms to the Poisson distribution in terms of statistical characteristics. Then the expected value of the kth distance value is expressed as:

[0032]

[0033] Where λ represents the expected value, i = 1, 2, ..., n, n represents the data dimension of the difference sequence D, and k∈[1, n] is a positive integer;

[0034] The average of the 4th distance values ​​of each row in the distance distribution matrix DIST is taken as the domain radius ε; the density threshold MinPts is the average number of points in the 4 neighborhoods of all data points, that is, the density threshold MinPts is equal to 4.

[0035] Optionally, the step of performing density clustering on the difference sequence using a density clustering algorithm according to the neighborhood radius and the density threshold to obtain a plurality of clusters after clustering specifically includes:

[0036] The DBSCAN density clustering algorithm is used to perform density clustering on the difference sequence according to the neighborhood radius and density threshold, and multiple clusters are formed after density clustering.

[0037] Optionally, calculating the extraction priority of each cluster and generating the PRI interval information to be extracted for the cluster with the highest priority specifically includes:

[0038] For the multiple clusters obtained after density clustering, calculate the extraction priority of each cluster separately:

[0039] Priority(i)=Cn(i) / [Cc(i)×Cj(i)]

[0040] Among them, Priority(i) represents the extraction priority of the PRI interval information of the i-th cluster, i=1,2,…,m, m represents the number of clusters, Cn(i) represents the number of points in the class corresponding to each cluster, Cc(i) represents the mean value of the points in the class corresponding to each cluster, and Cj(i) represents the length of the distribution interval of the points in the class corresponding to each cluster;

[0041] Then there is a positive integer k that satisfies:

[0042] Priority(k)≥Priority(i)

[0043] Among them, Priority(k) represents the extraction priority of the PRI interval information of the kth cluster;

[0044] At this time, the extraction priority corresponding to the kth cluster is the highest, so the PRI interval information of the cluster is extracted, and the format of the generated PRI interval information to be extracted is expressed as (PRI_cen, jitter);

[0045] Among them, PRI_cen=Cc(k) represents the center value, and jitter=Cj(k) / (2×PRI_cen) represents the jitter amplitude.

[0046] On the other hand, the present invention also proposes a radar signal sorting system based on PRI interval information, comprising:

[0047] A difference sequence generation module is used to perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence;

[0048] A density clustering parameter determination module, used to determine the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence;

[0049] A density clustering processing module, used to perform density clustering on the difference sequence according to the neighborhood radius and the density threshold using a density clustering algorithm to obtain a plurality of clusters after clustering;

[0050] The extraction priority calculation and PRI information generation module is used to calculate the extraction priority of each cluster and generate the PRI interval information to be extracted for the cluster with the highest priority;

[0051] The tolerance control and signal extraction module is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted by adopting an improved pulse sequence search method; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching the pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as the tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted to extract the corresponding radar pulse signal.

[0052] Optionally, also include:

[0053] An arrival time series acquisition module is used to derive the arrival time series of mixed overlapping pulses from the radar reconnaissance data to be sorted;

[0054] A radar target judgment module, used for judging whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence;

[0055] When the number of elements in the arrival time series is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and a difference sequence generation module is used to perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence;

[0056] When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated;

[0057] The radar pulse signal repeated extraction module is used to take the remaining radar pulse stream as input and return to the step of "determining whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence" to extract the remaining radar pulse signals until it is determined that there is no radar target.

[0058] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0059] The present invention proposes a radar signal sorting method and system based on PRI interval information. The method generates a difference sequence based on the arrival time by performing multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream. The difference sequence is density clustered by using a density clustering algorithm, and the priority of each cluster when extracting the PRI interval information is calculated. The PRI interval information is extracted in sequence according to the priority order, so that radar pulses with a large number of pulses, a small pulse interval and a small jitter amplitude can be preferentially extracted, thereby preferentially ensuring a higher extraction success rate and reducing the difficulty of subsequent pulse sorting.

[0060] After extracting the PRI interval information including the center value and the jitter amplitude, the present invention combines the pulse sequence search method with tolerance control to complete the sorting of the mixed overlapping pulse stream. In addition, the pulse sequence search method with tolerance control is adopted. The tolerance control refers to setting the TOA tolerance range when extracting the corresponding pulse according to different jitter amplitudes, thereby realizing flexible adjustment of the tolerance value, solving the problem that the setting of the tolerance value is too small, which is not conducive to the extraction of the pulse sequence with a large jitter amplitude, and the setting is too large, which is easy to cause wrong sorting, thereby improving the PRI valuation accuracy and sorting success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. The following drawings are not deliberately scaled according to the actual size, and the focus is on illustrating the main purpose of the present invention.

[0062] Figure 1 A flowchart of a radar signal sorting method based on PRI interval information provided in Example 1 of the present invention;

[0063] Figure 2 A principle flow chart of a radar signal sorting method based on PRI interval information provided in Example 1 of the present invention;

[0064] Figure 3 A schematic diagram of TOA difference scatter distribution before density clustering of radar pulse signals provided in Example 1 of the present invention;

[0065] Figure 4 A schematic diagram of TOA difference scatter distribution after density clustering of radar pulse signals provided in Example 1 of the present invention;

[0066] Figure 5 A curve comparison diagram of PRI estimation error rates of the present invention, the CDIF method and the PRI conversion method under different jitter amplitudes provided in Example 1 of the present invention;

[0067] Figure 6 A curve comparison diagram of the sorting success rates of the present invention, the CDIF method and the PRI conversion method under different jitter amplitudes provided in Example 1 of the present invention;

[0068] Figure 7 A schematic diagram of the PRI value extraction result of a pulse stream provided in Example 1 of the present invention;

[0069] Figure 8 A comparison chart of the experimental results of sorting success rates of the present invention, the CDIF method and the PRI transformation method provided in Example 1 of the present invention;

[0070] Fig. 9 This is a structural block diagram of a radar signal sorting system based on PRI interval information provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0072] As shown in the present invention and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0073] Although the present invention has made various references to certain modules in the system according to an embodiment of the present invention, any number of different modules can be used and run on a user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0074] The present invention uses a flow chart to illustrate the operations performed by the system according to an embodiment of the present invention. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0075] The purpose of the present invention is to provide a radar signal sorting method and system based on PRI interval information to improve the PRI estimation accuracy and information dimension of mixed overlapping pulses in complex environments, thereby achieving the purpose of improving the radar pulse signal sorting success rate and solving the problems of low PRI estimation accuracy and low sorting success rate in existing radar signal sorting methods.

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0077] Example 1

[0078] like Figure 1 As shown, this embodiment provides a radar signal sorting method based on PRI interval information, and the method specifically includes the following steps:

[0079] Step S1, deriving the arrival time sequence of mixed overlapping pulses from the radar reconnaissance data to be sorted.

[0080] Step S2: judging whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence.

[0081] When determining whether there is a radar target in the arrival time sequence, when the number of elements in the arrival time sequence is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and radar signal sorting is continued at this time, and step S3 is executed;

[0082] When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated.

[0083] In this embodiment, the first preset threshold is set to 20, and the second preset threshold is set to 0.1 / ms. If the conditions are met, step S3 is entered, otherwise it is determined that there is no radar target and the sorting is terminated. It is easy to understand that these two preset thresholds are not fixed and unique, and the judgment standard can be flexibly adjusted according to the actual application scenario and object.

[0084] Step S3, performing multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream to generate a difference sequence. Specifically comprising:

[0085] Step S3.1: The arrival time sequence TOA1, TOA2, ..., TOA n Perform multi-level difference calculation, where TOA1 represents the arrival time of the first pulse, TOA n represents the arrival time of the nth pulse, then the multi-level difference of the arrival time series is expressed as:

[0086] diff1=[TOA2-TOA1,TOA3-TOA2,…,TOA n -TOA n-1 ]

[0087] diff2=[TOA3-TOA1,TOA4-TOA2,…,TOA n -TOA n-2 ]

[0088] diff3=[TOA4-TOA1,TOA5-TOA2,…,TOA n -TOA n-3 ]

[0089]

[0090] Among them, diff1, diff2, and diff3 represent the first-level difference sequence, the second-level difference sequence, and the third-level difference sequence of the pulse arrival time, respectively.

[0091] It is easy to understand that the difference sequence levels of the first-level difference sequence, the second-level difference sequence and the third-level difference sequence in this embodiment are only examples, and this level is not a fixed value, that is, more levels of difference sequences can be set after diff3. The specific level depends on the number of radiation sources that may exist in the application scenario. The number of radiation sources is positively correlated with the set level. The level generally takes an integer between 3 and 10, which can be set according to actual conditions.

[0092] Step S3.2, the generated difference sequence D is expressed as:

[0093] D=[diff1,diff2,diff3,…].

[0094] Step S4, according to the distribution characteristics of the difference sequence, determine the neighborhood radius and density threshold of the density clustering algorithm. That is, for each data point in the difference sequence D in step S3, the relative distance between the data point and other data points is calculated by Euclidean distance to form a distance distribution matrix DIST, and then according to the distance distribution matrix DIST, the parameter neighborhood radius ε and density threshold MinPts of the DBSCAN density clustering algorithm are determined.

[0095] Step S4 specifically includes:

[0096] Step S4.1: Generate a distance distribution matrix DIST according to the difference sequence D, expressed as:

[0097] DIST=[dist(p,q)|p,q∈D]

[0098] Where p and q are any elements in the difference sequence D, and dist represents the Euclidean distance function;

[0099] Step S4.2: Rearrange the distance values ​​of each row in the distance distribution matrix DIST from small to large. The kth distance value of each row conforms to the Poisson distribution in terms of statistical characteristics. Then the expected value of the kth distance value is expressed as:

[0100]

[0101] Where λ represents the expected value, i = 1, 2, ..., n, n represents the data dimension of the difference sequence D, and k∈[1, n] is a positive integer;

[0102] Step S4.3, the average value of the 4th distance value of each row in the distance distribution matrix DIST is used as the domain radius ε; the density threshold MinPts is the average number of points in the 4 neighborhoods of all data points, that is, the density threshold MinPts is equal to 4. In this embodiment, the density threshold MinPts is determined to be 4 and the domain radius is determined to be 4 based on experience and actual data verification. When , the density clustering result is optimal.

[0103] Step S5: Based on the neighborhood radius ε and the density threshold MinPts, a density clustering algorithm is used to perform density clustering on the difference sequence to obtain a plurality of clusters after clustering.

[0104] The density clustering algorithm used in this embodiment is the DBSCAN density clustering algorithm, which uses the DBSCAN density clustering algorithm to perform density clustering on the difference sequence according to the neighborhood radius ε and the density threshold MinPts, thereby forming multiple clusters after density clustering.

[0105] The DBSCAN density clustering algorithm of the present invention specifically comprises the following steps:

[0106] Assume x is any element in the data set to be clustered. In this embodiment, the data set to be clustered is specifically a difference sequence D. Define the ε neighborhood of x as N ε (x), expressed as:

[0107] N ε (x)={y∈D:dist(y,x)<ε}

[0108] Where y represents the element in the difference sequence D that is within the ε neighborhood of x, and dist represents the Euclidean distance function.

[0109] Define the density of x as ρ(x) = |N ε (x)|, where N ε (x) is the ε neighborhood of x, and the density ρ(x) of x is the number of elements in the ε neighborhood of x. If the density ρ(x) of an element x>MinPts, then the element is called a core object, otherwise it is called a non-core object or noise.

[0110] The DBSCAN density clustering algorithm in the present invention has three conceptual definitions, namely direct density reachability, density reachability and density connection. Specifically, there are three cases:

[0111] (1) Direct density is reachable, set D c is the set of all core objects, if x∈D c , y∈N ε (x), then y is said to be directly density-reachable by x.

[0112] (2) The density can be reached. Let P1, P2, ..., P m ∈D, where m is a positive integer and 2≤m≤n, if P i+1 It is from P i Direct density reachable, i = 1, 2, ···, m-1 is the subscript, then it is called P m is reachable from P1 density.

[0113] (3) Density connected. Let x, y, z∈D. If y and z are density-reachable with x respectively, then y and z are said to be density-connected.

[0114] In the DBSCAN density clustering algorithm, from the pulse arrival time difference sequence y∈N ε (x), then y is said to be directly density-reachable by x; density-reachable, let P1,P2,...,P m ∈D, where m is a positive integer and 2≤m≤n, if P i+1 It is from P i Direct density reachable, i = 1, 2, ···, m-1 is the subscript, then it is called P m It is density reachable from P1; density connected, let x,y,z∈D, if y and z are density reachable from x respectively, then y and z are said to be density connected.

[0115] The basic steps of the DBSCAN density clustering algorithm in the present invention are: take an unmarked point from the pulse arrival time difference sequence D, if the point is a core object, then find all objects that are density-reachable with the point to form a cluster; otherwise, after marking the point as noise, continue to perform the above operation on other unmarked points until all points in the set D are traversed. If the pulse sequence to be sorted does not contain radar signals but only random chaotic pulse noise, then the difference sequence D will be marked as noise after density clustering and the sorting will end; conversely, if the pulse sequence to be sorted contains radar signals, then the difference sequence D will form several clusters after density clustering.

[0116] Step S6, calculating the extraction priority of each cluster, and generating the PRI interval information to be extracted for the cluster with the highest priority; the PRI interval information to be extracted includes the center value and jitter amplitude of the PRI to be extracted. Specifically including:

[0117] For the multiple clusters obtained after density clustering, calculate the extraction priority of each cluster separately:

[0118] Priority(i)=Cn(i) / [Cc(i)×Cj(i)]

[0119] Among them, Priority(i) represents the extraction priority of the PRI interval information of the i-th cluster, i=1,2,…,m, m represents the number of clusters, Cn(i) represents the number of points in the class corresponding to each cluster, Cc(i) represents the mean value of the points in the class corresponding to each cluster, and Cj(i) represents the length of the distribution interval of the points in the class corresponding to each cluster;

[0120] Then there is a positive integer k that satisfies:

[0121] Priority(k)≥Priority(i)

[0122] Wherein, Priority(k) represents the extraction priority of the PRI interval information of the k-th cluster.

[0123] At this time, the extraction priority corresponding to the kth cluster is the highest, so the PRI interval information of the cluster is extracted, and the format of the generated PRI interval information to be extracted is expressed as (PRI_cen, jitter), where PRI_cen=Cc(k) represents the center value, and jitter=Cj(k) / (2×PRI_cen) represents the jitter amplitude.

[0124] The present invention calculates the extraction priority of each cluster and generates the PRI interval information to be extracted for the cluster with the highest priority. This has two advantages. The first is that the pulse string with a small PRI value (i.e., PRI interval information), a large number of pulses, and a small jitter amplitude is extracted. Due to the small jitter, it is easy to ensure a high extraction accuracy, thereby ensuring the sorting success rate; the second is that a small PRI value means a large number of pulses and a high repetition frequency, so that as many pulses as possible can be extracted at a time, thereby reducing the subsequent sorting pressure and improving the overall sorting success rate. In general, by selecting the highest priority for extraction, it can be ensured that radar pulses with a large number of pulses, a small pulse interval, and a small jitter amplitude can be preferentially extracted in each round of extraction, thereby improving the sorting success rate.

[0125] Step S7: According to the PRI interval information to be extracted obtained in step S6, the radar pulse signal corresponding to the PRI interval information to be extracted is extracted using an improved pulse sequence search method. Specifically, the method comprises:

[0126] An improved pulse sequence search method is used to extract the corresponding radar pulse signal according to the PRI interval information to be extracted; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching for a pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as a tolerance value, and the tolerance value is dynamically adjusted according to different PRI interval information to be extracted to extract the corresponding radar pulse signal.

[0127] The present invention adopts an improved pulse sequence search method, that is, a pulse sequence search method with tolerance control. In the traditional pulse sequence search method, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is added as the TOA tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted. For example, when the radar signal corresponding to PRI1 is extracted by the pulse sequence search method in the first round, the tolerance used is the jitter amplitude corresponding to PRI1; after PRI1 is extracted, when the pulse sequence search method is used to extract PRI2, the jitter amplitude corresponding to PRI2 is used, and so on. The conventional pulse sequence search method uses a fixed tolerance value, which is generally 5% or 10% of the PRI value. The disadvantage of the fixed tolerance value is that if it is set too small, it is not conducive to the extraction of pulse sequences with large jitter amplitudes, and if it is set too large, it is easy to cause incorrect sorting. The present invention adopts a pulse sequence search method with tolerance control. Tolerance control refers to setting the TOA tolerance range when extracting corresponding pulses according to different jitter amplitudes, thereby achieving flexible adjustment of the tolerance value, solving the problem that setting a too small tolerance value is not conducive to the extraction of pulse sequences with large jitter amplitudes, and setting it too large easily causes incorrect sorting, thereby effectively improving the PRI valuation accuracy and sorting success rate.

[0128] Assume that the radar pulse stream to be sorted contains three radar pulse signals with pulse repetition intervals of PRI1, fixed repetition frequency (jitter amplitude <0.1%), PRI2, 1% jitter amplitude, and PRI3, 5% jitter amplitude. The three different jitter amplitudes set here are examples. In fact, the jitter amplitude can be set according to the actual situation. Different jitter amplitude settings will result in different extraction orders of subsequent PRI interval information. According to the above PRI parameters and jitter settings, according to the extraction priority in step S6, the radar pulse signal corresponding to PRI1 will be extracted first. Then, the pulse signal string is extracted using the pulse sequence search method with tolerance control in step S7. The specific steps for extracting the pulse signal string using the pulse sequence search method with tolerance control are as follows:

[0129] ①First, set the search parameter PRI to PRI1, and the corresponding tolerance range is ±0.1%;

[0130] ② Taking the arrival time of the first pulse as the starting point, determine whether there are pulses that meet the tolerance range at intervals of one times PRI1, two times PRI1, and three times PRI1. If there are pulses, it is considered that the starting part of the pulse signal string has been found, and the pulse is marked and transferred to step ④. Otherwise, proceed to step ③.

[0131] ③The starting point moves back one pulse and goes to step ②;

[0132] ④ Starting from the last marked pulse, search backward to see if there is a pulse in the interval from PRI1(1-0.1%) to PRI1(1+0.1%). If found, mark the pulse and repeat step ④. Otherwise, go to step ⑤.

[0133] ⑤ Starting from the last marked pulse, search backward to see if there is a pulse in the interval from 2PRI1 (1-0.1%) to 2PRI1 (1+0.1%) or from 3PRI1 (1-0.1%) to 3PRI1 (1+0.1%). If there is, mark the pulse and return to step ④. Otherwise, consider the pulse train to be over and go to step ⑥.

[0134] ⑥ This round of pulse search is completed, and all marked pulses are extracted from the pulse stream to be sorted.

[0135] Step S8, taking the remaining radar pulse stream as input, returns to the step of "determining whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence" to extract the remaining radar pulse signals until it is determined that there is no radar target.

[0136] In this embodiment, after extracting the PRI interval information with the highest priority, the above process is repeated for the remaining radar pulse streams until any of the following termination conditions is met: (1) in step ② of step S7, the starting part of the pulse string is still not found after traversing all the pulses to be sorted; (2) in step S6, the mean value of the in-class points corresponding to all clusters is Where i = 1, 2, ..., m, m is the number of clusters, PRI min , PRI max The minimum PRI value and the maximum PRI value are defined. These two values ​​can be flexibly set according to different application scenarios. Generally, PRI is used in the sorting of marine radar signals. min =100us, PRI max =10ms.

[0137] like Figure 2As shown in FIG. 1 , the principle of radar signal sorting of the present invention is as follows: first, the arrival time series of mixed overlapping pulses is obtained from the radar reconnaissance data to be sorted; then, it is determined whether the number of elements in the arrival time series is greater than 20 or the pulse density is greater than 0.1 pulses / ms; if the conditions are met, the next step is entered; otherwise, it is determined that there is no radar target and the sorting is terminated; then, multi-level difference calculation is performed on the arrival time series to generate a difference sequence D; then, for each data point in the difference sequence D, the relative distance between the point and other data points is calculated using the Euclidean distance to form a distance distribution matrix DIST; then, according to the distance distribution matrix DIST, the parameter neighborhood radius ε and the density threshold MinPts of the DBSCAN density clustering algorithm are determined; then, according to the determined neighborhood semicircle The difference sequence D is clustered by using the DBSCAN density clustering algorithm based on the diameter ε and density threshold MinPts to obtain multiple clusters after clustering. Then, for each cluster after clustering, the mean and distribution interval of each cluster are calculated, and then the extraction priority Priority of each cluster is calculated. The cluster with the largest Priority value is selected to generate the PRI interval information to be extracted (PRI_cen, jitter), where PRI_cen is the center value of the PRI to be extracted, which can be obtained by the average value of the difference scatter points in the corresponding clustering cluster, and jitter is the jitter amplitude of the corresponding PRI to be extracted. Then, according to the extracted PRI interval information, the pulse sequence of the corresponding radar is extracted in combination with the pulse sequence search method with tolerance control. Finally, the remaining pulse sequence is used as input, and the above steps are repeated to continue the PRI interval information estimation and pulse sequence extraction until the pulse sequence does not meet the minimum sorting requirements, then the sorting is terminated, and the sorting of the entire pulse sequence is completed.

[0138] To further illustrate the method of the present invention, the Matlab simulation platform is used to implement the method of the present invention. In the previous PRI sorting method, only the PRI value can be extracted but the PRI jitter amplitude information is lacking, which results in that the subsequent sequence extraction can only set the tolerance according to the maximum jitter amplitude in the pulse sequence, which is prone to missorting. Figure 3 The figure shows the TOA difference scatter distribution before density clustering of radar pulse signals with a PRI setting of 855us and a jitter amplitude of 2%. Figure 4 The figure shows the TOA difference scatter distribution after density clustering. According to the distribution of the points in the cluster obtained after clustering, the jitter amplitude range of the corresponding radar PRI value can be obtained. The PRI value here is the PRI interval information. Assuming that the scatter distribution in the cluster corresponding to a radar pulse train is [d1, d2], the format of the PRI value extracted by the method of the present invention is (PRI_cen, jitter), where PRI_cen is the PRI center value, and its value is equal to the mean of the scatter points in the corresponding cluster. jitter = (d2-d1) / (2×PRI_cen) is the jitter amplitude of the corresponding PRI, which is generally expressed in percentage, such as Figure 4 The extraction result of the algorithm of the present invention is (855.2, 2.1%), which is basically consistent with the radar PRI parameter setting value.

[0139] In order to verify the superiority of the method of the present invention in PRI valuation and sorting success rate, simulation comparison experiments were carried out on the method of the present invention, the typical difference histogram method (CDIF method) and the PRI transformation method in different degrees of PRI jitter, and a comparative analysis was performed from two aspects: PRI value extraction accuracy and sorting success rate.

[0140] It is assumed that there is only one radar, the pulse repetition period is set to 175.4μs, the PRI jitter is set to 1%, 2%, 5%, and 10% in sequence, and 200% of the total number of pulses of stray noise pulses are added to simulate a complex electromagnetic environment. The simulation time is set to 20ms. The method of the present invention is compared and analyzed with the CDIF method and PRI transformation method commonly used in PRI sorting. Each group of experiments is simulated 100 times with Monte Carlo simulations, and the results are shown in Table 1.

[0141] Table 1 Comparison of PRI value extraction results

[0142]

[0143]

[0144] The PRI estimation error rate and sorting success rate of each PRI sorting method under different jitter amplitudes are as follows: Figure 5 and Figure 6 As shown. It can be seen that as the PRI jitter amplitude continues to increase, the PRI estimation error rates of the three methods continue to rise, and the sorting success rate continues to decline. However, the method of the present invention performs better than the other two methods in both estimation error rate and sorting success rate. In particular, when the jitter amplitude is large, the performance of the CDIF method and the PRI transformation method drops sharply, but the method of the present invention can still maintain a low estimation error rate and a high sorting success rate, and has a strong ability to cope with PRI jitter.

[0145] In actual battlefield environments, multiple radar signals often overlap with each other, and the pulse width, carrier frequency, arrival angle and other parameters overlap seriously. At this time, PRI-based sorting is particularly important. In order to verify the multi-target sorting capability of the method of the present invention in the above situation, 4 groups of radar pulse signals are set: the first group PRI value is 250μs, jitter 1%; the second group PRI value is 333μs, jitter 5%; the third group PRI value is 855μs, jitter 5%; the fourth group PRI value is 583μs, jitter 2%. Set the simulation time to 20ms, and add 20% of the total number of pulses.

[0146] The method of the present invention is used to extract the PRI value of the simulated mixed overlapping pulse stream, and then combined with the sequence search method to perform refined sorting based on PRI. Figure 7 The figure shows the result of the refined extraction of the PRI value contained in the pulse stream by the method of the present invention. It can be seen from the figure that the extraction result is basically consistent with the set PRI value and PRI jitter range. Among them, the pulse signal of radar 1 is extracted first, because its PRI value and jitter amplitude are relatively small, that is, it is the most dense in the TOA difference scatter point distribution; followed by radar 2 and radar 4, whose TOA difference scatter point distribution density is relatively similar; the pulse signal of radar 3 is extracted last, because its PRI value and jitter amplitude are relatively large, so the TOA difference point distribution is relatively scattered. Only when other radar pulses are extracted, the high-density area where radar 3 is located can be revealed.

[0147] The above radar pulses were sorted based on PRI using the method of the present invention, CDIF method and PRI transformation method. Each group of experiments was simulated 100 times by Monte Carlo. The experimental results were compared. Figure 8 As shown. Comparative analysis shows that for the same radar pulse signal, since the PRI value extracted by the method of the present invention has a relatively high accuracy and contains PRI jitter amplitude information, the pulse sorting success rate is higher than that of the other two methods. For different radar pulse signals, the method of the present invention can set the corresponding pulse extraction TOA tolerance range according to the different radar PRI jitter amplitudes, give priority to extracting radar pulses with a large number of pulses and a small jitter amplitude, give priority to ensuring a higher extraction success rate, and reduce the difficulty of subsequent pulse sorting. The other two methods can only be set according to the maximum tolerance, so it is difficult to improve the sorting success rate, and it will also increase interference such as pulse loss and messy pulses in subsequent sorting.

[0148] The radar signal sorting method based on PRI interval information of the invention is adopted. It generates a time difference sequence of arrival time by making multi-level differences in the arrival time TOA of the mixed overlapping pulses, and uses a density clustering algorithm to extract PRI interval information, including the PRI center value and the PRI jitter amplitude, and then combines the pulse sequence search method with tolerance control to complete the sorting of the mixed overlapping pulse stream. Compared with traditional sorting methods such as the CDIF method and the PRI transformation method, the sorting method proposed by the present invention can improve the PRI estimation accuracy and enrich the PRI information dimension. For different radar pulse signals, the present invention can set the corresponding pulse extraction TOA tolerance range according to the different PRI jitter amplitudes of different radars, give priority to extracting radar pulses with a large number of pulses and a small jitter amplitude, give priority to ensuring a high extraction success rate, and at the same time reduce the difficulty of subsequent pulse sorting.

[0149] Moreover, compared with traditional sorting methods such as the CDIF method and the PRI transformation method, which can only perform signal sorting according to the maximum tolerance, the sorting method proposed in the present invention can improve the PRI estimation accuracy and enrich the PRI information dimension, thereby effectively improving the sorting success rate and having a strong ability to cope with PRI jitter.

[0150] Example 2

[0151] like Fig. 9 As shown, this embodiment provides a radar signal sorting system based on PRI interval information, and the system specifically includes:

[0152] The arrival time series acquisition module M1 is used to derive the arrival time series of mixed overlapping pulses from the radar reconnaissance data to be sorted;

[0153] A radar target judgment module M2 is used to judge whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence;

[0154] When the number of elements in the arrival time series is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and a difference sequence generation module M3 is used to perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence;

[0155] When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated;

[0156] The difference sequence generation module M3 is used to perform multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream to generate a difference sequence;

[0157] The density clustering parameter determination module M4 is used to determine the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence;

[0158] Density clustering processing module M5, used to perform density clustering on the difference sequence according to the neighborhood radius and density threshold using a density clustering algorithm to obtain a plurality of clusters after clustering;

[0159] The extraction priority calculation and PRI information generation module M6 is used to calculate the extraction priority of each cluster and generate the PRI interval information to be extracted for the cluster with the highest priority;

[0160] The tolerance control and signal extraction module M7 is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted by adopting an improved pulse sequence search method; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching for a pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as a tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted, so as to extract the corresponding radar pulse signal;

[0161] The radar pulse signal repeated extraction module M8 is used to take the remaining radar pulse stream as input and return to the step of "determining whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence" to extract the remaining radar pulse signals until it is determined that there is no radar target.

[0162] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0163] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily appreciated by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.

Claims

1. A radar signal sorting method based on PRI interval information, characterized in that: include: Perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence; Determining the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence; Determining the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence specifically includes: The distance distribution matrix DIST is generated according to the difference sequence D, which is expressed as: DIST=[dist(p,q)|p,q∈D] Where p and q are any elements in the difference sequence D, and dist represents the Euclidean distance function; Rearrange the distance values ​​of each row in the distance distribution matrix DIST from small to large. The kth distance value of each row conforms to the Poisson distribution in terms of statistical characteristics. Then the expected value of the kth distance value is expressed as: Where λ represents the expected value, i = 1, 2, ..., n, n represents the data dimension of the difference sequence D, and k∈[1, n] is a positive integer; The average value of the 4th distance value of each row in the distance distribution matrix DIST is taken as the domain radius ε; the density threshold MinPts is the average number of points in the 4 neighborhoods of all data points, that is, the density threshold MinPts is equal to 4; According to the neighborhood radius and the density threshold, the difference sequence is density clustered using a density clustering algorithm to obtain a plurality of clusters after clustering; Calculate the extraction priority of each cluster and generate the PRI interval information to be extracted for the cluster with the highest priority; The step of calculating the extraction priority of each cluster and generating the PRI interval information to be extracted for the cluster with the highest priority specifically includes: For the multiple clusters obtained after density clustering, calculate the extraction priority of each cluster separately: Priority(i)=Cn(i) / [Cc(i)×Cj(i)] Among them, Priority(i) represents the extraction priority of the PRI interval information of the i-th cluster, i=1,2,…,m, m represents the number of clusters, Cn(i) represents the number of points in the class corresponding to each cluster, Cc(i) represents the mean value of the points in the class corresponding to each cluster, and Cj(i) represents the length of the distribution interval of the points in the class corresponding to each cluster; Then there is a positive integer k that satisfies: Priority(k)≥Priority(i) Among them, Priority(k) represents the extraction priority of the PRI interval information of the kth cluster; At this time, the extraction priority corresponding to the kth cluster is the highest, so the PRI interval information of the cluster is extracted, and the format of the generated PRI interval information to be extracted is expressed as (PRI_cen, jitter); Wherein, PRI_cen=Cc(k) represents the center value, and jitter=Cj(k) / (2×PRI_cen) represents the jitter amplitude; An improved pulse sequence search method is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching the pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as the tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted to extract the corresponding radar pulse signal.

2. The radar signal sorting method based on PRI interval information according to claim 1 is characterized in that: Before the step of "performing multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream to generate a difference sequence", the method further includes: deriving the arrival time series of mixed overlapping pulses from the radar reconnaissance data to be sorted; judging whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence; When the number of elements in the arrival time series is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and the step of "performing multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence" is executed; When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated.

3. The radar signal sorting method based on PRI interval information according to claim 2 is characterized in that: After the step of "using an improved pulse sequence search method to extract the radar pulse signal corresponding to the PRI interval information to be extracted", the method further includes: With the remaining radar pulse stream as input, return to the step of "determining whether there is a radar target in the arrival time series according to the number of elements and the pulse density of the arrival time series" to extract the remaining radar pulse signals until it is determined that there is no radar target.

4. The radar signal sorting method based on PRI interval information according to claim 1 is characterized in that: The step of performing multi-level difference calculation on the arrival time sequence of the mixed overlapping radar pulse stream to generate a difference sequence specifically includes: The arrival time series TOA1, TOA2, ..., TOA of the mixed overlapping radar pulse stream n Perform multi-level difference calculation, where TOA1 represents the arrival time of the first pulse, TOA n represents the arrival time of the nth pulse, then the multi-level difference of the arrival time series is expressed as: diff1=[TOA2-TOA1,TOA3-TOA2,…,TOA n -TOA n-1 ] diff2=[TOA3-TOA1,TOA4-TOA2,…,TOA n -TOA n-2 ] diff3=[TOA4-TOA1,TOA5-TOA2,…,TOA n -TOA n-3 ] Among them, diff1, diff2, and diff3 represent the first-level difference sequence, the second-level difference sequence, and the third-level difference sequence of the pulse arrival time, respectively. The generated difference sequence D is expressed as: D=[diff1,diff2,diff3,…].

5. The radar signal sorting method based on PRI interval information according to claim 1 is characterized in that: The step of performing density clustering on the difference sequence according to the neighborhood radius and the density threshold using a density clustering algorithm to obtain a plurality of clusters after clustering specifically includes: The DBSCAN density clustering algorithm is used to perform density clustering on the difference sequence according to the neighborhood radius and density threshold, and multiple clusters are formed after density clustering.

6. A radar signal sorting system based on PRI interval information, characterized in that: include: A difference sequence generation module is used to perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence; The density clustering parameter determination module is used to determine the neighborhood radius and density threshold of the density clustering algorithm according to the distribution characteristics of the difference sequence, and specifically includes: The distance distribution matrix DIST is generated according to the difference sequence D, which is expressed as: DIST=[dist(p,q)|p,q∈D] Where p and q are any elements in the difference sequence D, and dist represents the Euclidean distance function; Rearrange the distance values ​​of each row in the distance distribution matrix DIST from small to large. The kth distance value of each row conforms to the Poisson distribution in terms of statistical characteristics. Then the expected value of the kth distance value is expressed as: Where λ represents the expected value, i = 1, 2, ..., n, n represents the data dimension of the difference sequence D, and k∈[1, n] is a positive integer; The average value of the 4th distance value of each row in the distance distribution matrix DIST is taken as the domain radius ε; the density threshold MinPts is the average number of points in the 4 neighborhoods of all data points, that is, the density threshold MinPts is equal to 4; A density clustering processing module, used to perform density clustering on the difference sequence according to the neighborhood radius and the density threshold using a density clustering algorithm to obtain a plurality of clusters after clustering; The extraction priority calculation and PRI information generation module is used to calculate the extraction priority of each cluster and generate the PRI interval information to be extracted for the cluster with the highest priority, specifically including: For the multiple clusters obtained after density clustering, calculate the extraction priority of each cluster separately: Priority(i)=Cn(i) / [Cc(i)×Cj(i)] Among them, Priority(i) represents the extraction priority of the PRI interval information of the i-th cluster, i=1,2,…,m, m represents the number of clusters, Cn(i) represents the number of points in the class corresponding to each cluster, Cc(i) represents the mean value of the points in the class corresponding to each cluster, and Cj(i) represents the length of the distribution interval of the points in the class corresponding to each cluster; Then there is a positive integer k that satisfies: Priority(k)≥Priority(i) Among them, Priority(k) represents the extraction priority of the PRI interval information of the kth cluster; At this time, the extraction priority corresponding to the kth cluster is the highest, so the PRI interval information of the cluster is extracted, and the format of the generated PRI interval information to be extracted is expressed as (PRI_cen, jitter); Wherein, PRI_cen=Cc(k) represents the center value, and jitter=Cj(k) / (2×PRI_cen) represents the jitter amplitude; The tolerance control and signal extraction module is used to extract the radar pulse signal corresponding to the PRI interval information to be extracted by adopting an improved pulse sequence search method; the improved pulse sequence search method refers to a pulse sequence search method with tolerance control. When searching the pulse sequence, the jitter amplitude corresponding to the PRI interval information to be extracted in each pulse sequence search is used as the tolerance value, and the tolerance value is dynamically adjusted according to the different PRI interval information to be extracted to extract the corresponding radar pulse signal.

7. The radar signal sorting system based on PRI interval information according to claim 6, characterized in that: Also includes: An arrival time series acquisition module is used to derive the arrival time series of mixed overlapping pulses from the radar reconnaissance data to be sorted; A radar target judgment module, used for judging whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence; When the number of elements in the arrival time series is greater than a first preset threshold or the pulse density is greater than a second preset threshold, it is determined that there is a radar target, and the difference sequence generation module is used to perform multi-level difference calculation on the arrival time series of the mixed overlapping radar pulse stream to generate a difference sequence; When the number of elements of the arrival time series is less than or equal to the first preset threshold or the pulse density is less than or equal to the second preset threshold, it is determined that there is no radar target and the radar signal sorting is terminated; The radar pulse signal repeated extraction module is used to take the remaining radar pulse stream as input and return to the step of "determining whether there is a radar target in the arrival time sequence according to the number of elements and pulse density of the arrival time sequence" to extract the remaining radar pulse signals until it is determined that there is no radar target.

Images