Sliding window flow point trace condensation processing method and system based on three-coordinate radar

Through the sliding window flow-through treatment method and the optimized center of mass method, the problems of high complexity and low accuracy of radar point trace condensation treatment technology are solved, and efficient real-time target tracking is achieved in complex scenarios.

CN115508787BActive Publication Date: 2024-11-08四川九洲防控科技有限责任公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211046440.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-30
Publication Date
2024-11-08
Estimated Expiration
2042-08-30

AI Technical Summary

Technical Problem

The existing radar point trace condensation processing technology has high complexity, large computing resources, long processing delay, and low accuracy in high-density multi-target scenarios, making it difficult to meet the real-time target tracking requirements under complex interference and clutter.

Method used

Using sliding window flow-flowing point trace condensation treatment method based on three-coordinate radar, the center of mass method is optimized by abnormal echo point trace removal, intra-frame point trace condensation and inter-frame point trace condensation treatment, combined with the secondary and first-level flow-flow treatment architectures , reduce data volume and complexity, and improve accuracy.

Benefits of technology

It effectively reduces the data volume and complexity of point trace aggregation processing, reduces processing delay, improves the accuracy of point trace aggregation processing, and realizes real-time target tracking in complex scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115508787B_ABST
    Figure CN115508787B_ABST
Patent Text Reader

Abstract

The present invention proposes a sliding window pipeline point trace condensation processing method and system based on three-coordinate radar. The method includes: firstly, through hierarchical and dimensional pipeline processing, the original point trace condensation processing within the CPI is realized, then the point trace condensation processing between CPIs within the beam is realized, and finally the point trace condensation processing between different beams is realized, and the original point trace data is condensed in three dimensions of distance, azimuth, and pitch. At the same time, through the sliding window pipeline processing, the number of points involved in the condensation is continuously compressed one by one, effectively reducing the data volume and complexity of the point trace condensation processing, reducing the processing delay, and improving the accuracy of the point trace condensation processing through the continuous statistics, extraction and optimized centroid estimation method of the condensed point trace feature information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of radar point trace condensation processing, and in particular relates to a sliding window flow type point trace condensation processing method and system based on three-coordinate radar. Background Art

[0002] With the increasing complexity of radar detection environment, especially for the current scenario of multiple batches and multiple targets under complex interference and clutter, in order to achieve accurate detection and accurate and stable tracking of targets, the radar system and working mode have become more complex, and the waveform of multi-beam and multi-pulse group has been more widely used. The algorithm complexity, echo data volume and calculation amount of the corresponding radar terminal processing system are increasing, and the data rate is faster. The radar terminal processing system needs to complete the processing of echo data in a very short time to provide target information, requiring the radar to have high real-time performance and accuracy.

[0003] As an important link between the previous and the next, it plays an indispensable role in obtaining accurate target parameters, ensuring stable tracking of targets, reducing clutter and interference, and compressing the amount of track input point data. The great increase in the amount of input point data has also put higher demands on its processing equipment, especially in the current international environment where domestic servers, boards, and operating systems are vigorously promoted and applied, but the processing performance cannot keep up. It is particularly important to improve the accuracy and processing efficiency of point condensation processing and reduce its occupation of computing and storage resources.

[0004] Disadvantages of existing technology

[0005] At present, the radar point trace aggregation processing technology sets the distance, azimuth, and pitch thresholds according to the target characteristics to form a three-dimensional spatial wave gate, and finds the point trace data belonging to the same target from the original point trace data according to the spatial wave gate criterion, obtains the point trace clusters of multiple targets, puts them into the storage buffer, and waits for all the original points in the CPI (pulse group) to be input in sequence to make an association judgment with each point trace of each point trace cluster, merges them into a certain target point trace cluster or generates a new target point trace cluster, until the target splitting end criterion (set according to the target characteristics) is reached, and then the point trace cluster is condensed according to the centroid method to obtain the target estimated position and other characteristic parameters.

[0006] It can be seen that the traditional radar point trace condensation processing method includes three layers of large number of point trace groups and cyclic comparison and association of original point traces. It is highly complex and requires a lot of computing resources. It also takes up a lot of storage space in the processor to cache the intermediate process point trace group data. Moreover, the complexity will continue to increase in the process of merging and resolving the original point traces, especially in the high-density multi-target air situation. According to the target characteristics (splitting end criterion), at least two processing sectors must be delayed before the target point traces can be output. In addition, due to the possible presence of clutter and interference in the point trace group, the target information obtained based on the single centroid method may have a large deviation and low accuracy. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes a technical solution of a sliding window pipeline point trace condensation processing method and system based on three-coordinate radar to solve the above technical problems.

[0008] The first aspect of the present invention discloses a sliding window flow point trace condensation processing method based on three-coordinate radar, the method comprising:

[0009] Step S1, abnormal echo point trace removal: the point trace data is removed according to the N-fold beam width range and the sum-difference ratio in the original point trace data;

[0010] Step S2, intra-frame point trace condensation: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces in the pulse group after the abnormal echo point traces are removed in the distance dimension, put the condensed points in the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation process in the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation process, and again classify, merge and condense all the data in the same beam in the cache in the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete the intra-frame point trace condensation process once;

[0011] Step S3, inter-frame point trace condensation and output: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data directly into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, use the new target condensation point information to update the old point trace information in the buffer queue, and the point trace information of the newly added target is directly added to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target widening characteristics in azimuth, and after the target point trace end criterion is reached, it is pipelined output.

[0012] According to the method of the first aspect of the present invention, in step S1, the specific method of removing the point trace data according to the N-fold beam width range includes:

[0013] Eliminate the point data outside the range of N times the beam width, where 1.2≤N≤1.5;

[0014] The specific method for removing the point trace data according to the sum-difference ratio in the original point trace data includes:

[0015] When the sum-difference ratio of the original points in the pulse group has both values ​​within the boundary conditions and values ​​outside the boundary conditions, the point data of the values ​​outside the boundary conditions are removed; when the sum-difference ratio is the value outside the boundary conditions, all the original points are retained.

[0016] According to the method of the first aspect of the present invention, in step S2, the basis for distinguishing the traces of different targets in the distance dimension includes: target signal-to-noise ratio, distance and Doppler.

[0017] According to the method of the first aspect of the present invention, in step S2, the classification and merging criteria include:

[0018] The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, the point data that does not match the signal-to-noise ratio is first filtered out, and then it is determined whether the distance unit difference of the point data is less than the distance threshold and whether the Doppler scale difference is less than the Doppler threshold.

[0019] According to the method of the first aspect of the present invention, in step S2, the classification and merging criteria also include: if the signal processing does not provide Doppler information, the change characteristics of the target amplitude are used to replace the Doppler information to distinguish target points that are adjacent or intersecting in distance.

[0020] According to the method of the first aspect of the present invention, in step S2, the method further includes:

[0021] The traditional centroid method is optimized, and then amplitude weighting or power weighting is used to obtain the distance, azimuth and elevation of the condensed point traces.

[0022] According to the method of the first aspect of the present invention, in step S3, the merging criteria for the inter-frame point trace condensation include:

[0023] The same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold.

[0024] The second aspect of the present invention discloses a sliding window flow point trace condensation processing system based on three-coordinate radar, the system comprising:

[0025] The first processing module is configured to remove abnormal echo traces: remove the trace data according to the N-fold beam width range and the sum-difference ratio in the original trace data;

[0026] The second processing module is configured to: condense the point traces within the frame: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces within the pulse group after the abnormal echo point traces are removed, put the condensed points within the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation processing within the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation processing within the pulse group, and again classify, merge and condense all the data in the same beam in the cache on the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete one frame point trace condensation processing;

[0027] The third processing module is configured to condense and output point traces between frames: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data directly into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, update the old point trace information in the buffer queue with the new target condensation point information, and directly add the point trace information of the newly added target to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target's widening characteristics in azimuth, and after the target point trace end criterion is reached, it is output in a pipeline manner.

[0028] According to the system of the second aspect of the present invention, the first processing module is configured to remove the point trace data according to the N-fold beam width range, specifically including:

[0029] Eliminate the point data outside the range of N times the beam width, where 1.2≤N≤1.5;

[0030] The specific method for removing the point trace data according to the sum-difference ratio in the original point trace data includes:

[0031] When the sum-difference ratio of the original points in the pulse group has both values ​​within the boundary conditions and values ​​outside the boundary conditions, the point data of the values ​​outside the boundary conditions are removed; when the sum-difference ratio is the value outside the boundary conditions, all the original points are retained.

[0032] In the system according to the second aspect of the present invention, the second processing module is configured to distinguish the traces of different targets in the distance dimension based on the following criteria: target signal-to-noise ratio, distance, and Doppler.

[0033] According to the system of the second aspect of the present invention, the second processing module is configured such that the criteria for classification and merging include:

[0034] The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, the point data that does not match the signal-to-noise ratio is first filtered out, and then it is determined whether the distance unit difference of the point data is less than the distance threshold and whether the Doppler scale difference is less than the Doppler threshold.

[0035] According to the system of the second aspect of the present invention, the second processing module is configured such that the classification and merging criteria also include: if the signal processing does not provide Doppler information, then using the change characteristics of the target amplitude to replace the Doppler information to distinguish target points that are adjacent or intersecting in distance.

[0036] According to the system of the second aspect of the present invention, the second processing module is configured to optimize the traditional centroid method and then use amplitude weighting or power weighting to obtain the distance, azimuth and elevation of the condensed point traces.

[0037] According to the system of the second aspect of the present invention, the third processing module is configured as follows: the merging criteria of the inter-frame point trace condensation include:

[0038] The same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold.

[0039] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in any one of the sliding window flow point trace condensation processing methods based on three-coordinate radar in the first aspect of the present disclosure are implemented.

[0040] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the sliding window pipeline point trace condensation processing methods based on three-coordinate radar in the first aspect of the present disclosure are implemented.

[0041] The solution proposed in the present invention first realizes the condensation processing of original point traces within the CPI, then realizes the condensation processing of point traces between CPIs within the beam, and finally realizes the condensation processing of point traces between different beams through hierarchical and dimensional, level-by-level pipeline processing, completing the correlation condensation processing of original point trace data in three dimensions of distance, azimuth, and pitch. At the same time, through sliding window pipeline processing, the number of points involved in condensation is continuously compressed level by level, effectively reducing the data volume and complexity of point trace condensation processing, reducing processing delay, and improving the accuracy of point trace condensation processing through continuous statistics, extraction and optimized centroid estimation method of condensed point trace feature information. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 It is a flow chart of a sliding window pipeline point trace condensation processing method based on three-coordinate radar according to an embodiment of the present invention;

[0044] Figure 2 is a beam distribution diagram according to an embodiment of the present invention;

[0045] Figure 3 is a flow chart of intra-frame condensation processing according to an embodiment of the present invention;

[0046] Figure 4 is a flow chart of inter-frame condensation processing according to an embodiment of the present invention;

[0047] Figure 5a is the original point trace data according to an embodiment of the present invention;

[0048] Figure 5b The original point trace data on the two-dimensional plane according to the embodiment of the present invention;

[0049] Figure 5c is a condensed point trace within the CPI according to an embodiment of the present invention;

[0050] Figure 5d is a cohesion point trace between CPIs within a beam according to an embodiment of the present invention;

[0051] Figure 5e is an inter-beam condensation point trace according to an embodiment of the present invention;

[0052] Figure 5f is a cohesion point trace between beams on a two-dimensional plane according to an embodiment of the present invention;

[0053] Figure 6 It is a structural diagram of a sliding window pipeline point trace condensation processing system based on a three-coordinate radar according to an embodiment of the present invention;

[0054] Figure 7 The figure is a structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0056] The first aspect of the present invention discloses a sliding window flow point trace condensation processing method based on three-coordinate radar. Figure 1 FIG. 1 is a flow chart of a sliding window flow point trace condensation processing method based on a three-coordinate radar according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0057] Step S1, abnormal echo point trace removal: the point trace data is removed according to the N-fold beam width range and the sum-difference ratio in the original point trace data;

[0058] Step S2, intra-frame point trace condensation: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces in the pulse group after the abnormal echo point traces are removed in the distance dimension, put the condensed points in the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation process in the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation process, and again classify, merge and condense all the data in the same beam in the cache in the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete the intra-frame point trace condensation process once;

[0059] Step S3, inter-frame point trace condensation and output: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data directly into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, use the new target condensation point information to update the old point trace information in the buffer queue, and the point trace information of the newly added target is directly added to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target widening characteristics in azimuth, and after the target point trace end criterion is reached, it is pipelined output.

[0060] In step S1, abnormal echo point traces are eliminated: point trace data are eliminated according to the N-fold beam width range and the sum-difference ratio contained in the original point trace data.

[0061] In some embodiments, in step S1, the specific method of removing point trace data according to the N-fold beam width range includes:

[0062] Eliminate the point data outside the range of N times the beam width, where 1.2≤N≤1.5;

[0063] The specific method for removing the point trace data according to the sum-difference ratio in the original point trace data includes:

[0064] When the sum-difference ratio of the original points in the pulse group has both values ​​within the boundary conditions and values ​​outside the boundary conditions, the point data of the values ​​outside the boundary conditions are removed; when the sum-difference ratio is the value outside the boundary conditions, all the original points are retained.

[0065] Specifically, since the original point data output by signal processing may contain clutter, noise, interference, etc. in addition to the real target data, obviously abnormal data can be filtered out in advance before the point correlation aggregation processing, thereby reducing the amount of subsequent processing data and the invalid calculation complexity.

[0066] According to the radar receiving beam characteristics, such as Figure 2 As shown in the figure, all the original points in the CPI should theoretically be within the receiving beam width. Based on the current radar beam center pointing (azimuth, pitch), the original points of each CPI recorded are processed for over-limit judgment in azimuth and elevation dimensions, and the elimination criteria are set: the point data outside the range of N times the beam width are eliminated. The parameter N can be dynamically configured according to the current working environment of the radar. Usually, N=1.2 or N=1.5 can be set to prevent small targets from entering from the side lobes and side lobes.

[0067] At the same time, according to the sum-difference ratio in the original point data, the data is screened. According to the scanning wave position characteristics of the radar beam in pitch, the effective boundary values ​​of the sum-difference ratio are different, which can be divided into low elevation angle area and high elevation angle area. When the original point data and the difference ratio in the CPI have both values ​​within the boundary conditions and values ​​outside the boundary conditions, the values ​​outside the boundary conditions are removed; when the sum-difference ratios are all values ​​outside the boundary conditions, all the original point data are retained and not filtered for the time being.

[0068] However, point filtering is generally only required when the processing capacity is saturated. Otherwise, all original point data should be retained as much as possible to avoid filtering out valid target points, which may have an adverse effect on subsequent processing in a low-observable target environment.

[0069] In step S2, intra-frame point trace condensation: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces in the pulse group after the abnormal echo point traces are removed, put the condensed points in the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group to complete the distance dimension point trace condensation in the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation in the pulse group, and again classify, merge and condense all the data in the same beam in the cache on the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete the intra-frame point trace condensation once.

[0070] In some embodiments, in step S2, the main basis for distinguishing the traces of different targets in the distance dimension includes: target signal-to-noise ratio, distance and Doppler.

[0071] The classification criteria include:

[0072] The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, the point data that does not match the signal-to-noise ratio is first filtered out, and then it is determined whether the distance unit difference of the point data is less than the distance threshold and whether the Doppler scale difference is less than the Doppler threshold.

[0073] The classification and merging criteria also include: if the signal processing does not provide Doppler information, then using the target amplitude variation characteristics to replace the Doppler information to distinguish target points that are adjacent or intersecting in distance.

[0074] The method further comprises:

[0075] The traditional centroid method is optimized, and then amplitude weighting or power weighting is used to obtain the distance, azimuth and elevation of the condensed point traces.

[0076] Specifically, intra-frame point trace aggregation refers to the association and aggregation of the original point traces of the same beam in the distance dimension. For the working waveform of the multi-beam multi-pulse group of the three-coordinate radar, the differences in azimuth and elevation of the original point traces in the same beam are very small, and the azimuth angle is roughly the same as the elevation angle, and the difference will not exceed one beam width. However, the extension of the general target echo in the distance is relatively narrow, and the extension in the azimuth is relatively wide, and most of them are greater than one CPI. Therefore, intra-frame point trace aggregation first classifies, merges and aggregates the original point traces of the same beam in the distance dimension, and then performs subsequent processing on the condensed points in the distance dimension.

[0077] For the point trace data in each CPI that is entered sequentially, a two-level pipeline processing architecture is adopted. In terms of distance dimension, the point trace data in the pulse group after the abnormal echo point traces are removed is first classified, merged and condensed. The condensed points in the pulse group after classification, merging and condensation are put into the cache, and the next pulse group point trace data is waited for to arrive. The distance dimension point trace condensation processing in the pulse group is completed until all the pulse groups in the same beam complete the distance dimension point trace condensation processing in the pulse group. All the data in the same beam in the cache are classified, merged and condensed in the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result and complete the intra-frame point trace condensation processing. In each level of pipeline processing, the targets are batched in advance in terms of distance, the original point trace data is compressed, and the complexity is reduced. Figure 3 The pipeline processing flow is shown below.

[0078] The main bases for distinguishing the traces of different targets in the distance dimension include: target signal-to-noise ratio, distance and Doppler.

[0079] The classification criteria include:

[0080] The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, first filter out the point data that does not match the signal-to-noise ratio, and then judge whether the distance unit difference of the point is less than the distance threshold, and whether the Doppler scale difference is less than the Doppler threshold. Generally, the distance threshold △R=2, which can be configured, and the Doppler threshold △Fd is determined according to the accuracy of the current radar filter. If the signal processing does not give Doppler information, the change characteristics of the target amplitude are used to replace the Doppler information to distinguish the target points that are adjacent or intersecting in distance. At this time, whether the original point amplitude difference is less than the amplitude threshold △Amp, the amplitude threshold △Amp is determined by the pulse width and the number of sampling points. If adjacent amplitude peak points appear, the relationship between the amplitude difference and the pulse compression main-to-sub ratio and the distance between them is judged.

[0081] In the process of associating and merging the original points, characteristic parameter information such as the target's range spread, starting and ending azimuth / pitch, number of pulses, amplitude envelope, etc. will be extracted and counted to facilitate subsequent evaluation of the target point quality and track quality and determine the target credibility.

[0082] In order to improve the estimation accuracy of the condensed points, we only select the average amplitude of all the original points participating in the condensation whose amplitude is greater than ceof=80% (configurable) The target location information is estimated by using the point traces, optimizing the traditional centroid method, and then using amplitude weighting or power weighting to obtain the distance, azimuth, elevation angle and other information of the condensed point traces. The calculation formula based on amplitude weighting is as follows:

[0083]

[0084]

[0085]

[0086] Among them, Σ 关联点 Represents the magnitude of the associated point, R 关联点 Indicates the distance of the associated point, Az 关联点 Indicates the orientation of the associated point, El 关联点 Indicates the elevation angle of the associated point.

[0087] The amplitude of the condensed point is the maximum amplitude of all the original points that meet the association criteria:

[0088] Condensation point amplitude = max{Σ 起前 ,Σ 起 ,Σ 存1 ,…,Σ 存N}

[0089] Among them, Σ i Indicates the magnitude of each associated trace.

[0090] In step S3, inter-frame point trace condensation and output: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data as the first inter-beam condensation result directly into the buffer queue, wait for the next beam point trace data to arrive, and classify and judge all the intermediate points in the buffer queue in turn, and directly complete the condensation processing for the point trace data identified as the same target, use the new target condensation point information to update the old point trace information in the buffer queue, and the point trace information of the newly added target is directly added to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target widening characteristics in azimuth, and after the target point trace end criterion is reached, it is pipelined output.

[0091] In some embodiments, in step S3, the merging criteria for the inter-frame point trace condensation include:

[0092] The same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold.

[0093] Specifically, inter-frame point trace condensation refers to the association and condensation of point traces between different beams in azimuth and elevation dimensions on the basis of intra-frame condensation, at which time the target's distance, azimuth, elevation and other parameter information are considered at the same time to form a spatial correlation wave gate for point trace condensation. The merging criteria for inter-frame point trace condensation include: same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold. The azimuth and elevation thresholds △A and △E are both related to the current antenna beam width. Generally, △A can be set to 1.5 times the 3dB azimuth beam width, and △E can be set to 1.5 times the 3dB elevation beam width, which can be configured. Similarly, if the signal processing does not provide Doppler information, the amplitude difference of the target in the azimuth sidelobe area and the main lobe width and the azimuth amplitude envelope characteristics are used to separate the points of adjacent observed targets.

[0094] For the point trace data of each beam input sequentially, at different pitches or azimuths, a one-level pipeline processing architecture is adopted to classify, merge and condense the beam point trace data that arrive in sequence according to the spatial correlation gate and merging criteria. The first beam point trace data that arrives is directly put into the buffer queue as the first inter-beam condensation result, and the next beam point trace data is waited for to arrive and all the intermediate points in the buffer queue are classified and judged in turn. The point trace data identified as the same target is directly condensed, and the old point trace information in the buffer queue is updated with the new target condensation point information, and the point trace information of the newly added target is directly added to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value, completing the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target's widening characteristics in azimuth, and after the target point trace end criterion is reached, it is pipelined output. In the pipeline processing, the target is separated in azimuth and elevation, and the target information is updated by real-time aggregation of existing traces, which will not cause the accumulation of trace data, reduce the data volume and complexity, and is compatible with elevation sequence or random scanning wave position mode, such as Figure 4 The pipeline processing flow is shown below.

[0095] The point trace condensation between frames also uses the optimized centroid method to calculate the point trace position after condensation based on the statistical amplitude information. Just like the point trace condensation within the frame, the characteristic parameters of the target's range widening number, starting and ending azimuth / pitch, and number of pulses are continuously updated during the association merging process, and the characteristic information such as the target's azimuth amplitude envelope is extracted.

[0096] After the inter-frame dot trace condensation is completed, the target dot trace end criteria are judged in turn according to the target's widening characteristics in azimuth. After reaching the target dot trace end criteria, pipeline output is performed. In azimuth, the dot trace data is divided into azimuth sectors. According to the azimuth beam width, the full range is divided into N = 128 (configurable) sectors, and the extension of the target in azimuth generally does not exceed △N = 2 (configurable) sectors. Therefore, the target dot trace end criterion (target splitting / output criterion) is that the difference between the dot trace azimuth and the current beam pointing is greater than 2 sectors, that is, if there is a target dot trace, there will be a dot trace output after scanning 2 sectors. At the same time, in addition to the ring mode test signal, the target dot trace whose start and end azimuths generally continue for more than 10° (configurable) can also end the output.

[0097] The above point trace condensation processing method is verified by using antenna ring mode test data. The original point trace data of the antenna scanning process with the nth circle = 13, the sector is [28°, 36°], and the pitch full wave position is recorded. The processing results are as follows Figures 5a-5f As shown in the figure, it can be seen that the targets of the ring model test are all around 200km and exist at 4 altitude levels.

[0098] In summary, the scheme proposed in the present invention can realize the condensation processing of the original point traces within the CPI through hierarchical and dimensional, level-by-level pipeline processing, firstly, realize the condensation processing of the point traces between the CPIs within the beam, and finally realize the condensation processing of the point traces between different beams, and complete the three-dimensional correlation condensation processing of the original point trace data in distance, azimuth, and pitch. At the same time, through the sliding window pipeline processing, the number of points involved in the condensation is continuously compressed level by level, which effectively reduces the data volume and complexity of the point trace condensation processing, reduces the processing delay, and improves the accuracy of the point trace condensation processing through the continuous statistics, extraction and optimized centroid estimation method of the condensed point trace feature information.

[0099] The second aspect of the present invention discloses a sliding window flow point trace aggregation processing system based on three-coordinate radar. Figure 6 FIG. 4 is a structural diagram of a sliding window flow point trace condensation processing system based on a three-coordinate radar according to an embodiment of the present invention; Figure 6 As shown, the system 100 includes:

[0100] The first processing module 101 is configured to remove abnormal echo traces: remove the trace data according to the N-fold beam width range and the sum-difference ratio in the original trace data;

[0101] The second processing module 102 is configured to: condense the point traces within the frame: adopt a two-level pipeline processing architecture, first classify and merge the point traces within the pulse group after the abnormal echo point traces are removed in the distance dimension, put the condensed points within the pulse group after the classification, merging and condensation processing into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation processing within the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation processing within the pulse group, and again classify and merge and condense all the data in the same beam in the cache in the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete one frame point trace condensation processing;

[0102] The third processing module 103 is configured to perform inter-frame point trace condensation and output: adopt a one-level pipeline processing architecture, directly put the first arriving beam point trace data into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, update the old point trace information in the buffer queue with the new target condensation point information, and directly add the point trace information of the newly added target to the buffer queue, after the condensation is completed, perform target end judgment on the condensation results in turn, directly output the target point trace that meets the end criterion as the target estimation value, and complete the inter-frame point trace condensation processing flow; after completing the inter-frame point trace condensation, judge the target point trace end criterion in turn according to the target's widening characteristics in azimuth, and after reaching the target point trace end criterion, perform pipeline output.

[0103] According to the system of the second aspect of the present invention, the first processing module 101 is configured to remove the point trace data according to the N-fold beam width range, specifically including:

[0104] Eliminate the point data outside the range of N times the beam width, where 1.2≤N≤1.5;

[0105] The specific method for removing the point trace data according to the sum-difference ratio in the original point trace data includes:

[0106] When the sum-difference ratio of the original points in the pulse group has both values ​​within the boundary conditions and values ​​outside the boundary conditions, the point data of the values ​​outside the boundary conditions are removed; when the sum-difference ratio is the value outside the boundary conditions, all the original points are retained.

[0107] In the system according to the second aspect of the present invention, the second processing module 102 is configured to distinguish the traces of different targets in the distance dimension based on the following criteria: target signal-to-noise ratio, distance, and Doppler.

[0108] According to the system of the second aspect of the present invention, the second processing module 102 is configured such that the classification and merging criteria include:

[0109] The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, the point data that does not match the signal-to-noise ratio is first filtered out, and then it is determined whether the distance unit difference of the point data is less than the distance threshold and whether the Doppler scale difference is less than the Doppler threshold.

[0110] According to the system of the second aspect of the present invention, the second processing module 102 is configured such that the classification and merging criteria also include: if the signal processing does not provide Doppler information, the change characteristics of the target amplitude are used to replace the Doppler information to distinguish target points that are adjacent or intersecting in distance.

[0111] According to the system of the second aspect of the present invention, the second processing module 102 is configured to optimize the traditional centroid method and then use amplitude weighting or power weighting to obtain the distance, azimuth and elevation of the condensed point traces.

[0112] According to the system of the second aspect of the present invention, the third processing module 103 is configured as follows: the merging criteria of the inter-frame point trace condensation include:

[0113] The same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold.

[0114] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the sliding window flow point trace condensation processing methods based on three-coordinate radar disclosed in the first aspect of the present invention are implemented.

[0115] Figure 7 is a structural diagram of an electronic device according to an embodiment of the present invention, such as Figure 7 As shown, the electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a button, a trackball or a touch pad set on the housing of the electronic device, or an external keyboard, touch pad or mouse, etc.

[0116] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a structural diagram of the part related to the technical solution of the present disclosure, and does not constitute a limitation on the electronic device to which the technical solution of the present application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0117] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the steps of the sliding window pipeline point trace condensation processing method based on three-coordinate radar in any one of the first aspects of the present invention are implemented.

[0118] Please note that the technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. The above embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several variations and improvements can be made, which all belong to the scope of protection of the present application. Therefore, the scope of protection of the patent in this application shall be based on the attached claims.

Claims

1. A sliding window flow point trace condensation processing method based on three-coordinate radar, characterized in that: The method comprises: Step S1, abnormal echo point trace removal: remove the point trace data according to the N-fold beam width range and the sum-difference ratio in the original point trace data; Step S2, intra-frame point trace condensation: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces in the pulse group after the abnormal echo point traces are removed in the distance dimension, put the condensed points in the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation process in the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation process, and again classify, merge and condense all the data in the same beam in the cache in the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete the intra-frame point trace condensation process once; Step S3, inter-frame point trace condensation and output: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data directly into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, use the new target condensation point information to update the old point trace information in the buffer queue, and the point trace information of the newly added target is directly added to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target widening characteristics in azimuth, and after the target point trace end criterion is reached, it is pipelined output.

2. According to the sliding window flow point trace condensation processing method based on three-coordinate radar according to claim 1, it is characterized in that: In step S1, the specific method of removing the point trace data according to the N-fold beam width range includes: Eliminate the point data outside the range of N times the beam width, where 1.2≤N≤1.5; The specific method of eliminating the dot data according to the sum-difference ratio in the original dot data includes: When the sum-difference ratio of the original points in the pulse group has both values ​​within the boundary conditions and values ​​outside the boundary conditions, the point data of the values ​​outside the boundary conditions are removed; when the sum-difference ratio is the value outside the boundary conditions, all the original points are retained.

3. The sliding window flow point trace condensation processing method based on three-coordinate radar according to claim 1 is characterized in that: In step S2, the basis for distinguishing the traces of different targets in the distance dimension includes: target signal-to-noise ratio, distance and Doppler.

4. The method for processing sliding window flow-type point trace condensation based on three-coordinate radar according to claim 3, characterized in that: In step S2, the classification and merging criteria include: The signal-to-noise ratio is greater than zero, the distance scale is continuous, and the Doppler scale is continuous. That is, when the signal processing gives the noise mean, the point data that does not match the signal-to-noise ratio is first filtered out, and then it is determined whether the distance unit difference of the point data is less than the distance threshold and whether the Doppler scale difference is less than the Doppler threshold.

5. The method for processing sliding window flow-type point trace condensation based on three-coordinate radar according to claim 4, characterized in that: In step S2, the classification and merging criteria also include: if the signal processing does not provide Doppler information, then using the target amplitude variation characteristics to replace the Doppler information to distinguish target points that are adjacent or intersecting in distance.

6. The method for processing sliding window flow point traces based on three-coordinate radar according to claim 1, characterized in that: In step S2, the method further includes: The traditional centroid method is optimized, and then amplitude weighting or power weighting is used to obtain the distance, azimuth and elevation of the condensed point traces.

7. The method for processing sliding window flow point traces based on three-coordinate radar according to claim 1, characterized in that: In step S3, the merging criteria for the inter-frame point trace condensation include: The same distance unit, azimuth scale continuity, elevation scale continuity and Doppler FD scale continuity, that is, whether the distance difference of the point trace data is less than 1 or 2 distance units, whether the azimuth difference is less than the azimuth threshold, whether the elevation difference is less than the elevation threshold, and whether the Doppler FD scale difference is less than the Doppler threshold.

8. A sliding window flow point trace condensation processing system based on three-coordinate radar, characterized in that: The system comprises: The first processing module is configured to remove abnormal echo traces: remove the trace data according to the N-fold beam width range and the sum-difference ratio in the original trace data; The second processing module is configured to: condense the point traces within the frame: adopt a two-level pipeline processing architecture, first classify, merge and condense the point traces within the pulse group after the abnormal echo point traces are removed, put the condensed points within the pulse group after classification, merging and condensation into the cache, and wait for the arrival of the point trace data in the next pulse group, complete the distance dimension point trace condensation processing within the pulse group, until all the pulse groups in the same beam complete the distance dimension point trace condensation processing within the pulse group, and again classify, merge and condense all the data in the same beam in the cache on the distance dimension between the pulse groups in the same beam to obtain the same beam condensation result, and complete one frame point trace condensation processing; The third processing module is configured to condense and output point traces between frames: adopt a one-level pipeline processing architecture, put the first arriving beam point trace data directly into the buffer queue as the first inter-beam condensation result, wait for the arrival of the next beam point trace data and classify and judge all the intermediate points in the buffer queue in turn, directly complete the condensation processing for the point trace data identified as the same target, update the old point trace information in the buffer queue with the new target condensation point information, and directly add the point trace information of the newly added target to the buffer queue. After the condensation is completed, the target end judgment is made on the condensation results in turn, and the target point trace that meets the end criterion is directly output as the target estimation value to complete the inter-frame point trace condensation processing flow; after the inter-frame point trace condensation is completed, the target point trace end criterion is judged in turn according to the target's widening characteristics in azimuth, and after the target point trace end criterion is reached, it is output in a pipeline manner.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps in a sliding window pipeline point trajectory condensation processing method based on a three-coordinate radar as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps in the sliding window pipeline point trajectory condensation processing method based on three-coordinate radar described in any one of claims 1 to 7 are implemented.