Method, system and equipment for identifying false track of detection radar during travel

Abstract

The invention provides a method, a system and equipment for identifying a false track of an inter-travel detection radar, and the method comprises the steps: obtaining track information based on the false track and a real track, and building a sample data set; calculating information entropy, association probability, course volatility, speed volatility, acceleration volatility and position range ratio of the false track and the real track according to the track information; training a machine learning classifier by taking the track information and the calculated information entropy, association probability, course volatility, speed volatility, acceleration volatility and position range ratio as identification features; and obtaining identification features of a track to be identified, performing classification through the trained classifier, and outputting identification results of a real track and a false track. According to the method, the real track and the false track are identified by using the SVM classifier, so that the false track is effectively inhibited, and the method is small in calculation amount, high in applicability and high in real-time performance and has an engineering application prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of on-the-go detection radar data processing, and in particular to a method, system and device for identifying false tracks of on-the-go detection radar. Background Art

[0002] Conventional fixed-station radars are set up on fixed platforms. When the working platform changes, the radar needs to be withdrawn, moved, set up and leveled before it can start working. It does not have the ability to detect intelligence information during the movement process and lacks the ability to respond to emergencies. Compared with fixed-station radars, moving detection radars can detect and track the enemy while moving, have better mobility and timeliness, and thus have higher battlefield survivability and combat effectiveness.

[0003] Radars that work on the move must be able to detect and track targets during the movement of the carrier platform. The movement of the platform will cause two main problems: position change and attitude change. Changes in the radar position and attitude lead to relative movement between the target and the radar, resulting in the coupling of the platform's movement and attitude disturbances into the target information observed by the radar. If the radar observation data is directly used to track the target trajectory, the tracking performance will be reduced or even track association and prediction errors will be made. In addition, due to the movement of the platform on the platform, the ground clutter, sea clutter, and cloud and rain clutter that enter from the radar main lobe or side lobe are constantly changing relative to the radar, resulting in the broadening of the clutter spectrum, the movement of the clutter relative to the radar's Doppler and distance, and it is impossible to filter it out from the signal processing level, which easily forms false tracks, resulting in a large number of false targets in the intelligence report, affecting the operator's judgment, reducing work efficiency, and in severe cases causing computer saturation and the radar failing to work properly. Summary of the invention

[0004] Aiming at the problem of many false tracks in modern radars, especially detection radars on the move, the present invention proposes a method, system and equipment for identifying false tracks of detection radars on the move. The false track identification method of the present invention has small calculation amount, strong applicability and high real-time performance, so as to ensure that the radar can adapt to the complex and changeable battlefield environment on the move.

[0005] In a first aspect, the present invention provides a method for identifying false tracks of a moving detection radar, the method comprising the following steps:

[0006] Obtain track information based on false tracks and real tracks, and establish a sample data set;

[0007] According to the track information, information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track are calculated;

[0008] The track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate are used as identification features for identifying false tracks and real tracks, and a machine learning classifier is trained;

[0009] The identification features of the track to be identified are obtained, and the trained classifier is used to classify it, and the identification results of the real track and the false track are output.

[0010] As a further improvement of the present invention, the step of obtaining track information based on false tracks and real tracks, and establishing a sample data set also includes obtaining position information based on the false tracks and real tracks through radar detection, generating track information based on the false tracks and real tracks through coordinate transformation, and establishing a sample data set.

[0011] As a further improvement of the present invention, the step of coordinate conversion specifically includes:

[0012] Extract the distance ρ based on the location information t , direction α t and pitch β t , after converting from the antenna array polar coordinate system to the antenna array rectangular coordinate system, the carrier platform rectangular coordinate system, and the radar station center coordinate system, the radar station center coordinate system is converted to the earth-centered earth-fixed coordinate system;

[0013] The process of converting the polar coordinate system of the antenna array surface to the rectangular coordinate system of the antenna array surface is as follows:

[0014] The antenna array coordinate system O-XYZ takes the center of the radar antenna as the origin O of the coordinate system. The X-axis is within the array and is the intersection of the local horizontal plane and the array. The Y-axis is the array normal, and the Z-axis is perpendicular to the array and upwards to form a right-handed coordinate system:

[0015]

[0016] The process of converting the rectangular coordinate system of the antenna array to the rectangular coordinate system of the carrier platform is:

[0017] The carrier platform rectangular coordinate system Oc-XcYcZc takes the center of the radar antenna as the origin Oc of the coordinate system, the Yc axis points to the direction of the carrier head, the Xc axis points to the right side of the carrier, and the Zc axis is perpendicular to the carrier baseline and upward to form a right-handed coordinate system:

[0018]

[0019] Among them, As and Es are the azimuth angle of the servo turntable and the preset tilt angle of the antenna respectively;

[0020] The process of converting from the carrier platform rectangular coordinate system to the radar station center coordinate system is:

[0021] The radar station center coordinate system Or-XrYrZr takes the center of the radar antenna as the origin of the coordinate system Or, the Zr axis coincides with the ellipsoid normal, upward is positive, the Yr axis coincides with the minor semi-axis of the ellipsoid, and the Zr axis coincides with the major semi-axis of the earth ellipsoid to form a right-handed coordinate system:

[0022]

[0023] Where ψ, θ, and γ are the yaw angle, pitch angle, and roll angle of the radar installation plane, respectively;

[0024] The process of converting the radar station center coordinate system to the earth-centered earth-fixed coordinate system is:

[0025] The radar station center coordinate system Og-XgYgZg takes the center of the earth as the coordinate origin Og, the Zg axis is parallel to the earth's axis and points to the North Pole, the Xg axis points to the intersection of the prime meridian and the equator, and the Yg axis is perpendicular to the XgOgZg plane to form a right-handed coordinate system:

[0026]

[0027] In the formula, B r , L r , H r are the latitude, longitude and altitude of the radar installation plane respectively; N is the radius of curvature of the reference ellipsoid, e is the eccentricity of the ellipsoid, a is the major radius of the reference ellipsoid, and b is the polar radius.

[0028] As a further improvement of the present invention, in the step of calculating the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position range rate of the false track and the real track, the information entropy f is calculated by the following formulas respectively: H (k), association probability f M (k) Heading fluctuation V φ (k) Speed ​​fluctuation V v (k) Acceleration fluctuation V a (k) and position range rate δ P (k):

[0029]

[0030] Among them, information entropy generates the radar track state vector at time k based on the IMM algorithm: The matching probability μ under model j j (k) is calculated, where M is the number of motion models in the IMM algorithm; l is the model number with the highest probability of matching the track motion in the IMM algorithm; △ l (k) is the filter residual of model l; S l (k) is the new information covariance; K is the number of sliding window periods; φ, V φare the heading angle and its variance respectively; v, V v are the track speed and its variance respectively; A, V a are the track acceleration and its variance respectively; δ x (k),δ y (k),δ z (k) are the position extreme error rates of the three axes in the Earth-centered Earth-fixed coordinate system, T is the sampling period; x(i), y(i), z(i) are the position information of the three axes of the track in the Og-XgYgZg coordinate system at time i.

[0031] As a further improvement of the present invention, the step of training the machine learning classifier further includes labeling the sample data set to obtain a training data set D = {(x 1 ,y 1 ) , (x 2 ,y 2 ),…,(x N ,y N )}, where N is the number of training samples, The feature vector representing the input track, x i =(f H (k), f M (k), V φ (k), V v (k), V a (k),δ P (k)) T ,y i ∈{-1,+1} is the category label of the track, +1 represents the real track, and -1 represents the false track. The machine learning classifier is trained through the training data set.

[0032] As a further improvement of the present invention, the classifier also includes constructing a convex quadratic programming model and solving it to obtain the optimal separation hyperplane, specifically:

[0033] Construct a convex quadratic programming model:

[0034]

[0035] After solving, we get: * =(λ 1 * ,λ 2 * ,…,λ M * ,0,…,0) T , select λ in the open interval (0, C) j * Quantity, calculate:

[0036]

[0037] Solve to construct the optimal separating hyperplane:

[0038]

[0039] In the formula, λ i is the Lagrange multiplier; K(·,·) is the kernel function; sign represents symbolic operation, and M is the number of support vectors.

[0040] As a further improvement of the present invention, the step of training the machine learning classifier also includes selecting the SVM algorithm, ELM algorithm or RKELM algorithm in machine learning as the machine learning classifier.

[0041] In a second aspect, the present invention further proposes a system for identifying false tracks of a moving detection radar, the system comprising:

[0042] An acquisition module is used to acquire track information based on false tracks and real tracks and establish a sample data set;

[0043] A calculation module, used to calculate the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track according to the track information;

[0044] A classifier module, used to train a machine learning classifier by using the track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate as identification features for identifying false tracks and real tracks;

[0045] The output module is used to obtain the identification features of the track to be identified, classify it through the trained classifier, and output the identification results of the real track and the false track.

[0046] In a third aspect, the present invention proposes an electronic device, comprising a processor and a memory storing a computer program, wherein the processor, when executing the computer program, implements the steps of the method for identifying false tracks of a moving detection radar as described in the first aspect.

[0047] In a fourth aspect, the present invention proposes a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for identifying false tracks of a moving detection radar as described in the first aspect.

[0048] The present invention utilizes the real-time attitude information and position information of the carrier platform provided by the inertial navigation device to dynamically compensate the radar detection target position, thereby ensuring the stable tracking of the detection radar while moving and the accurate extraction of track characteristic parameters; at the same time, by constructing a real track and a false track training set and extracting track characteristics, the SVM classifier is used to identify the real track and the false track, thereby achieving effective suppression of the false track. The method has small calculation amount, strong applicability, high real-time performance, and has engineering application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of the method for identifying false tracks of a moving detection radar proposed by the present invention.

[0050] Figure 2 This is a framework diagram of the false track identification system of the moving detection radar proposed by the present invention.

[0051] Figure 3 Schematic diagram of an electronic device. DETAILED DESCRIPTION

[0052] For ease of understanding, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying 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 creative work are within the scope of protection of the present invention.

[0053] The terms used herein are only used for the purpose of describing specific example embodiments and are not restrictive. For example, unless the context clearly indicates otherwise, as used herein, the singular forms "a", "an" and "the" may also include plural forms. When used in this specification, the terms "include", "comprise" and / or "contain" mean that the associated integers, steps, operations, elements and / or components exist, but do not exclude the existence of one or more other features, integers, steps, operations, elements, components and / or groups or that other features, integers, steps, operations, elements, components and / or groups may be added in the system / method.

[0054] In view of the following description, these and other features of the present specification, as well as the operation and function of the related elements of the structure, and the economy of the combination and manufacture of the parts can be significantly improved. Reference is made to the accompanying drawings, all of which form a part of this specification. However, it should be clearly understood that the drawings are for illustration and description purposes only and are not intended to limit the scope of this specification. It should also be understood that the drawings are not drawn to scale.

[0055] The flowcharts used in this specification illustrate the operations implemented by the system according to some embodiments in this specification. It should be clearly understood that the operations of the flowcharts may not be implemented in sequence. On the contrary, the operations may be implemented in reverse order or simultaneously. In addition, one or more other operations may be added to the flowchart. One or more operations may be removed from the flowchart.

[0056] Figure 1 The specific process of the method for identifying false tracks of the moving detection radar is illustrated, such as Figure 1 As shown, the method for identifying the false track of the moving detection radar includes the following steps:

[0057] Step S101: Acquire track information based on the false track and the real track, and establish a sample data set.

[0058] Step S102. Calculate the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track according to the track information.

[0059] Step S103. Use the track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate as identification features for identifying false tracks and real tracks, and train a machine learning classifier.

[0060] Step S104: Obtain identification features of the track to be identified, classify it using the trained classifier, and output the identification results of the real track and the false track.

[0061] In this embodiment, the process of obtaining the track information based on the real track and the false track in step S101 is actually a data processing process of the target and clutter detected by the radar. After obtaining the position information of the target and clutter detected by the radar, coordinate conversion is performed to generate the track information based on the false track and the real track, and establish a sample data set. The specific process of coordinate conversion is as follows:

[0062] Extract the distance ρ based on the location information t , direction α t and pitch β t , after converting from the antenna array polar coordinate system to the antenna array rectangular coordinate system, the carrier platform rectangular coordinate system, and the radar station center coordinate system, the radar station center coordinate system is converted to the earth-centered earth-fixed coordinate system;

[0063] The process of converting the polar coordinate system of the antenna array surface to the rectangular coordinate system of the antenna array surface is as follows:

[0064] The antenna array coordinate system O-XYZ takes the center of the radar antenna as the origin O of the coordinate system. The X-axis is within the array and is the intersection of the local horizontal plane and the array. The Y-axis is the array normal, and the Z-axis is perpendicular to the array and upwards to form a right-handed coordinate system:

[0065]

[0066] The process of converting the rectangular coordinate system of the antenna array to the rectangular coordinate system of the carrier platform is:

[0067] The carrier platform rectangular coordinate system Oc-XcYcZc takes the center of the radar antenna as the origin Oc of the coordinate system, the Yc axis points to the direction of the carrier head, the Xc axis points to the right side of the carrier, and the Zc axis is perpendicular to the carrier baseline and upward to form a right-handed coordinate system:

[0068]

[0069] Among them, As and Es are the azimuth angle of the servo turntable and the preset tilt angle of the antenna respectively;

[0070] The process of converting the carrier platform rectangular coordinate system to the radar station center coordinate system (ENU coordinate system) is as follows:

[0071] The radar station center coordinate system Or-XrYrZr takes the center of the radar antenna as the origin of the coordinate system Or, the Zr axis coincides with the ellipsoid normal, upward is positive (sky direction), the Yr axis coincides with the minor semi-axis of the ellipsoid (north direction), and the Zr axis coincides with the major semi-axis of the earth ellipsoid (east direction) to form a right-handed coordinate system:

[0072]

[0073] Where ψ, θ, and γ are the yaw angle, pitch angle, and roll angle of the radar installation plane, respectively, which are provided in real time by the inertial navigation device.

[0074] The process of converting from the radar station center coordinate system (ENU coordinate system) to the earth-centered earth-fixed coordinate system (ECEF coordinate system) is:

[0075] The radar station center coordinate system (ECEF coordinate system) Og-XgYgZg takes the center of the earth as the coordinate origin Og, the Zg axis is parallel to the earth's axis and points to the North Pole, the Xg axis points to the intersection of the prime meridian and the equator, and the Yg axis is perpendicular to the XgOgZg plane (that is, the intersection of 90 degrees east longitude and the equator) to form a right-handed coordinate system:

[0076]

[0077] In the formula, B r , L r , H rare the latitude, longitude and altitude of the radar installation plane respectively; N is the radius of curvature of the reference ellipsoid, e is the eccentricity of the ellipsoid, a is the major radius of the reference ellipsoid, and b is the polar radius.

[0078] It should be noted that in this embodiment, the nearest neighbor association, probability data association, joint probability data association and other algorithms can also be used to establish the membership relationship between the echo point track and the track to achieve point-to-track association. At the same time, the track filtering based on multiple models (IMM) describes the possible motion state of the target during the radar operation process through multiple motion models, and finally estimates the motion state through effective weighted fusion, which can well overcome the problem of large single model estimation error. The motion model can be a combination of CV, CA, CT, Singer, CS, Jerk, etc.

[0079] In this embodiment, in step S102, the information entropy f is calculated by the following formulas: H (k), association probability f M (k) Heading fluctuation V φ (k) Speed ​​fluctuation V v (k) Acceleration fluctuation V a (k) and position range rate δ P (k):

[0080]

[0081] Among them, information entropy generates the radar track state vector at time k based on the IMM algorithm: The matching probability μ under model j j (k) is calculated, where M is the number of motion models in the IMM algorithm; l is the model number with the highest probability of matching the track motion in the IMM algorithm; △ l (k) is the filter residual of model l; S l (k) is the new information covariance; K is the number of sliding window periods; φ, V φ are the heading angle and its variance respectively; v, V v are the track speed and its variance respectively; A, V a are the track acceleration and its variance respectively; δ x (k),δ y (k),δ z (k) are the position extreme error rates of the three axes in the Earth-centered Earth-fixed coordinate system, T is the sampling period; x(i), y(i), z(i) are the position information of the three axes of the track in the Og-XgYgZg coordinate system at time i.

[0082] It should be noted that information entropy is a measure of system uncertainty. The more stable the system, the smaller the information entropy. For clutter false tracks, their motion state is chaotic, and the degree of matching with each model is relatively low, so the information entropy is large. Track association probability is used to evaluate the probability of association between point tracks and tracks. False tracks usually do not conform to conventional motion models, so △ l (k) is larger, f M (k) is small. The heading angle of the clutter false track fluctuates greatly, V φ (k) is relatively large. The clutter false track speed fluctuates greatly, V φ (k) is relatively large. The acceleration fluctuation of the clutter false track is relatively large, V a (k) is relatively large. Fixed ground clutter and slow meteorological clutter have small changes in the three-axis position over a long period of time, and δ P (k) is smaller.

[0083] In step S103, the step of training the machine learning classifier also includes selecting the SVM algorithm, ELM algorithm or RKELM algorithm in machine learning as the machine learning classifier. In this embodiment, the SVM algorithm is used for illustration, and the sample data set is marked to obtain the training data set D = {(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x N ,y N )}, where N is the number of training samples, Feature vector representing the input track.

[0084] Among them, x i =(f H (k), f M (k), V φ (k), V v (k), V a (k),δ P (k)) T ,y i ∈{-1,+1} is the category label of the track, +1 represents the real track, and -1 represents the false track. The machine learning classifier is trained through the training data set.

[0085] Construct and solve a convex quadratic programming problem:

[0086]

[0087] The optimal solution can be obtained through the SMO (Sequential Minimal Optimization) algorithm (not general, assuming that the first M samples constitute the support vector):

[0088] λ* =(λ 1 * ,λ 2 * ,…,λ M * ,0,…,0) T

[0089]

[0090] Thus we get the optimal separating hyperplane:

[0091]

[0092] Where sign represents symbolic operation, b * represents the classification threshold, and M is the number of support vectors. It should be noted that the SVM algorithm, ELM algorithm or RKELM algorithm can be selected as the machine learning classifier, and the SVM algorithm is used as an example in this embodiment.

[0093] In step S104, when the radar is in normal working condition, the identification features of the track to be identified are obtained, the features of the track are extracted, the target is identified based on the currently trained SVM classifier (as shown in the optimal separating hyperplane formula), the trained classifier is used for classification, and the identification results of the real track and the false track are output, the false track is eliminated, and the real target track is retained as the final output result.

[0094] The technical solution of the present invention mainly includes two aspects:

[0095] First: Eliminate the impact of carrier platform movement and attitude changes on radar detection information

[0096] Through the real-time radar attitude, position and speed information provided by the "integrated navigation equipment", the carrier attitude and position compensation is realized, and the disturbance of the carrier platform in the radar detection information is eliminated. In order to reduce the impact of the carrier attitude and position changes on the tracking stability, it is necessary to convert the target information detected by the radar at different times into the same fixed coordinate system before tracking the target. The Earth-centered Earth-fixed coordinate system (ECEF) is a coordinate system based on the earth, which is a fixed coordinate system. For this reason, the present invention selects the earth-centered earth-fixed coordinate system as the tracking coordinate system. First, the radar antenna receives the target echo to complete signal processing and detection, and obtains the position information of the target under the antenna array coordinate system (Oa-XaYaZa), and converts it to the carrier coordinate system (Oc-XcYcZc) according to the radar servo angle and the preset tilt angle of the array; then, the attitude data (yaw angle, pitch angle, roll angle) of the radar is obtained in real time according to the inertial navigation device installed on the radar plane, and it is converted to the station center coordinate system (ENU coordinate system) (Or-XrYrZr) according to the attitude data, and then it is converted to the earth-centered earth-fixed coordinate system (ECEF coordinate system) (Og-XgYgZg) according to the longitude and latitude of the inertial group. Through the above coordinate conversion, the influence of platform motion on the target position information is eliminated from the target position information, so that the target information only contains its own motion characteristics before track tracking, which provides the prerequisite for subsequent stable tracking and false track identification.

[0097] Second: Identify clutter and false tracks

[0098] The radar detection environment is harsh, with a large amount of strong ground clutter. Especially as the radar carrier moves, the clutter spectrum will be broadened. The existing clutter suppression technology is still insufficient in dealing with strong clutter environments such as ground objects and weather. It cannot filter out clutter well, and there is still a large amount of residual clutter. The residual clutter will form a stable tracking track when it enters the track processing stage. The false track of the radar is usually related to some disordered clutter points. Its motion trajectory is mostly random, with large changes in speed and heading, which usually does not conform to the classical physics motion model. The present invention starts from the perspective of studying the clutter characteristics in a strong clutter environment, and proposes a method to eliminate the influence of the carrier platform movement and the change of attitude at the track processing level, and then statistically analyze the differences between the clutter track and the real target track in track information entropy, association probability, heading angle, heading speed, heading acceleration and other characteristics, build a training sample library and train an SVM classifier, and then use the SVM classifier to identify whether the target is a clutter false target, thereby eliminating the false track, and verify the feasibility of the method in combination with the measured data. The method is applicable to moving detection radars and also to fixed station radars.

[0099] Figure 2The framework diagram of the false track identification system of the moving detection radar is shown as follows: Figure 2 As shown, the system 200 includes:

[0100] An acquisition module 201 is used to acquire track information based on a false track and a real track and establish a sample data set;

[0101] A calculation module 202 is used to calculate the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track according to the track information;

[0102] A classifier module 203 is used to train a machine learning classifier by using the track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate as identification features for identifying false tracks and real tracks;

[0103] The output module 204 is used to obtain the identification features of the track to be identified, classify it through the trained classifier, and output the identification results of the real track and the false track.

[0104] Figure 3 An example of a physical structure diagram of an electronic device is shown. Figure 3 As shown, the electronic device 300 may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call a computer program in the memory 330 to perform the steps of the method for identifying false tracks of a moving detection radar.

[0105] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0106] On the other hand, an embodiment of the present application also provides a processor-readable storage medium, which stores a computer program, and the computer program is used to enable the processor to execute the steps of the method for identifying false tracks of the moving detection radar provided in the above embodiments.

[0107] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state drive (SSD)), etc.

[0108] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying false tracks of a moving detection radar, characterized in that: The method comprises the following steps: Obtain track information based on false tracks and real tracks, and establish a sample data set; According to the track information, information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track are calculated; The track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate are used as identification features for identifying false tracks and real tracks, and a machine learning classifier is trained; The identification features of the track to be identified are obtained, and the trained classifier is used to classify it, and the identification results of the real track and the false track are output.

2. The method for identifying false tracks of a moving detection radar according to claim 1, characterized in that: The step of obtaining track information based on the false track and the real track and establishing a sample data set also includes obtaining position information based on the false track and the real track through radar detection, generating track information based on the false track and the real track through coordinate conversion, and establishing a sample data set.

3. The method for identifying false tracks of a moving detection radar according to claim 2, characterized in that: The steps of coordinate transformation specifically include: Extract the distance ρ based on the location information t , direction α t and pitch β t , after converting from the antenna array polar coordinate system to the antenna array rectangular coordinate system, the carrier platform rectangular coordinate system, and the radar station center coordinate system, the radar station center coordinate system is converted to the earth-centered earth-fixed coordinate system; The process of converting the polar coordinate system of the antenna array surface to the rectangular coordinate system of the antenna array surface is as follows: The antenna array coordinate system O-XYZ takes the center of the radar antenna as the origin O of the coordinate system. The X-axis is within the array and is the intersection of the local horizontal plane and the array. The Y-axis is the array normal, and the Z-axis is perpendicular to the array and upwards to form a right-handed coordinate system: The process of converting the rectangular coordinate system of the antenna array to the rectangular coordinate system of the carrier platform is: The carrier platform rectangular coordinate system Oc-XcYcZc takes the center of the radar antenna as the origin Oc of the coordinate system, the Yc axis points to the direction of the carrier head, the Xc axis points to the right side of the carrier, and the Zc axis is perpendicular to the carrier baseline and upward to form a right-handed coordinate system: Among them, As and Es are the azimuth angle of the servo turntable and the preset tilt angle of the antenna respectively; The process of converting from the carrier platform rectangular coordinate system to the radar station center coordinate system is: The radar station center coordinate system Or-XrYrZr takes the center of the radar antenna as the origin of the coordinate system Or, the Zr axis coincides with the ellipsoid normal, upward is positive, the Yr axis coincides with the minor semi-axis of the ellipsoid, and the Zr axis coincides with the major semi-axis of the earth ellipsoid to form a right-handed coordinate system: Where ψ, θ, and γ are the yaw angle, pitch angle, and roll angle of the radar installation plane, respectively; The process of converting the radar station center coordinate system to the earth-centered earth-fixed coordinate system is: The radar station center coordinate system Og-XgYgZg takes the center of the earth as the coordinate origin Og, the Zg axis is parallel to the earth's axis and points to the North Pole, the Xg axis points to the intersection of the prime meridian and the equator, and the Yg axis is perpendicular to the XgOgZg plane to form a right-handed coordinate system: In the formula, B r , L r , H r are the latitude, longitude and altitude of the radar installation plane respectively; N is the radius of curvature of the reference ellipsoid, e is the eccentricity of the ellipsoid, a is the major radius of the reference ellipsoid, and b is the polar radius.

4. The method for identifying false tracks of a moving detection radar according to claim 3, characterized in that: In the steps of calculating the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position range rate of the false track and the real track, the information entropy f is calculated by the following formulas respectively: H (k), association probability f M (k) Heading fluctuation V φ (k) Speed ​​fluctuation V v (k) Acceleration fluctuation V a (k) and position range rate δ P (k): Among them, information entropy generates the radar track state vector at time k based on the IMM algorithm: The matching probability μ under model j j (k) is calculated, where M is the number of motion models in the IMM algorithm; l is the model number with the highest probability of matching the track motion in the IMM algorithm; △ l (k) is the filter residual of model l; S l (k) is the new information covariance; K is the number of sliding window periods; φ, V φ are the heading angle and its variance respectively; v, V v are the track speed and its variance respectively; A, V a are the track acceleration and its variance respectively; δ x (k),δ y (k),δ z (k) are the position extreme error rates of the three axes in the Earth-centered Earth-fixed coordinate system, T is the sampling period; x(i), y(i), z(i) are the position information of the three axes of the track in the Og-XgYgZg coordinate system at time i.

5. The method for identifying false tracks of a moving detection radar according to claim 4, characterized in that: The step of training the machine learning classifier also includes labeling the sample data set to obtain a training data set D = {(x1, y1), (x2, y2), ..., (x N ,y N )}, where N is the number of training samples, The feature vector representing the input track, x i =(f H (k), f M (k), V φ (k), V v (k), V a (k),δ P (k)) T ,y i ∈{-1,+1} is the category label of the track, +1 represents the real track, and -1 represents the false track. The machine learning classifier is trained through the training data set.

6. The method for identifying false tracks of a moving detection radar according to claim 5, characterized in that: The classifier also includes constructing a convex quadratic programming model and solving it to obtain the optimal separation hyperplane, specifically: Construct a convex quadratic programming model: After solving, we get: * =(λ1 * ,λ2 * ,…,λ M * ,0,…,0) T , select λ in the open interval (0, C) j * Quantity, calculate: Solve to construct the optimal separating hyperplane: In the formula, λ i is the Lagrange multiplier; K(·,·) is the kernel function; sign represents symbolic operation, and M is the number of support vectors.

7. The method for identifying false tracks of a moving detection radar according to claim 1, characterized in that: The step of training the machine learning classifier also includes selecting an SVM algorithm, an ELM algorithm or an RKELM algorithm in machine learning as the machine learning classifier.

8. The identification system of the false track identification method of the moving detection radar according to any one of claims 1 to 7, characterized in that: The system comprises: An acquisition module is used to acquire track information based on false tracks and real tracks and establish a sample data set; A calculation module, used to calculate the information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate of the false track and the real track according to the track information; A classifier module, used to train a machine learning classifier by using the track information and the calculated information entropy, association probability, heading volatility, speed volatility, acceleration volatility and position extreme rate as identification features for identifying false tracks and real tracks; The output module is used to obtain the identification features of the track to be identified, classify it through the trained classifier, and output the identification results of the real track and the false track.

9. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the method for identifying false tracks of the moving detection radar as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying false tracks of a moving detection radar as described in any one of claims 1 to 7 are implemented.

Images