A radar signal sorting method and system based on cyclic stationary analysis

Abstract

The application provides a radar signal sorting method and system based on cyclic stationary analysis, superimposed pulses of different radar transmitting sources are collected to obtain a multi-source mixed pulse signal sequence; based on the physical acoustic propagation principle, a compensated pulse signal sequence that suppresses the distortion of complex medium propagation is generated; according to a preset pulse repetition interval feature extraction rule, a feature data set containing pulse repetition interval features is obtained; the compensated pulse signal sequence is subjected to cyclic stationary analysis to obtain a cyclic stationary feature set containing periodic component information; according to a preset radar signal sorting rule, the feature data set and the cyclic stationary feature set are subjected to fusion sorting processing to generate a sorting result of the multi-source mixed radar pulse signal sequence; the technical scheme provided by the application realizes efficient sorting and feature extraction of the multi-source mixed radar pulse signal, and improves the accuracy and stability of radar signal processing in a complex medium environment.

Description

Technical Field

[0001] This application relates to the field of radar signal processing technology, and in particular to a radar signal sorting method and system based on cyclostationary analysis. Background Technology

[0002] In urban environments, with the rapid development of intelligent transportation systems and environmental monitoring networks, radar equipment is widely deployed in complex urban spaces. These radar systems generate a large number of dense and mutually interfering signals during operation. Especially under conditions of multiple sources coexisting and complex propagation paths, accurately separating the pulse sequences of each radar source from the aliased signals has become a key technical challenge to ensure system reliability and data integrity. Therefore, a technical solution capable of achieving efficient and accurate signal sorting under strong interference is urgently needed.

[0003] Current research has proposed a radar signal sorting method based on a combination of adaptive filtering and statistical feature extraction. This method performs time-frequency analysis on the received signal, combines short-time Fourier transform to obtain the signal's local frequency characteristics, and suppresses background noise. Then, it uses statistical methods such as principal component analysis to extract key feature parameters of the signal, and finally uses a clustering algorithm to classify and identify different radar sources. However, existing solutions have certain limitations. For example, because they do not consider the influence of the physical medium on the signal propagation path and intensity, the signal feature extraction stage is easily affected by amplitude attenuation and phase distortion caused by non-uniform media, thus affecting the final classification accuracy. Furthermore, they struggle to capture the periodic or quasi-periodic structural information in radar signals, leading to misjudgments or missed detections when facing highly overlapping pulse sequences, thus limiting overall sorting performance. Summary of the Invention

[0004] This application provides a radar signal sorting method and system based on cyclic stationary analysis to solve the problems in the prior art, such as insufficient classification accuracy due to easy interference in the signal feature extraction stage, misjudgment or missed detection when facing highly overlapping pulse sequences, and limited overall sorting performance.

[0005] In a first aspect, this application provides a radar signal sorting method based on cyclic stationary analysis, including:

[0006] By collecting superimposed pulses from different radar sources, a multi-source aliased pulse signal sequence is obtained;

[0007] Based on the principle of physical acoustic propagation, the multi-source aliasing pulse signal sequence is subjected to medium attenuation compensation processing to generate a compensated pulse signal sequence that suppresses propagation distortion in complex media.

[0008] According to the preset pulse repetition interval feature extraction rules, feature extraction is performed on the compensated pulse signal sequence to obtain a feature data set containing pulse repetition interval features;

[0009] Cyclic stationarity analysis is performed on the compensated pulse signal sequence to obtain a cyclic stationarity feature set containing periodic component information;

[0010] According to the preset radar signal sorting rules, the feature data set and the cyclic stationary feature set are fused and sorted to generate the sorting result of the multi-source aliased radar pulse signal sequence.

[0011] Optionally, based on the principle of physical acoustic propagation, the multi-source aliasing pulse signal sequence is subjected to medium attenuation compensation processing to generate a compensated pulse signal sequence that suppresses propagation distortion in complex media, including:

[0012] Based on the physical acoustic propagation principle, the propagation path of radar signals in complex media is analyzed to obtain the set of path node coordinates.

[0013] Identify the medium attribute data of each path node in the set of path node coordinates, so as to separate the medium density parameter set and the medium propagation speed parameter set from the medium attribute data;

[0014] Based on the set of medium density parameters and the set of propagation velocity parameters, the path scattering loss and the path phase delay are calculated respectively, and the path scattering loss and the path phase delay are combined to obtain a set of attenuation compensation coefficients.

[0015] Identify the propagation path type of each pulse signal in the multi-source aliasing pulse signal sequence to obtain an intermediate pulse signal sequence with propagation path identifiers;

[0016] The intermediate pulse signal sequence is corrected according to the set of attenuation compensation coefficients to obtain the corrected pulse signal sequence;

[0017] The corrected pulse signal sequence is recombined to obtain a compensated pulse signal sequence that suppresses propagation distortion in complex media.

[0018] Optionally, according to a preset pulse repetition interval feature extraction rule, feature extraction is performed on the compensated pulse signal sequence to obtain a feature data set containing pulse repetition interval features, including:

[0019] Based on the time marking conditions of the preset pulse repetition interval feature extraction rules, the pulse signals in the compensated pulse signal sequence are marked with positions to obtain a pulse time position set.

[0020] Calculate the time interval between adjacent pulse time positions in the pulse time position set to generate the original pulse repetition interval set;

[0021] Based on a preset pulse repetition interval range, effective pulse repetition intervals are extracted from the original pulse repetition interval set to obtain an effective pulse repetition interval set.

[0022] Based on preset statistical distribution characteristics, the effective pulse repetition interval set is grouped to generate a pulse signal group set containing pulse repetition interval characteristics;

[0023] The pulse signal group set is matched with the preset radar source identifier to obtain the feature data set.

[0024] Optionally, cyclostationary analysis is performed on the compensated pulse signal sequence to obtain a cyclostationary feature set containing periodic component information, including:

[0025] The compensated pulse signal sequence is segmented to obtain multiple pulse signal segments;

[0026] Perform time-frequency conversion on each pulse signal segment to generate a time-frequency distribution set;

[0027] Based on a preset periodic stability condition, the time-frequency regions in the time-frequency distribution set where the energy intensity fluctuates periodically are detected to determine candidate time-frequency regions that meet the preset periodic fluctuation amplitude change threshold.

[0028] Measure the time interval between adjacent energy peaks in each candidate time-frequency region to generate a set of regional period values;

[0029] Calculate the difference between the highest and lowest frequencies of each candidate time-frequency region in the candidate time-frequency region set to obtain the region frequency span set;

[0030] The set of regional frequency spans and the set of regional period values ​​are fused to generate a set of period component parameters.

[0031] Based on the preset radar source periodic characteristics, the set of periodic component parameters is classified by source, resulting in a cyclic stationary feature set containing periodic component information.

[0032] Optionally, based on a preset periodic stability condition, time-frequency regions in the time-frequency distribution set where the energy intensity exhibits periodic fluctuations are detected to determine candidate time-frequency regions that satisfy a preset threshold for the amplitude change of periodic fluctuations, including:

[0033] Based on a preset periodic stability condition, the time-frequency regions in the time-frequency distribution set where the energy intensity fluctuates periodically are detected to determine multiple peak positions;

[0034] Calculate the time interval variance between the peak positions. When the time interval variance is less than a preset peak interval stability threshold, the peak position is selected as a candidate peak position.

[0035] Based on a preset time overlap threshold and a preset peak interval similarity threshold, time feature matching is performed on the candidate peak positions to select the target peak positions that simultaneously meet the time overlap requirement and the interval similarity requirement.

[0036] The energy amplitude variation of the time-frequency region where the target peak position is located is verified, and the time-frequency region where the amplitude difference between adjacent peaks is less than the preset periodic fluctuation amplitude variation threshold is selected as the candidate time-frequency region.

[0037] Optionally, the time interval between adjacent energy peaks in each candidate time-frequency region is measured to generate a set of regional periodic values, including:

[0038] Arrange the target peak positions corresponding to each candidate time-frequency region in ascending order to obtain the corresponding sorted target peak position sequence.

[0039] Calculate the time interval between adjacent target peak positions in each sorted target peak position sequence to generate multiple original time interval sequences;

[0040] Calculate the absolute difference between adjacent time intervals in each original time interval sequence, and select a stable time interval sequence from the corresponding original time interval sequence whose absolute difference does not exceed the preset interval tolerance;

[0041] The average value of each stable time interval sequence is used as the regional period value of the corresponding candidate time-frequency region, and the regional period values ​​of all candidate time-frequency regions are aggregated to obtain the set of regional period values.

[0042] Optionally, according to preset radar signal sorting rules, the feature data set and the cyclically stationary feature set are fused and sorted to generate a sorting result for the multi-source aliased radar pulse signal sequence, including:

[0043] Based on the preset radar signal sorting rules, the pulse repetition interval feature in the feature data set and the periodic component information in the cyclic stationary feature set are matched to generate a fused feature group set.

[0044] Based on the preset radar signal sorting rules, the auxiliary matching rules match the period parameters of each fusion feature group in the fusion feature group set with the preset radar source feature template to generate a fusion feature group set with radar source identifier.

[0045] The pulse signal corresponding to the time feature parameter of the fused feature group set is identified from the compensated pulse signal sequence, and the feature group pulse set is generated.

[0046] Based on the acquisition time sequence of the multi-source aliased radar pulse signal sequence, the pulse signals with the same radar source identifier in the feature group pulse set are recombined to obtain the sorting result of the multi-source aliased radar pulse signal sequence.

[0047] Secondly, this application provides a radar signal sorting system based on cyclic stationary analysis, comprising:

[0048] The acquisition module is used to acquire superimposed pulses from different radar sources to obtain a multi-source aliased pulse signal sequence;

[0049] The compensation module is used to perform medium attenuation compensation processing on the multi-source aliasing pulse signal sequence based on the physical acoustic propagation principle, so as to generate a compensated pulse signal sequence that suppresses the propagation distortion of complex media.

[0050] The extraction module is used to extract features from the compensated pulse signal sequence according to a preset pulse repetition interval feature extraction rule, so as to obtain a feature data set containing pulse repetition interval features.

[0051] The analysis module is used to perform cyclostationary analysis on the compensated pulse signal sequence to obtain a cyclostationary feature set containing periodic component information.

[0052] The sorting module is used to perform fusion sorting processing on the feature data set and the cyclic stationary feature set according to the preset radar signal sorting rules, and generate the sorting result of the multi-source aliased radar pulse signal sequence.

[0053] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement a radar signal sorting method based on cyclic stationary analysis as described in the first aspect above.

[0054] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a radar signal sorting method based on cyclic stationary analysis as described in the first aspect.

[0055] In this application, superimposed pulses from different radar sources are acquired to obtain a multi-source aliased pulse signal sequence. Based on the principle of physical acoustic propagation, the multi-source aliased pulse signal sequence is subjected to medium attenuation compensation processing to generate a compensated pulse signal sequence that suppresses propagation distortion in complex media. According to a preset pulse repetition interval feature extraction rule, features are extracted from the compensated pulse signal sequence to obtain a feature data set containing pulse repetition interval features. Cyclic stationarity analysis is performed on the compensated pulse signal sequence to obtain a cyclic stationary feature set containing periodic component information. According to a preset radar signal sorting rule, the feature data set and the cyclic stationary feature set are fused and sorted to generate a sorting result for the multi-source aliased radar pulse signal sequence. The technical solution provided in this application captures mixed signals from multiple radar sources, providing a raw data foundation for subsequent signal processing and avoiding information loss due to incomplete sampling. Based on the principle of physical acoustic propagation, it simulates and compensates for signal amplitude attenuation and phase distortion caused by non-uniform media such as buildings and terrain. Its technical effect is to reduce the impact of environmental factors on signal integrity and improve the accuracy of subsequent feature extraction. Features are extracted from the compensated signal to serve as a crucial basis for identifying radar emission sources, providing structured data support for subsequent fusion analysis. Cyclic stationary analysis is used to uncover hidden periodic components in radar signals, enhancing the distinguishability between radar signals and improving identification capabilities in highly overlapping signals. A comprehensive judgment and classification based on multiple feature information is then performed to achieve automatic sorting of radar signals. This application addresses the feature distortion problem caused by existing methods failing to consider the influence of complex media on signal propagation by introducing the principle of physical acoustic propagation, thereby improving signal fidelity along the propagation path. Simultaneously, through cyclostationary analysis technology, the potential periodic structural information in radar signals is deeply mined, overcoming the limitations of traditional methods based on static statistical features in handling highly overlapping signals. This enables the system to possess stronger anti-interference capabilities and higher sorting accuracy in urban radar signal scenarios with multiple sources and severe interference, thus breaking through the performance bottleneck of existing technologies in practical applications.

[0056] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1A flowchart of a radar signal sorting method based on cyclic stationary analysis provided in this application is shown;

[0059] Figure 2 A schematic diagram of a radar signal sorting system based on cyclic stationary analysis provided in this application is shown.

[0060] Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0061] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0062] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] In urban environments, with the widespread deployment of radar equipment, the mutual interference of multi-source radar signals under complex propagation paths is becoming increasingly serious. Accurately separating the pulse sequences of each radar source from highly aliased signals has become a key technical challenge. While existing technologies employ adaptive filtering combined with time-frequency analysis and statistical feature extraction, they fail to consider the influence of the physical medium on signal propagation. This leads to susceptibility to amplitude attenuation and phase distortion caused by non-uniform media during feature extraction. Furthermore, the lack of characterization of the periodic structure in radar signals makes it difficult to address misjudgments and missed detections caused by highly overlapping signals. To address these shortcomings, this application proposes a radar signal sorting method based on cyclostationary analysis. By introducing the principle of physical acoustic propagation, it compensates for signal attenuation through the medium, improving signal fidelity in complex environments. Simultaneously, by combining cyclostationary analysis technology, it deeply mines the periodic features in radar signals, enhancing the distinguishability between different radar sources. Finally, by fusing pulse repetition interval features and cyclostationary features, it achieves high-precision and robust signal sorting, thus overcoming the technical bottleneck of limited performance of existing technologies in complex urban electromagnetic environments.

[0065] Figure 1 A flowchart of a radar signal sorting method based on cyclic stationary analysis is provided in this application embodiment, as follows: Figure 1 As shown, the method includes:

[0066] Step 101: Collect superimposed pulses from different radar sources to obtain a multi-source aliased pulse signal sequence.

[0067] In this step, superimposed pulses refer to composite waveforms generated by the time-domain overlap of pulse signals emitted by multiple radar sources, including a mixture of pulses with different frequencies, amplitudes, and arrival times. Multi-source aliased pulse signal sequences refer to sequences composed of superimposed pulses arranged in chronological order, containing signals from multiple interfering radar sources.

[0068] In this embodiment of the application, pulse signals from multiple radar sources that overlap in the time domain are acquired to obtain a superimposed pulse sequence containing signals from different radar sources; the arrival time, amplitude, and frequency parameters of each pulse are recorded using signal synchronous acquisition technology to obtain a multi-source aliased pulse signal sequence.

[0069] Step 102: Based on the principle of physical acoustic propagation, perform medium attenuation compensation processing on the multi-source aliasing pulse signal sequence to generate a compensated pulse signal sequence that suppresses the propagation distortion of complex media.

[0070] In this step, the physical acoustic propagation principle refers to the attenuation model of sound waves propagating in a medium, including the quantized relationship between path scattering loss and path phase delay. The compensated pulse signal sequence refers to the pulse sequence after eliminating medium distortion through inverse gain adjustment, containing corrected amplitude and phase information.

[0071] In this embodiment, based on the principle of physical acoustic propagation, the path scattering loss and path phase delay of the signal in the multi-medium path are calculated, each pulse signal is corrected to eliminate pulse signal distortion, and a compensated pulse signal sequence that suppresses the propagation distortion of complex media is generated.

[0072] In this embodiment, based on the principle of physical acoustic propagation, the scattering loss and phase delay of the signal in multiple media paths such as buildings and air layers are calculated, and the amplitude of each pulse is reversed to eliminate signal distortion caused by the urban environment, thereby generating a compensated pulse signal sequence after amplitude and phase correction.

[0073] Step 103: Based on the preset pulse repetition interval feature extraction rules, perform feature extraction on the compensated pulse signal sequence to obtain a feature data set containing pulse repetition interval features.

[0074] In this step, the preset pulse repetition interval feature extraction rule refers to the set effective range and statistical distribution threshold of the pulse interval. The feature data set refers to the pulse repetition interval dataset grouped by radar source, which includes the interval value and its statistical distribution attributes.

[0075] In this embodiment, based on a preset pulse repetition interval feature extraction rule, the compensated pulse signal sequence is subjected to pulse time marking, adjacent pulse interval calculation and interval clustering processing to extract the effective pulse repetition interval characterizing the radar source working mode and generate a feature data set containing pulse repetition interval features.

[0076] Step 104: Perform cyclostationary analysis on the compensated pulse signal sequence to obtain a cyclostationary feature set containing periodic component information.

[0077] In this step, the cyclic stationary feature set refers to the parameter set extracted through time-frequency conversion analysis.

[0078] In this embodiment, the compensated pulse signal sequence is decomposed into a time-frequency distribution set by time-frequency conversion technology, the time-frequency region with periodic fluctuations in energy intensity is detected, the period value and frequency span of the region are measured, the source is classified after parameter fusion, and a cyclic stationary feature set of periodic component information is generated.

[0079] Step 105: According to the preset radar signal sorting rules, the feature data set and the cyclic stationary feature set are fused and sorted to generate the sorting result of the multi-source aliased radar pulse signal sequence.

[0080] In this step, the preset radar signal sorting rules refer to the set feature matching and pulse backtracking logic. The sorting result refers to the time-series pulse sequences classified according to radar source.

[0081] In this embodiment of the application, the pulse repetition interval feature and the periodic component information are associated and fused according to the preset radar source feature matching rules, the fused feature set is bound to the radar source identifier, and the sorting result of the multi-source aliased radar pulse signal sequence is generated.

[0082] This application's embodiments solve the problem of source identifier loss caused by multipath distortion and signal aliasing by integrating physical acoustic compensation, cyclic stationarity analysis, and dual-feature sorting; improve signal integrity by utilizing medium attenuation compensation processing, and enhance feature discrimination by combining time-frequency period detection, ultimately achieving high-precision radar signal sorting and improving sorting robustness in complex scenarios.

[0083] This application provides a specific embodiment. Step 102 involves performing medium attenuation compensation processing on the multi-source aliasing pulse signal sequence based on the principle of physical acoustic propagation to generate a compensated pulse signal sequence that suppresses propagation distortion in complex media. This specifically includes the following steps:

[0084] Step 201: Based on the principle of physical acoustic propagation, analyze the propagation path of radar signals in complex media to obtain the set of path node coordinates.

[0085] In this step, the propagation path refers to the physical trajectory of the radar signal propagating in a complex medium. The set of path node coordinates refers to the sequence of three-dimensional coordinate points formed by discretizing the propagation path.

[0086] In this embodiment, based on the principle of physical acoustic propagation, the propagation trajectory of radar signals in a group of buildings is simulated by ray tracing technology. The path is discretized into line segment nodes and their three-dimensional coordinates are recorded to obtain a set of path node coordinates.

[0087] Step 202: Identify the medium attribute data of each path node in the set of path node coordinates, so as to separate the medium density parameter set and the medium propagation speed parameter set from the medium attribute data.

[0088] In this step, the medium property data refers to the dataset of physical characteristics of the medium at each path node. The medium density parameter set refers to the set of medium density values ​​for each path node; density equals mass divided by volume. The medium propagation velocity parameter set refers to the set of sound wave propagation velocity values ​​for each path node.

[0089] In this embodiment of the application, the material properties at the coordinates of each path node in the path node coordinate set are queried, and the medium density parameter set and the medium propagation speed parameter set are extracted through the attribute classification algorithm.

[0090] Step 203: Based on the medium density parameter set and the propagation velocity parameter set, calculate the path scattering loss and the path phase delay respectively, and combine the path scattering loss and the path phase delay to obtain the attenuation compensation coefficient set.

[0091] In this step, the set of attenuation compensation coefficients refers to the set of compensation values ​​for fused scattering loss and phase delay.

[0092] In this embodiment, path phase delay = path length ÷ speed. The path scattering loss value and path phase delay value are weighted and added together by node to obtain the attenuation compensation coefficient. The specific calculation formula is: attenuation compensation coefficient = path scattering loss value + path phase delay value × 2π × frequency parameter, where the frequency parameter is obtained by signal synchronous acquisition technology. Each path node is assigned a corresponding attenuation compensation coefficient to obtain the attenuation compensation coefficient set.

[0093] Step 204: Identify the propagation path type of each pulse signal in the multi-source aliasing pulse signal sequence to obtain an intermediate pulse signal sequence with propagation path identifiers.

[0094] In this step, the propagation path type refers to the classification of the propagation mode of the pulse signal. The intermediate pulse signal sequence refers to the pulse sequence with the path type identifier added.

[0095] In this embodiment of the application, by matching the pulse arrival time difference and phase characteristics, three types of propagation paths, namely direct, reflected and diffracted, are distinguished. A propagation path type identifier is added to each pulse signal in the multi-source aliased pulse signal sequence to generate an intermediate pulse signal sequence with propagation path identifier.

[0096] Step 205: Correct the intermediate pulse signal sequence according to the set of attenuation compensation coefficients to obtain the corrected pulse signal sequence.

[0097] In this step, the corrected pulse signal sequence refers to the pulse sequence that has undergone amplitude compensation.

[0098] In this embodiment of the application, according to the set of attenuation compensation coefficients, the corresponding attenuation compensation coefficients are called to perform a correction operation on the amplitude of each pulse signal in the intermediate pulse signal sequence, and a corrected pulse signal sequence after amplitude correction is generated, wherein the corrected amplitude = original amplitude × 1 / attenuation compensation coefficient.

[0099] Step 206: Reassemble the corrected pulse signal sequence to obtain a compensated pulse signal sequence that suppresses propagation distortion in complex media.

[0100] In the embodiments of this application, all corrected pulse signal sequences are integrated in the original time sequence to reconstruct the signal waveform and generate a compensated pulse signal sequence that suppresses the propagation distortion of complex media.

[0101] The embodiments of this application accurately quantify the medium attenuation effect through physical acoustic modeling, and achieve pulse-level directional compensation by combining path type recognition, thereby suppressing signal distortion and energy attenuation and providing a high-fidelity signal foundation for subsequent feature extraction.

[0102] This application provides a specific embodiment. Step 103 involves extracting features from the compensated pulse signal sequence according to a preset pulse repetition interval feature extraction rule to obtain a feature data set containing pulse repetition interval features. Specifically, this includes the following steps:

[0103] Step 301: According to the time marking conditions of the preset pulse repetition interval feature extraction rules, the pulse signals in the compensated pulse signal sequence are marked with positions to obtain a pulse time position set.

[0104] In this step, the time stamp condition refers to the amplitude trigger threshold for pulse front detection. The pulse time position set refers to the ordered set of all pulse start timestamps in the compensated pulse signal sequence.

[0105] In this embodiment, the pulse leading edge detection threshold in the preset pulse repetition interval feature is used as the time stamp condition. The pulse start time point is determined by comparing the intersection of the rising edge of the pulse amplitude of the pulse signal with the time stamp condition. This provides a precise timestamp for each pulse in the compensated pulse signal sequence, resulting in a set of pulse time positions.

[0106] Step 302: Calculate the time interval between adjacent pulse time positions in the pulse time position set to generate the original pulse repetition interval set.

[0107] In this step, the original pulse repetition interval set refers to the original dataset of the time position differences between adjacent pulses.

[0108] In this embodiment, the time points in the pulse time position set are arranged in chronological order, the difference between adjacent time points is calculated, and the original pulse repetition interval set is generated.

[0109] Step 303: Based on the preset pulse repetition interval range, extract the effective pulse repetition interval from the original pulse repetition interval set to obtain the effective pulse repetition interval set.

[0110] In this step, the preset pulse repetition interval range refers to the effective interval value range set according to the radar type, for example, 1-100ms for urban traffic radar. The effective pulse repetition interval set refers to the pulse interval dataset that conforms to the preset range.

[0111] In this embodiment of the application, according to the preset pulse repetition interval range, the interval values ​​in the original pulse repetition interval set that fall within this range are filtered, and after removing outliers, a valid pulse repetition interval set is generated.

[0112] Step 304: Based on the preset statistical distribution characteristics, the effective pulse repetition interval set is grouped to generate a pulse signal group set containing the pulse repetition interval characteristics.

[0113] In this step, the preset statistical distribution characteristic refers to the width threshold of the interval value clustering range, which is used to define the maximum permissible deviation of the interval between signals from the same source. The pulse signal group set refers to the set of pulse groups divided according to the interval value clustering characteristic, with each group representing a potential radar signal from the same source.

[0114] In this embodiment of the application, based on the distribution density of statistical interval values ​​of the clustering interval width based on the preset statistical distribution characteristics, the interval values ​​within the density peak interval are grouped together to generate a pulse signal group set containing pulse repetition interval characteristics.

[0115] Step 305: Match the pulse signal group set with the preset radar source identifier to obtain the feature data set.

[0116] In this step, the preset radar source identifier refers to the pre-established radar source feature template library.

[0117] In this embodiment of the application, the interval group in the pulse group set is matched with the preset radar source identifier and the source identifier is bound to generate a feature data set.

[0118] This application embodiment solves the problem of interval feature fragmentation caused by signal loss and interference by dynamically statistically grouping and matching radar source identifiers, thereby improving the radar source identification accuracy in multi-source aliasing scenarios.

[0119] This application provides a specific embodiment. Step 104 involves performing cyclostationary analysis on the compensated pulse signal sequence to obtain a cyclostationary feature set containing periodic component information. This specifically includes the following steps:

[0120] Step 401: The compensated pulse signal sequence is segmented to obtain multiple pulse signal segments.

[0121] In this step, multiple pulse signal segments refer to discrete signal units formed by dividing the signal into time windows.

[0122] In this embodiment, the continuous pulse stream is divided into equal-length segments by cutting the compensated pulse signal sequence with a fixed time window, thus obtaining multiple pulse signal segments.

[0123] Step 402: Perform time-frequency conversion on each pulse signal segment to generate a time-frequency distribution set.

[0124] In this step, the time-frequency distribution set refers to the two-dimensional energy matrix generated by the time-frequency conversion of the pulse signal segment, with the horizontal axis representing time units and the vertical axis representing frequency units, and the unit value being energy intensity.

[0125] In this embodiment, the short-time Fourier transform technique is used to decompose each pulse signal segment into two-dimensional units of time and frequency, and the signal energy intensity in each unit is calculated. The energy intensity of all units constitutes a time-frequency distribution set.

[0126] Step 403: Based on the preset periodic stability condition, detect the time-frequency regions in the time-frequency distribution set where the energy intensity fluctuates periodically, so as to determine the candidate time-frequency regions that meet the preset periodic fluctuation amplitude change threshold.

[0127] In this step, the preset periodic stability condition refers to the combination of two thresholds for periodic determination. The preset periodic fluctuation amplitude change threshold refers to the maximum allowable difference in amplitude between adjacent energy peaks. The candidate time-frequency region refers to the rectangular time-frequency block that passes the periodic stability detection.

[0128] In this embodiment, based on the dual conditions that the energy peak interval variance is less than a preset peak interval stability threshold and the difference between adjacent peak amplitudes is less than a preset periodic fluctuation amplitude change threshold, the time-frequency plane is scanned to identify periodic regions and a candidate time-frequency region set is generated.

[0129] Step 404: Measure the time interval between adjacent energy peaks in each candidate time-frequency region to generate a set of regional period values.

[0130] In this step, adjacent energy peaks refer to two local energy maxima that occur consecutively along the time axis within the time-frequency region. The set of regional periodic values ​​refers to the set of periodic measurements for each candidate time-frequency region.

[0131] In this embodiment of the application, the positions of continuous energy peaks within the candidate time-frequency region are extracted, the peak time difference is calculated, and a set of regional periodic values ​​is generated.

[0132] Step 405: Calculate the difference between the highest and lowest frequencies of each candidate time-frequency region in the candidate time-frequency region set to obtain the region frequency span set.

[0133] In this step, the regional frequency span set refers to the set of bandwidths of each candidate time-frequency region.

[0134] In this embodiment, the frequency boundaries of the candidate time-frequency regions are located, and the bandwidth value is obtained by subtracting the lowest frequency from the highest frequency. The bandwidth values ​​of all candidate time-frequency regions are integrated according to the region number to form a set of region frequency spans.

[0135] Step 406: Perform parameter fusion on the set of regional frequency spans and the set of regional periodic values ​​to generate a set of periodic component parameters.

[0136] In this step, the periodic component parameter set refers to a two-dimensional parameter set that fuses the period value and bandwidth, with each parameter corresponding to a feature vector of a candidate time-frequency region.

[0137] In this embodiment of the application, the periodic value and frequency span value of the same region are combined into a two-dimensional feature vector to obtain a set of periodic component parameters.

[0138] Step 407: Based on the preset radar source periodic characteristics, classify the source of the periodic component parameter set to include a cyclic stationary feature set containing periodic component information.

[0139] In this step, the preset radar source periodic characteristics refer to the database of known radar source periodic and bandwidth characteristic ranges.

[0140] In this embodiment of the application, the period and bandwidth feature range in the preset radar source periodic features are compared, and the matching periodic component parameters are classified into the corresponding radar source category to generate a cyclic stationary feature set of periodic component information.

[0141] This application embodiment enhances the ability to capture the cyclic features of frequency hopping signals by extracting the periodic characteristics in the time and frequency domain and fusing two parameters, solves the problem of periodic annihilation, and provides highly discriminative features for dense signal sorting.

[0142] This application provides a specific embodiment. Step 403 involves detecting time-frequency regions in the time-frequency distribution set where the energy intensity fluctuates periodically, based on a preset periodic stability condition, to determine candidate time-frequency regions that meet a preset threshold for the amplitude change of periodic fluctuations. This specifically includes the following steps:

[0143] Step 411: Based on the preset periodic stability condition, detect the time-frequency regions in the time-frequency distribution set where the energy intensity fluctuates periodically, in order to determine multiple peak positions.

[0144] In this step, the peak position refers to the coordinate point on the time-frequency plane where the energy intensity is locally maximized.

[0145] In this embodiment, based on a preset periodic stability condition, each frequency unit of the time-frequency distribution set is scanned by a sliding window to identify the maximum point where the energy intensity is greater than that of the adjacent time-frequency unit and fluctuates periodically. The time and frequency coordinates are recorded to determine multiple peak positions.

[0146] Step 412: Calculate the time interval variance between the peak positions. When the time interval variance is less than a preset peak interval stability threshold, the peak position is taken as a candidate peak position.

[0147] In this step, the preset peak interval stability threshold refers to the upper limit of the allowable peak time interval fluctuation, used to determine periodic stability. Candidate peak positions refer to the set of reliable peak positions filtered by interval variance, reflecting the locations of potential signal sources that satisfy periodic consistency.

[0148] In this embodiment of the application, the peak positions of the same frequency unit are sorted by time, and the variance of the time interval between adjacent peaks is calculated. If the variance is less than the preset peak interval stability threshold, all peak positions of the frequency unit are retained as candidate peak positions.

[0149] Step 413: Based on the preset time overlap threshold and the preset peak interval similarity threshold, perform time feature matching on the candidate peak positions to select the target peak positions that simultaneously meet the time overlap requirement and the interval similarity requirement.

[0150] In this step, the preset time coincidence threshold refers to the minimum required proportion of peak time overlap between different frequency units. The preset peak interval similarity threshold refers to the maximum allowable difference in peak interval between different frequency units. The time coincidence requirement means that the proportion of peak time coordinate overlap between spatially adjacent frequency units must be greater than the preset threshold, reflecting the signal's time-domain synchronization. The interval similarity requirement means that the difference in peak interval between spatially adjacent frequency units must be less than the similarity threshold, characterizing the signal's periodic consistency. The target peak position refers to the set of peak positions that simultaneously satisfy both time coincidence and interval similarity, identifying the precise distribution of the same source signal in the time-frequency plane.

[0151] In this embodiment of the application, candidate peak positions of spatially adjacent frequency units are matched. First, it is verified whether the time overlap meets the preset time overlap threshold. Second, the peak interval difference is calculated. If the interval difference is less than the preset peak interval similarity threshold, the target peak position that meets both the time overlap requirement and the interval similarity requirement is selected.

[0152] Step 414: Verify the energy amplitude change in the time-frequency region where the target peak position is located, and select the time-frequency region where the difference between adjacent peak amplitudes is less than the preset periodic fluctuation amplitude change threshold as the candidate time-frequency region.

[0153] In this embodiment of the application, continuous energy peaks within the time-frequency region where the target peak location is located are extracted, and the absolute value of the amplitude difference between adjacent peaks is calculated. If all amplitude differences are less than the preset periodic fluctuation amplitude change threshold, the time-frequency block is marked as a candidate time-frequency region.

[0154] The embodiments of this application accurately separate co-origin periodic signals through a spatiotemporal dual verification mechanism, overcoming the feature annihilation problem caused by frequency hopping interference and signal fragmentation, and improving the robustness of periodic region detection.

[0155] This application provides a specific embodiment. Step 404 involves measuring the time interval between adjacent energy peaks in each candidate time-frequency region to generate a set of regional periodic values. This specifically includes the following steps:

[0156] Step 421: Sort the target peak positions corresponding to each candidate time-frequency region in ascending order to obtain the corresponding sorted target peak position sequence;

[0157] In this step, the sorted target peak position sequence refers to the linear sequence formed by arranging the target peak positions in ascending order of time values. The sequence elements are time coordinate values, reflecting the temporal order of the energy peaks.

[0158] In this embodiment of the application, the time coordinate values ​​in the target peak position set are extracted and reordered in ascending order of value to generate a sorted target peak position sequence.

[0159] Step 422: Calculate the time interval between adjacent target peak positions in each sorted target peak position sequence to generate multiple original time interval sequences.

[0160] In this step, multiple original time interval sequences refer to the interval value sequence generated for each candidate time-frequency region, and the sequence elements are the time differences between adjacent target peak positions.

[0161] In this embodiment of the application, two consecutive target peak positions are taken from the sorted sequence, and the difference between their time values ​​is calculated to generate the original time interval sequence.

[0162] Step 423: Calculate the absolute difference between adjacent time intervals in each original time interval sequence, and select a stable time interval sequence from the corresponding original time interval sequence whose absolute difference does not exceed the preset interval tolerance.

[0163] In this step, the preset interval tolerance refers to the upper limit of the allowable fluctuation between adjacent periods, used to determine the period stability. The stable time interval sequence refers to the continuous interval segment that has passed the tolerance screening, reflecting high-stability periodic characteristics.

[0164] In this embodiment of the application, the absolute difference is calculated for consecutive interval pairs of the original time interval sequence, and only consecutive interval segments with the difference less than or equal to the preset interval tolerance are retained to generate a stable time interval sequence that meets the stability requirements.

[0165] Step 424: Take the average value of each stable time interval sequence as the regional period value of the corresponding candidate time-frequency region, and aggregate the regional period values ​​of all candidate time-frequency regions to obtain the set of regional period values.

[0166] In this step, the regional periodic value refers to the arithmetic mean of the stable time interval sequence.

[0167] In this embodiment of the application, the arithmetic mean of all interval values ​​within the stable time interval sequence is calculated, and the result is used as the regional period value of the corresponding candidate time-frequency region. Finally, the regional period values ​​of all candidate time-frequency regions are summarized to obtain the set of regional period values.

[0168] The embodiments of this application accurately extract the anti-interference period value through a dual stability verification mechanism, which solves the problem of the superposition of Doppler period jitter and instantaneous interference caused by vehicle movement, and improves the reliability of the period characteristics.

[0169] This application provides a specific embodiment. Step 105 involves fusing and sorting the feature data set and the cyclically stationary feature set according to a preset radar signal sorting rule to generate a sorting result for a multi-source aliased radar pulse signal sequence. This specifically includes the following steps:

[0170] Step 501: Based on the preset main matching rule of radar signal sorting rules, the pulse repetition interval feature in the feature data set and the periodic component information in the cyclic stationary feature set are matched to generate a fused feature group set.

[0171] In this step, the main matching rule refers to the association logic between the pulse repetition interval feature and the periodic component information. The pulse repetition interval feature refers to the pulse interval dataset grouped by radar source. The periodic component information refers to the time-frequency periodic parameter set. The fused feature set refers to the set of feature combinations successfully associated through the main matching rule.

[0172] In this embodiment of the invention, based on the preset radar signal sorting rules, the correlation strength between the pulse repetition interval feature and the periodic component information in the cyclic stationary feature set is calculated. When the interval value of the pulse repetition interval feature and the periodic value of the periodic component information satisfy the time ratio constraint, feature association is established. The successfully matched feature groups are merged to generate a fused feature group set.

[0173] Step 502: Based on the preset radar signal sorting rules and auxiliary matching rules, match the periodic parameters of each fusion feature group in the fusion feature group set with the preset radar source feature template to generate a fusion feature group set with radar source identifier.

[0174] In this step, the auxiliary matching rule refers to the matching logic between the fused feature group and the radar source template. The preset radar source feature template refers to the database of feature parameters of known radar sources.

[0175] In this embodiment of the invention, based on the auxiliary matching rules of the preset radar signal sorting rules, the periodic parameters of each fusion feature group in the fusion feature group set are extracted, and the inclusion of the parameter range in the radar source feature template library is verified. Radar source identifiers are assigned to the successfully matched feature groups, and a fusion feature group set with radar source identifiers is generated.

[0176] Step 503: Identify the pulse signal corresponding to the time feature parameter of the fused feature group set from the compensated pulse signal sequence, and generate the feature group pulse set.

[0177] In this step, the feature group pulse set refers to the pulse dataset located by matching through a time window.

[0178] In this embodiment of the invention, based on the time feature parameters of the fused feature group set, pulses located within the time window in the compensated pulse signal sequence are scanned to generate a feature group pulse set.

[0179] Step 504: Based on the acquisition time sequence of the multi-source aliased radar pulse signal sequence, reassemble the pulse signals with the same radar source identifier in the feature group pulse set to obtain the sorting result of the multi-source aliased radar pulse signal sequence.

[0180] In this step, the same radar source identifier refers to the unique code of the radar source assigned by the auxiliary matching rules, which is used to identify signal pulses from the same source.

[0181] In this embodiment of the invention, all pulse signals with the same radar source identifier are sorted according to the acquisition time sequence of the original multi-source aliased radar pulse signal sequence. After removing duplicate pulses, a time-continuous radar source pulse sequence is generated, which constitutes the sorting result of the multi-source aliased radar pulse signal sequence.

[0182] This application embodiment solves the problem of feature and pulse correlation breakage caused by signal aliasing by using a two-level feature matching and time window backtracking mechanism, thereby achieving high-precision separation and recombination of multi-source radar pulses.

[0183] Figure 2 This application provides a schematic diagram of a radar signal sorting system based on cyclic stationary analysis, as shown in the embodiment. Figure 2 As shown, the system includes:

[0184] Acquisition module 21 is used to acquire superimposed pulses from different radar sources to obtain a multi-source aliased pulse signal sequence;

[0185] The compensation module 22 is used to perform medium attenuation compensation processing on the multi-source aliasing pulse signal sequence based on the physical acoustic propagation principle, and generate a compensated pulse signal sequence that suppresses the propagation distortion of complex media.

[0186] Extraction module 23 is used to extract features from the compensated pulse signal sequence according to a preset pulse repetition interval feature extraction rule, so as to obtain a feature data set containing pulse repetition interval features;

[0187] Analysis module 24 is used to perform cyclostationary analysis on the compensated pulse signal sequence to obtain a cyclostationary feature set containing periodic component information;

[0188] The sorting module 25 is used to perform fusion sorting processing on the feature data set and the cyclic stationary feature set according to the preset radar signal sorting rules, and generate the sorting result of the multi-source aliased radar pulse signal sequence.

[0189] Figure 2 The radar signal sorting system based on cyclic stationary analysis described above can perform... Figure 1 The implementation principle and technical effects of the radar signal sorting method based on cyclostationary analysis described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the radar signal sorting system based on cyclostationary analysis in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0190] In one possible design, Figure 2 The radar signal sorting system based on cyclostationary analysis shown in the embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0191] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0192] The processing component 32 is used for the above Figure 1 The embodiment describes a radar signal sorting method based on cyclic stationary analysis.

[0193] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0194] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0195] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0196] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0197] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0198] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0199] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is a radar signal sorting method based on cyclic stationary analysis.

[0200] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0201] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0202] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A radar signal sorting method based on cyclic spectral analysis, characterized in that, The method comprises the following steps: Collecting superimposed pulses of different radar emission sources to obtain a multi-source mixed superimposed pulse signal sequence; Based on the principle of physical acoustics propagation, the medium attenuation compensation processing is performed on the multi-source mixed superimposed pulse signal sequence to generate a compensated pulse signal sequence that suppresses the distortion of complex medium propagation; According to a preset pulse repetition interval feature extraction rule, feature data set containing pulse repetition interval features is obtained by extracting features from the compensated pulse signal sequence; Cyclic stationary analysis is performed on the compensated pulse signal sequence to obtain a cyclic stationary feature set containing periodic component information; According to a preset radar signal sorting rule, the feature data set and the cyclic stationary feature set are fused and sorted to generate a sorting result of the multi-source mixed superimposed radar pulse signal sequence.

2. The method of claim 1, wherein, Based on the principle of physical acoustics propagation, the medium attenuation compensation processing is performed on the multi-source mixed superimposed pulse signal sequence to generate a compensated pulse signal sequence that suppresses the distortion of complex medium propagation, comprising: Based on the principle of physical acoustics propagation, the propagation path of the radar signal in the complex medium is analyzed to obtain a path node coordinate set; The medium attribute data of each path node in the path node coordinate set is identified to separate the medium density parameter set and the medium propagation velocity parameter set from the medium attribute data; Based on the medium density parameter set and the propagation velocity parameter set, the path scattering loss and the path phase delay are calculated respectively, and the path scattering loss and the path phase delay are combined to obtain an attenuation compensation coefficient set; The propagation path type of each pulse signal in the multi-source mixed superimposed pulse signal sequence is identified to obtain an intermediate pulse signal sequence with propagation path identification; According to the attenuation compensation coefficient set, the intermediate pulse signal sequence is corrected to obtain a corrected pulse signal sequence; The corrected pulse signal sequence is recombined to obtain a compensated pulse signal sequence that suppresses the distortion of complex medium propagation.

3. The method of claim 1, wherein, According to a preset pulse repetition interval feature extraction rule, feature data set containing pulse repetition interval features is obtained by extracting features from the compensated pulse signal sequence, comprising: According to the time mark condition of the preset pulse repetition interval feature extraction rule, the pulse signals in the compensated pulse signal sequence are position marked to obtain a pulse time position set; The time interval between adjacent pulse time positions in the pulse time position set is calculated to generate an original pulse repetition interval set; Based on a preset pulse repetition interval range, valid pulse repetition intervals are extracted from the original pulse repetition interval set to obtain a valid pulse repetition interval set; According to a preset statistical distribution feature, the valid pulse repetition interval set is grouped to generate a pulse signal grouping set containing pulse repetition interval features; The pulse signal grouping set and the preset radar source identification are matched to obtain the feature data set.

4. The method of claim 1, wherein, Cyclic stationary analysis is performed on the compensated pulse signal sequence to obtain a cyclic stationary feature set containing periodic component information, comprising: The compensated pulse signal sequence is segmented to obtain a plurality of pulse signal segments; perform time-frequency conversion on each pulse signal segment to generate a time-frequency distribution set; detect time-frequency regions with periodic energy intensity fluctuation in the time-frequency distribution set according to a preset periodic stability condition to determine candidate time-frequency regions that satisfy a preset periodic fluctuation amplitude change threshold; measure time intervals between adjacent energy peaks of each candidate time-frequency region to generate a region period value set; calculate a difference between the highest frequency and the lowest frequency of each candidate time-frequency region in the candidate time-frequency region set to obtain a region frequency span set; perform parameter fusion on the region frequency span set and the region period value set to generate a periodic component parameter set; classify the periodic component parameter set according to a preset radar source periodic characteristic to obtain a cyclostationary feature set containing periodic component information.

5. The method of claim 4, wherein, detect time-frequency regions with periodic energy intensity fluctuation in the time-frequency distribution set according to a preset periodic stability condition to determine candidate time-frequency regions that satisfy a preset periodic fluctuation amplitude change threshold, including: detect time-frequency regions with periodic energy intensity fluctuation in the time-frequency distribution set according to a preset periodic stability condition to determine multiple peak position; calculate a time interval variance between the peak positions, and when the time interval variance is less than a preset peak interval stability threshold, the peak positions are taken as candidate peak positions; perform time feature matching on the candidate peak positions based on a preset time coincidence degree threshold and a preset peak interval similarity threshold to select target peak positions that satisfy both the time coincidence degree requirement and the interval similarity requirement; perform energy amplitude change verification on a time-frequency region where the target peak position is located to take a time-frequency region with an adjacent peak amplitude difference less than a preset periodic fluctuation amplitude change threshold as a candidate time-frequency region.

6. The method of claim 4, wherein, measure time intervals between adjacent energy peaks of each candidate time-frequency region to generate a region period value set, including: arrange target peak positions corresponding to each candidate time-frequency region in ascending order to obtain a corresponding sorted target peak position sequence; calculate time intervals between adjacent target peak positions in each sorted target peak position sequence to generate multiple original time interval sequences; calculate absolute differences of adjacent time intervals in each original time interval sequence, and select stable time interval sequences with absolute differences not exceeding a preset interval tolerance from the corresponding original time interval sequences; take an average value of each stable time interval sequence as a region period value of the corresponding candidate time-frequency region, and aggregate region period values of all candidate time-frequency regions to obtain a region period value set.

7. The method of claim 1, wherein, perform fusion and sorting processing on the feature data set and the cyclostationary feature set according to a preset radar signal sorting rule to generate a sorting result of the multi-source mixed radar pulse signal sequence, including: perform feature association matching on pulse repetition interval features in the feature data set and periodic component information in the cyclostationary feature set based on a main matching rule of the preset radar signal sorting rule to generate a fusion feature group set; According to an auxiliary matching rule based on a preset radar signal sorting rule, the period parameter of each fusion feature group in the fusion feature group set is matched with a preset radar source feature template, and a fusion feature group set with a radar source identifier is generated; Pulse signals corresponding to the time feature parameters of the fusion feature group set are identified from the compensated pulse signal sequence, and a feature group pulse set is generated; According to the acquisition time sequence of the multi-source mixed radar pulse signal sequence, the pulse signals with the same radar source identifier in the feature group pulse set are reorganized, and a sorting result of the multi-source mixed radar pulse signal sequence is obtained.

8. A radar signal sorting system based on cyclostationary analysis, characterized by, Comprise: The acquisition module is used for acquiring superimposed pulses of different radar transmitting sources, and obtaining a multi-source mixed pulse signal sequence; The compensation module is used for performing medium attenuation compensation processing on the multi-source mixed pulse signal sequence based on the physical acoustic propagation principle, and generating a compensated pulse signal sequence that suppresses complex medium propagation distortion; The extraction module is used for performing feature extraction on the compensated pulse signal sequence according to a preset pulse repetition interval feature extraction rule, and obtaining a feature data set containing pulse repetition interval features; The analysis module is used for performing cyclostationary analysis on the compensated pulse signal sequence, and obtaining a cyclostationary feature set containing period component information; The sorting module is used for performing fusion sorting processing on the feature data set and the cyclostationary feature set according to a preset radar signal sorting rule, and generating a sorting result of the multi-source mixed radar pulse signal sequence.

9. A computing device, comprising: Comprise a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component, and realize a radar signal sorting method based on cyclostationary analysis as claimed in any one of claims 1-7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, a radar signal sorting method based on cyclostationary analysis as claimed in any one of claims 1-7 is realized.

Images