Radar multi-feature constant false alarm rate detection method and system, terminal and medium

Abstract

The invention relates to the technical field of radar signal processing, and particularly provides a radar multi-feature constant false alarm detection method and system, a terminal and a medium, and the method comprises the steps: obtaining a radar echo signal of a to-be-detected unit; extracting multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector; substituting the multi-dimensional feature vector into a preset marginal cumulative distribution function to obtain a probability integral transformation value of each dimension of feature, inputting the probability integral transformation values based on all dimensions into a pre-constructed Copula function, and calculating to obtain a boundary threshold so as to obtain a detection statistical magnitude of the to-be-detected unit; comparing the detection statistical magnitude of the unit to be detected with a preset judgment threshold; if the detection statistical magnitude is greater than or equal to the judgment threshold, determining that a target exists in the unit to be detected; and if the detection statistical magnitude is smaller than the judgment threshold value, determining that the to-be-detected unit does not have the target. According to the invention, stable and reliable detection performance can still be maintained in the face of an actual complex, non-Gaussian and non-stationary sea clutter background.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar signal processing, and particularly relates to a radar multi-feature constant false alarm detection method, system, terminal and medium. Background Art

[0002] In modern radar systems, target detection is a core link for realizing key functions such as maritime surveillance, navigation support and military defense. Especially in complex sea conditions, due to the interference of various uncertain factors such as sea surface waves and water body scattering, the signals received by the radar often contain a large amount of strong sea clutter, resulting in the target signal being masked. Therefore, in order to ensure the stable performance of the radar system in complex backgrounds, the constant false alarm rate (CFAR, Constant False Alarm Rate) detection technology is generally adopted in the academic community and engineering practice. This method sets a detection threshold to make the system maintain a stable false alarm rate under different environmental conditions, so as to realize the automatic recognition and alarm of weak targets on the premise of ensuring reliability.

[0003] Most of the existing CFAR detection methods are based on single-feature analysis. Common variants include mean value (CA-CFAR), ordered statistics (OS-CFAR), maximum value (GO-CFAR) and minimum value (SO-CFAR), etc. These methods usually assume that the background noise has a Gaussian or Weibull distribution and derive the detection threshold based on this. However, in actual sea conditions, sea clutter exhibits non-Gaussian and non-stationary characteristics, and there are often complex non-linear correlations between features, which makes the traditional CFAR model prone to problems such as high false alarm rate and decreased detection rate in high sea conditions. In addition, although some studies have tried to introduce multi-feature fusion to improve the detection performance, they often assume that the features are independent of each other, or fail to establish a clear mathematical relationship between each feature and the false alarm rate, resulting in a lack of interpretability and controllability in the detection process. Summary of the Invention

[0004] Aiming at the problem that most of the existing CFAR detection methods are based on single-feature analysis, and multi-feature detection is difficult to accurately implement due to complex feature correlations, the present invention provides a radar multi-feature constant false alarm detection method, system, terminal and medium to solve the above technical problems.

[0005] In a first aspect, the present invention provides a radar multi-feature constant false alarm detection method, including: Obtain the radar echo signal of the unit to be measured; Extract the multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector; Substitute the multi-dimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature. Based on the probability integral transformation values of all dimensions, input them into the pre-constructed Copula function, calculate the boundary threshold, and thus obtain the detection statistic of the unit under test; Compare the detection statistic of the unit under test with the preset decision threshold: If the detection statistic is greater than or equal to the decision threshold, the unit under test has a target; if the detection statistic is less than the decision threshold, the unit under test does not have a target.

[0006] Furthermore, the construction method of the preset marginal cumulative distribution function includes: Obtain the radar echo signal of the background unit, and extract the multi-dimensional features of the radar echo signal of the background unit ; Construct the corresponding marginal cumulative distribution function and the corresponding probability integral transformation value , where , represents the marginal cumulative distribution function of the i-th dimension feature, represents the cumulative probability within its corresponding distribution range. Among them, the marginal cumulative distribution function is obtained by statistical fitting of the features of each dimension in multiple background units. The fitting methods used include maximum likelihood estimation and least squares method, and the distribution models used include Beta distribution, lognormal distribution, and exponential distribution.

[0007] Furthermore, the method for pre-constructing the Copula function includes: Obtain the radar echo signal of the background unit, and extract the multi-dimensional features of the background unit ; Based on the probability integral transformation values of the background unit, use the maximum likelihood estimation method to determine the parameters of the Copula function, and obtain the Copula function: , where represents the Copula function.

[0008] Furthermore, substituting the multi-dimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature, and based on the probability integral transformation values of all dimensions, inputting them into the pre-constructed Copula function, and obtaining the detection statistic of the unit under test based on the Copula function includes: Substitute the multi-dimensional feature vector in the unit under test into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature ; The probability integral transformation value is input into a pre-constructed Copula function to obtain the joint cumulative distribution value of the Copula function. ; Calculate the threshold of each dimension feature. , and transform the features of each dimension based on the threshold of each dimension feature, and the statistic obtained is: .

[0009] Further, the method for calculating the threshold of each dimension feature vector is as follows: The marginal cumulative distribution function at the threshold is:

[0010] Expressed using the Copula function as:

[0011] For a given false alarm probability , define:

[0012] Then

[0013] wherein, represents the probability that an event occurs where one eigenvalue is greater than the threshold , which is the false alarm probability and represents the probability of misjudging the existence of a target.

[0014] Further, compare the detection statistic of the unit under test with the preset decision threshold: if the detection statistic is greater than or equal to the decision threshold, there is a target in the unit under test; if the detection statistic is less than the decision threshold, there is no target in the unit under test, including: If the statistic then it is determined that there is a target; If the statistic , then it is determined that there is no target.

[0015] In a second aspect, the present invention provides a radar multi-feature constant false alarm detection system, including: A signal acquisition module for acquiring the radar echo signal of the unit under test; A feature extraction module for extracting multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector; The statistic calculation module substitutes the multi-dimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature. Based on the probability integral transformation values of all dimensions, it is input into the pre-constructed Copula function to calculate the boundary threshold, thereby obtaining the detection statistic of the unit under test. The target judgment module is used to compare the detection statistic of the unit under test with the preset judgment threshold: if the detection statistic is greater than or equal to the judgment threshold, there is a target in the unit under test; if the detection statistic is less than the judgment threshold, there is no target in the unit under test.

[0016] In a third aspect, a terminal is provided, including: A processor and a memory, where The memory is used to store a computer program, The processor is used to call and run the computer program from the memory, so that the terminal executes the method of the above terminal.

[0017] In a fourth aspect, a computer storage medium is provided. Instructions are stored in the computer-readable storage medium. When it runs on a computer, the computer is made to execute the methods described in the above aspects.

[0018] The radar multi-feature constant false alarm detection method, system, terminal and medium provided by the present invention introduce a Copula function to establish a non-linear correlation modeling framework between features, breaking through the dependence limitations of traditional CFAR methods on feature independence or specific distribution models (such as Gaussian, Weibull), so that in the face of actual complex, non-Gaussian, non-stationary sea clutter background, stable and reliable detection performance can still be maintained.

[0019] Secondly, this application constructs a multi-feature detection model and transforms the target detection problem into a hypothesis testing problem under the Copula framework. Using the decomposition form of the marginal distribution and the Copula function, a closed analytical solution of the detection threshold is derived, forming a clear mathematical mapping relationship between the false alarm rate and the detection threshold, thereby realizing strict controllability of the false alarm rate. This precise threshold calculation mechanism significantly improves the consistency and robustness of the detection system under different sea conditions.

[0020] Thirdly, compared with existing methods such as CA-CFAR and OS-CFAR, this application shows better detection rate and lower false alarm rate in the verification of measured data. Especially in high sea state environments, the detection rate can be increased by 20.11% to 28.12%, and the false alarm rate is closer to the theoretical target, showing obvious practical application value and engineering feasibility, and fundamentally solving the key technical bottleneck that traditional CFAR methods are prone to failure in complex backgrounds.

[0021] In addition, the design principle of the present invention is reliable, the structure is simple, and it has a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0023] Figure 1 is a schematic flowchart of a method according to an embodiment of the present invention.

[0024] Figure 2 is a schematic block diagram of a system according to an embodiment of the present invention.

[0025] Figure 3 is a schematic structural diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

[0028] The radar multi-feature constant false alarm detection method provided by the embodiment of the present invention is executed by a computer device. Correspondingly, the radar multi-feature constant false alarm detection system runs in the computer device.

[0029] Figure 1 is a schematic flowchart of a method according to an embodiment of the present invention. Among them, Figure 1 The execution subject can be a radar multi-feature constant false alarm detection system. According to different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0030] For the convenience of understanding the present invention, the following further describes the radar multi-feature constant false alarm detection method provided by the present invention with reference to the principle of the radar multi-feature constant false alarm detection method of the present invention.

[0031] This application reformulates the radar target detection problem as a binary hypothesis test and introduces the Copula function to simulate the complex non-linear dependence relationships among multivariate features.

[0032] Within this framework, this application proposes the Copula-based constant false alarm rate (Copula-CFAR) theory and derives the mathematical relationship between the specified PFA (denoted as ) and the detection threshold under common Copula structures.

[0033] Radar target detection can be expressed as a binary hypothesis test problem. Let (denoted as ) represent the received signal of the ranging unit. Assuming that the energy of the noise is significantly weaker than that of the sea clutter, the noise part is ignored in this analysis. If a ranging unit contains a target, includes the target signal and the sea clutter; otherwise, only includes the sea clutter. Therefore, target detection in sea clutter can be formulated as the following binary hypothesis test: (1) where and represent the signals of the measured cell and the reference cell, respectively. Here, represents the target signal, while and represent the sea clutter signals of the measured cell and the reference cell, respectively. Under the null hypothesis condition , the measured cell has no target signal and only sea clutter signal, so it is classified as a sea clutter cell. Under the alternative hypothesis , the measured cell contains a target signal in addition to the sea clutter, so it is classified as a target cell. The reference cell signal is modeled as , and the number of reference cells is set to . In addition, we assume that the sea clutter in the reference cells is spatially uniform, which means that the statistical characteristics of the reference cells are the same as those of the measured cell.

[0034] Specifically, as Figure 1 shown, the radar multi-feature constant false alarm rate detection method includes: S1. Obtain the radar echo signal of the unit to be measured.

[0035] Specifically, the target detected in this application is a light buoy, 2.97 nautical miles away from the radar setting point, with an azimuth angle of 10.7°, and it is in an anchored floating state. The target is cylindrical, red in color, with a tank-shaped top mark. It is made of steel, with a floating body diameter of 2.4 meters and a height above the sea surface of about 4.1 meters. The data acquisition device uses a self-developed portable radar data acquisition device, namely HD-LD-CJ-22. Each channel uses 14-bit quantization, with a peak sampling rate of 120 MSPS and a continuous storage capacity of 80 MB / s. The dataset can be downloaded from the specified website: https: / / radars.ac.cn / web / data / getData?dataType=DatasetofRadarDetectingSea.

[0036] S2. Extract the multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector.

[0037] S3. Substitute the multi-dimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension feature. Based on the probability integral transformation values of all dimensions, input them into the pre-constructed Copula function, and obtain the detection statistic of the unit to be measured based on the Copula function.

[0038] The construction method of the preset marginal cumulative distribution function includes: Obtain the radar echo signal of the background unit and extract the multi-dimensional features of the radar echo signal of the background unit . Construct the corresponding marginal cumulative distribution function and the corresponding probability integral transformation value , where , represents the marginal cumulative distribution function of the i-th dimension feature, represents the cumulative probability within its corresponding distribution range. Among them, the marginal cumulative distribution function is obtained by statistically fitting the features of each dimension in multiple background units. The fitting methods used include maximum likelihood estimation and least squares method, and the distribution models used include Beta distribution, lognormal distribution, and exponential distribution.

[0039] The method for pre-constructing the Copula function includes: Obtain the radar echo signal of the background unit and extract the multi-dimensional features of the background unit ; Based on the probability integral transformation values of the background unit, use the maximum likelihood estimation method to determine the parameters of the Copula function and obtain the Copula function: , where Denote the Copula function.

[0040] Substitute the multi-dimensional feature vector in the unit under test into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature ; Input the probability integral transformation value into the pre-constructed Copula function to obtain the joint cumulative distribution value of the Copula function ; Calculate the threshold of each dimension of the feature , and perform transformation on the feature of each dimension based on the threshold of each dimension of the feature to obtain the statistic: .

[0041] The method for calculating the threshold of each dimension of the feature vector to be includes: The threshold when the marginal cumulative distribution function is:

[0042] Expressed using the Copula function as:

[0043] For a given false alarm probability , define:

[0044] Then

[0045] where represents the probability that an event occurs where one eigenvalue is greater than the threshold , which is the false alarm probability and represents the probability of misjudging the existence of a target.

[0046] S4. Compare the detection statistic of the unit under test with the preset decision threshold: If the detection statistic is greater than or equal to the decision threshold, there is a target in the unit under test; if the detection statistic is less than the decision threshold, there is no target in the unit under test.

[0047] If the statistic then it is judged that there is a target. If the statistic , then it is judged that there is no target.

[0048] Specifically, this application will give several examples to derive the analytical expression of the feature threshold corresponding to a given , and then derive the detection probability . It should be noted that these derivations are all based on the equivalent contribution hypothesis, that is, in the process of radar target detection, each feature has the same importance.

[0049] Example 1: Independent Copula with the same marginal distribution Suppose the features and follow the exponential distribution with parameters . Their CDFs at the thresholds of the feature are (2) where and are independent random variables, and their dependence structure is described by the independent Copula (3) According to the equal contribution hypothesis, let (4) For a given , Theorem 1 implies that (5) Therefore (6) We obtain (7) In addition, under the hypothesis, the feature still follows the exponential distribution, but the parameter is different. Then the survival function under (8) The detection probability is (9) Because (10) where is the Copula function under the (11) Combined with

[0050] Since it can be simplified to (12) Example 2: Clayton Copula with marginal distributions of the same distribution type but different parameters To emphasize the lower tail dependence of the features, we consider using the Clayton Copula. Its form is as follows (13) Assume the feature follows an exponential distribution with parameter , and the feature follows an exponential distribution with parameter . At the threshold of the feature , we have,,.

[0051] According to the equal - contribution hypothesis, we have (14) According to Theorem 1, the following relationship holds (15) Solving for (16) The detection threshold is (17) Since the joint CDF under is modeled by a Clayton Copula with parameter (18) , according to , the detection probability is (19) Example 3: Gumbel Copula with different marginal distributions For the upper - tail dependence relationship between features, the dependence structure is modeled by a Gumbel Copula, and its expression is (20) If follows an exponential distribution with parameter , while follows a uniform distribution , then under the equal - contribution hypothesis, the Copula structure becomes (21) According to Theorem 1, the expression of (22) The analytical expression of the detection threshold is (23) In addition, using the Gumbel Copula and equal - contribution , combined with the CDF is . Therefore, the detection probability is (24) In some embodiments, the radar multi-feature constant false alarm detection system may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the radar multi-feature constant false alarm detection system can be stored in the memory of the computer device and executed by at least one processor to perform (see Figure 1 description) the functions of radar multi-feature constant false alarm detection.

[0052] In this embodiment, according to the functions it performs, the radar multi-feature constant false alarm detection system can be divided into multiple functional modules, as Figure 2 shown. The functional modules of system 200 may include: a signal acquisition module 210, a feature extraction module 220, a statistic calculation module 230, and a target judgment module 240. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and are stored in the memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0053] The signal acquisition module is used to acquire the radar echo signal of the unit to be measured.

[0054] The feature extraction module is used to extract the multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector.

[0055] The statistic calculation module substitutes the multi-dimensional feature vector into a preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension feature. Based on the probability integral transformation values of all dimensions, it is input into a pre-constructed Copula function to calculate and obtain a boundary threshold, thereby obtaining the detection statistic of the unit to be measured; The target judgment module is used to compare the detection statistic of the unit to be measured with a preset judgment threshold: if the detection statistic is greater than or equal to the judgment threshold, there is a target in the unit to be measured. If the detection statistic is less than the judgment threshold, there is no target in the unit to be measured.

[0056] Figure 3 FIG. 300 is a schematic structural diagram of a terminal 300 provided by an embodiment of the present invention, and this terminal 300 can be used to execute the radar multi-feature constant false alarm detection method provided by an embodiment of the present invention.

[0057] Among them, the terminal 300 may include: a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art can understand that the structure of the server shown in the figure does not constitute a limitation on the present invention. It can be a bus structure, a star structure, and may also include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0058] Among them, the memory 320 can be used to store the execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. When the execution instructions in the memory 320 are executed by the processor 310, the terminal 300 is enabled to execute some or all of the steps in the above method embodiments.

[0059] The processor 310 is the control center of the storage terminal, connecting various parts of the entire electronic terminal using various interfaces and lines. By running or executing the software programs and / or modules stored in the memory 320, and by invoking the data stored in the memory, it executes various functions of the electronic terminal and / or processes data. The processor may be composed of an integrated circuit (IC). For example, it may be composed of a single packaged IC, or may be composed of multiple packaged ICs with the same or different functions connected together. For example, the processor 310 may only include a central processing unit (CPU). In the embodiments of the present invention, the CPU may be a single arithmetic core or may include multiple arithmetic cores.

[0060] The communication unit 330 is used to establish a communication channel so that the storage terminal can communicate with other terminals. It receives user data sent by other terminals or sends user data to other terminals.

[0061] The present invention also provides a computer storage medium. Among them, the computer storage medium can store a program, and when the program is executed, it may include some or all of the steps in the various embodiments provided by the present invention. The storage medium may be a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0062] Therefore, by introducing the Copula function, the present invention establishes a non-linear correlation modeling framework between features, breaking through the dependence limitations of traditional CFAR methods on feature independence or specific distribution models (such as Gaussian, Weibull), enabling stable and reliable detection performance in the face of actual complex, non-Gaussian, and non-stationary sea clutter backgrounds.

[0063] Secondly, the present application constructs a multi-feature detection model and transforms the target detection problem into a hypothesis testing problem under the Copula framework. By using the decomposition form of the marginal distribution and the Copula function, a closed-form analytical solution for the detection threshold is derived, forming a clear mathematical mapping relationship between the false alarm rate and the detection threshold, thereby achieving strict controllability of the false alarm rate. This precise threshold calculation mechanism significantly improves the consistency and robustness of the detection system under different sea conditions.

[0064] Thirdly, compared with existing methods such as CA-CFAR and OS-CFAR, the present application shows better detection rates and lower false alarm rates in the verification of measured data. Especially in high-sea conditions, the detection rate can be increased by 20.11% to 28.12%, and the false alarm rate is closer to the theoretical target, demonstrating obvious practical application value and engineering feasibility, and fundamentally solving the key technical bottleneck of the easy failure of traditional CFAR methods in complex backgrounds. The technical effects achievable in this embodiment can be referred to the description above and will not be elaborated here.

[0065] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution in the embodiments of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc., which can store program codes, including several instructions for causing a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0066] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the terminal embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the description in the method embodiments.

[0067] In several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the system or module can be in electrical, mechanical or other forms.

[0068] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0069] In addition, in each embodiment of the present invention, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0070] Although the present invention has been described in detail by referring to the drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions should be within the scope of the present invention. / Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A radar multi-feature constant false alarm detection method, characterized in that: include: Obtaining the radar echo signal of the unit under test; Extract the multi-dimensional features of the radar echo signal to obtain a multi-dimensional feature vector; Substitute the multidimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature. Based on the probability integral transformation value of all dimensions, input it into the pre-built Copula function to calculate the boundary threshold, thereby obtaining the detection statistic of the unit to be tested; The detection statistic of the unit to be tested is compared with a preset determination threshold: if the detection statistic is greater than or equal to the determination threshold, the target exists in the unit to be tested; if the detection statistic is less than the determination threshold, the target does not exist in the unit to be tested.

2. The method according to claim 1, characterized in that The marginal cumulative distribution function is constructed by: Obtain the radar echo signal of the background unit and extract the multi-dimensional features of the radar echo signal of the background unit ; Build The corresponding marginal cumulative distribution function and the corresponding probability integral transform ,in, , represents the marginal cumulative distribution function of the i-th dimension feature, represent The cumulative probability within its corresponding distribution range, wherein the marginal cumulative distribution function is obtained by statistically fitting the characteristics of each dimension in multiple background units, the fitting methods adopted include maximum likelihood estimation and least squares method, and the distribution models adopted include Beta distribution, lognormal distribution and exponential distribution.

3. The method according to claim 1, characterized in that: The methods for constructing Copula functions include: Obtain the radar echo signal of the background unit and extract the multi-dimensional features of the background unit ; Based on the probability integral transformation value of the background unit characteristics, the maximum likelihood estimation method is used to determine the parameters of the Copula function to obtain the Copula function: ,in, Represents a Copula function.

4. The method according to claim 3, characterized in that: Substitute the multidimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature. Based on the probability integral transformation value of all dimensions, input it into the pre-built Copula function to calculate the boundary threshold, thereby obtaining the detection statistics of the unit to be tested, including: Substitute the multidimensional feature vector in the unit to be tested into the fitted marginal cumulative distribution function to obtain the probability integral transformation value of each dimensional feature ; Input the probability integral transformation value into the pre-built Copula function to obtain the joint cumulative distribution value of the Copula function ; Calculate the threshold of each dimension feature , based on the threshold of each dimension feature, the feature of each dimension is transformed, and the statistic is obtained as follows: 。 5. The method according to claim 4, characterized in that Calculate the threshold of each dimension of feature vector as The methods include: Threshold The marginal cumulative distribution function at time is: Using Copula function it can be expressed as: For a given false alarm probability ,definition: but in, is the false alarm probability, indicating that there is a characteristic value Greater than threshold The probability of an event occurring.

6. The method according to claim 1, characterized in that Compare the detection statistic of the unit to be tested with a preset decision threshold: if the detection statistic is greater than or equal to the decision threshold, the target exists in the unit to be tested; If the detection statistic is less than the judgment threshold, there is no target in the unit under test, including: If the statistic Then it is judged that there is a target; If the statistic , then it is judged that there is no target.

7. A radar multi-feature constant false alarm detection system, characterized in that: include: A signal acquisition module, used to acquire the radar echo signal of the unit under test; A feature extraction module is used to extract multi-dimensional features of radar echo signals and obtain multi-dimensional feature vectors; The statistics calculation module substitutes the multi-dimensional feature vector into the preset marginal cumulative distribution function to obtain the probability integral transformation value of each dimension of the feature. The probability integral transformation value based on all dimensions is input into the pre-built Copula function to calculate the boundary threshold, thereby obtaining the detection statistic of the unit to be tested; The target judgment module is used to compare the detection statistic of the unit to be tested with a preset judgment threshold: if the detection statistic is greater than or equal to the judgment threshold, there is a target in the unit to be tested; if the detection statistic is less than the judgment threshold, there is no target in the unit to be tested.

8. A terminal, characterized in that: include: A memory, used for storing a radar multi-feature constant false alarm detection program; A processor is used to implement the steps of the radar multi-feature constant false alarm detection method according to any one of claims 1 to 6 when executing the radar multi-feature constant false alarm detection program.

9. A computer-readable storage medium storing a computer program, characterized in that: The readable storage medium stores a radar multi-feature constant false alarm detection program, and when the radar multi-feature constant false alarm detection program is executed by the processor, the steps of the radar multi-feature constant false alarm detection method according to any one of claims 1 to 6 are implemented.

Images