Radar jamming signal design and optimization method based on relative entropy criterion

Abstract

The present application relates to a radar jamming signal design and optimization method based on the relative entropy criterion. The method comprises: expressing the jammed radar target detection problem as a binary hypothesis testing problem, wherein under the first hypothesis, the electromagnetic signal received by the radar only contains the jamming signal and the system noise, and under the second hypothesis, the electromagnetic signal received by the radar contains the target echo signal, the jamming signal, and the system noise; establishing likelihood functions under the two hypotheses respectively and calculating the expression of the relative entropy between the two likelihood functions, taking radio frequency noise interference as the decision variable and relative entropy as the objective function, establishing an interference signal optimization model based on relative entropy minimization under the constraint of radio frequency noise power; optimizing the interference signal optimization model using the upper bound minimization method, and convolving the optimized radio frequency noise interference with the radar transmission signal to obtain the optimized interference signal. The use of this method can improve the radar signal jamming effect.

Description

Technical Field

[0001] The present application relates to the technical field of radar jamming signal design and optimization, and in particular to a radar jamming signal design and optimization method based on the relative entropy criterion. Background Art

[0002] Active jamming is an effective strategy for avoiding detection by enemy radar systems. However, the rapid advancement of radar anti-jamming technology has made the jamming task increasingly difficult. For example, waveform agility and adaptive filtering techniques can significantly suppress jamming energy. As two sides of the same coin, jamming and radar anti-jamming technology advance hand in hand. Jammers equipped with digital radio frequency memory (DRFM) can intercept radar signals and intelligently make jamming decisions. Radar jamming and anti-jamming have been hot topics in the field of electronic warfare for many years.

[0003] Radar suppression jamming mainly takes the form of RF noise. This involves injecting a certain amount of noise into the radar receiver, masking the target in the noise and preventing it from being correctly detected by the radar. The effectiveness of the suppression jamming is closely related to the power spectrum of the RF noise. Carefully designed RF noise can achieve better jamming effects at the same power. According to currently available literature and patents, the design of RF noise is mainly based on the interference-to-signal ratio (JSR) criterion and the mutual information (MI) criterion. The JSR criterion designs the power spectrum of RF noise to maximize the ratio of the jamming signal power to the target signal at the radar receiver; the MI criterion designs the power spectrum of RF noise to minimize the mutual information between the radar received signal and the target response. The interference effect of both is low. Summary of the Invention

[0004] Based on this, it is necessary to provide a radar jamming signal design and optimization method based on the relative entropy criterion that can improve the radar signal jamming effect in response to the above technical problems.

[0005] A radar jamming signal design and optimization method based on relative entropy criterion, the method comprising:

[0006] Modeling the electromagnetic signals in the radar jamming scenario, modeling the electromagnetic signals in the scenario into radar transmission signals, target echo signals and jamming signals;

[0007] The radar target detection problem under interference is formulated as a binary hypothesis testing problem. Under the first hypothesis, the electromagnetic signal received by the radar contains only the interference signal and system noise. Under the second hypothesis, the electromagnetic signal received by the radar contains the target echo signal, the interference signal, and system noise. The likelihood functions under the two hypotheses are established and the expression of the relative entropy between the two likelihood functions is calculated, resulting in a functional relationship between the relative entropy and the RF noise interference.

[0008] Taking radio frequency noise interference as the decision variable and relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the constraint of radio frequency noise power. The interference signal optimization model is optimized using the upper bound minimization method to obtain the optimized radio frequency noise interference.

[0009] The optimized radio frequency noise interference is convolved with the radar transmission signal to obtain the optimized interference signal.

[0010] The above-mentioned radar jammer signal design and optimization method based on the relative entropy criterion is proposed in this application. By modeling the radar target detection problem as a binary hypothesis testing problem, the characteristics of the jammer and target signals under different hypotheses are described separately. Likelihood functions are established under each hypothesis, and the expression for the relative entropy between the two likelihood functions is calculated. Based on relative entropy in information theory, the degree of distinguishability of the radar received signal under the two hypotheses is measured. The smaller the relative entropy, the more difficult it is for the radar to distinguish between the two hypotheses, thereby reducing the reliability of target detection. By minimizing the relative entropy between the two hypotheses, the radar becomes less able to distinguish between the two hypotheses when detecting targets. Compared to traditional interference-to-signal ratio (JSR) and mutual information (MI) criteria, the relative entropy criterion directly takes the hypothesis testing perspective and is more compatible with radar detection problems. RF noise interference is used as the optimization decision variable, and a power constraint is introduced to ensure that the jammer achieves optimal performance within the power budget. Mathematical optimization techniques are used to solve the objective function, making the jammer signal design more theoretically feasible and practical. The optimized RF noise interference is convolved with the radar transmit signal to generate the final jammer signal. The convolution operation makes the generated interference signal more consistent with the characteristics of the radar signal, improving the effect of the interference signal in actual scenarios. At the same time, the optimized interference signal acts more accurately on the radar's signal processing link, significantly reducing the radar's ability to detect targets in noise. Compared with traditional methods based on JSR or MI, the interference signal designed based on the relative entropy criterion is more targeted. Currently, there is no method for designing interference signals based on the relative entropy criterion in the literature and patents. This application fills the gap in the current lack of designing interference signals based on the relative entropy criterion, and the interference signal obtained according to the proposed design method can effectively interfere with the radar detector, improve the radar signal interference effect, and significantly reduce the probability of radar detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 1 is a flow chart of a radar jamming signal design and optimization method based on the relative entropy criterion in one embodiment;

[0012] Figure 2 This is a schematic block diagram of the MM iterative optimization process in one embodiment;

[0013] Figure 3 FIG. 1 is a schematic diagram of interference performance evaluation in one embodiment. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0015] In one embodiment, Figure 1 As shown, a radar jamming signal design and optimization method based on relative entropy criterion is provided, comprising the following steps:

[0016] Step 102 : Modeling the electromagnetic signal in the radar jamming scenario, where the electromagnetic signal in the scenario is modeled as a radar transmission signal, a target echo signal, and an interference signal.

[0017] The electromagnetic signal in the radar jamming scenario is modeled as three main parts: the radar transmission signal s, the target echo signal r, and the interference signal i. The target echo signal r is represented as the convolution of the radar transmission signal s and the target response t, and the interference signal i is represented as the convolution of the radar transmission signal s and the radio frequency noise interference j. Gaussian noise n is used to simulate the radar system noise. At this point, the electromagnetic signal received in the radar jamming scenario can be represented as a combination of the target echo r, the interference signal i, and the system noise n.

[0018] Step 104: The radar target detection problem under interference is expressed as a binary hypothesis testing problem. Under the first hypothesis, the electromagnetic signal received by the radar contains only the interference signal and the system noise. Under the second hypothesis, the electromagnetic signal received by the radar contains the target echo signal, the interference signal, and the system noise. Likelihood functions are established under the two hypotheses respectively, and the expression of the relative entropy between the two likelihood functions is calculated to obtain the functional relationship between the relative entropy and the radio frequency noise interference.

[0019] The radar target detection problem under interference is formulated as a binary hypothesis testing problem. Under the H0 hypothesis, the electromagnetic signal received by the radar only contains the interference signal i and the system noise n. Under the H1 hypothesis, the electromagnetic signal received by the radar also contains the target echo signal r, the interference signal i and the system noise n.

[0020] In the presence of interference, when there is no target in the environment, the electromagnetic signal received by the radar is

[0021] H0:y=s*j+n(I)

[0022] Where n represents the system noise vector and * represents the convolution operator. When there is a target in the environment, the electromagnetic signal received by the radar is

[0023] H1:y=s*t+s*j+n(II)

[0024] Using the linear properties of convolution operation, equations (I) and (II) can be written in the following matrix form:

[0025] H0:y=St+n(III)

[0026] as well as

[0027] H1:y=St+Sj+n(IV)

[0028] Where S represents the augmented transmit signal matrix, and the specific expression is:

[0029]

[0030] Where L represents the discrete sampling length of the radar transmit signal.

[0031] Based on the established electromagnetic signal model, the likelihood functions P(y; H0) and P(y; H1) under the H0 hypothesis and H1 hypothesis are given respectively. The expression of the relative entropy between P(y; H0) and P(y; H1) is calculated using the likelihood function, and the functional relationship between the relative entropy and the radio frequency noise interference j is obtained.

[0032] Using Rice model Modeling the target response, where t0 represents the expected target response, R t represents the covariance matrix of the target response, Represents complex Gaussian distribution; using Gaussian distribution Modeling RF noise, where R represents the covariance matrix of interference; system noise n is represented by power σ 2 Gaussian white noise model.

[0033] Under the condition of H0, the probability density function of the radar receiving electromagnetic signals is:

[0034]

[0035] Where exp represents the exponential function, det represents the matrix determinant, I represents the identity matrix, N is the number of discrete samples of the received signal, and the superscript H denotes the conjugate transpose operator, the superscript -1 represents the matrix inversion operation; under the H1 condition, the probability density function of the radar receiving electromagnetic signal is

[0036]

[0037] The relative entropy expressions of P(y; H0) and P(y; H1) are:

[0038]

[0039] Where log represents the logarithmic function and tr represents the trace function of the matrix.

[0040] Target detection in an interference environment is essentially a binary detection problem, that is, the radar determines whether the received signal y comes from P(y; H0) or P(y; H1). Under the condition of a given interference signal power, the present invention designs the covariance matrix R of the interference signal by minimizing the relative entropy of P(y; H0) and P(y; H1), making it difficult for the radar to make a correct judgment. So far, the following interference signal design model based on relative entropy minimization is established:

[0041]

[0042] Where Ej represents the total power of interference noise, and the constraint expression tr(SRS H )=E j Indicates that the total power of the interference noise is determined, and the constraint expression R ≥ 0 indicates that the covariance matrix of the RF noise should have a semi-positive definite structure;

[0043] Step 106 , using radio frequency noise interference as a decision variable and relative entropy as an objective function, establish an interference signal optimization model based on relative entropy minimization under the constraint of radio frequency noise power; optimize the interference signal optimization model using an upper bound minimization method to obtain optimized radio frequency noise interference.

[0044] Taking the RF noise interference j as the decision variable and the relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the RF noise power constraint to complete the interference signal design modeling. The upper bound minimization method is used to optimize formula (VIII). From formula (VIII), it can be seen that the optimization of the interference signal is equivalent to optimizing the covariance matrix R of the RF noise j; it is optimized using the MM method, as shown in Figure 2 As shown in FIG, it is a schematic diagram of the MM iterative optimization process;

[0045] Step 108: Convolve the optimized radio frequency noise interference with the radar transmission signal to obtain an optimized interference signal.

[0046] After completing the interference signal optimization and obtaining the optimized RF noise interference j, it is convolved with the radar's transmit signal s to obtain the optimized interference signal i.

[0047] The aforementioned radar jammer signal design and optimization method based on the relative entropy criterion models the radar target detection problem as a binary hypothesis testing problem, describing the characteristics of the jammer and target signals under different hypotheses. Likelihood functions are established for each hypothesis, and the relative entropy between the two likelihood functions is calculated. Based on information theory, relative entropy measures the degree of distinguishability of the radar received signal under the two hypotheses. The smaller the relative entropy, the more difficult it is for the radar to distinguish between the two hypotheses, which reduces the reliability of target detection. By minimizing the relative entropy between the two hypotheses, the radar has difficulty distinguishing between the two hypotheses when detecting targets. Compared to traditional interference-to-signal ratio and mutual information criteria, the relative entropy criterion directly approaches hypothesis testing, making it more suitable for radar detection. RF noise interference is used as the optimization decision variable, and a power constraint is introduced to ensure that the jammer achieves optimal performance within the power budget. Mathematical optimization techniques are used to solve the objective function, making the jammer signal design more theoretically feasible and practical. The optimized RF noise interference is convolved with the radar transmit signal to generate the final jammer signal. This convolution operation makes the generated jammer signal more consistent with radar signal characteristics, improving the jammer's effectiveness in real-world scenarios. At the same time, the optimized interference signal acts more accurately on the radar's signal processing chain, significantly reducing the radar's ability to detect targets in noise. Compared with traditional methods based on JSR or MI, the interference signal designed based on the relative entropy criterion is more targeted. Currently, there is no method for designing interference signals based on the relative entropy criterion in the literature and patents. This application fills the gap in the current lack of interference signal design based on the relative entropy criterion. The interference signal obtained according to the proposed design method can effectively interfere with the radar detector, improve the radar signal interference effect, and significantly reduce the probability of radar detection.

[0048] In one embodiment, the probability distribution of the optimized interference signal is calculated; based on the probability distribution of the optimized interference signal, the Monte Carlo method is used to generate data samples that satisfy the likelihood function distribution; under the premise of a given false alarm probability, the detection probability of the Neyman-Pearson detector before and after the interference is compared, and the receiver operating characteristic curve is obtained to realize the calculation and estimation of the interference performance of the optimized interference signal.

[0049] In a specific embodiment, interference performance calculation and estimation includes the following specific steps:

[0050] S4-1: Use the Monte Carlo method to generate several samples that obey P(y; H0) and P(y; H1) to simulate the electromagnetic signal received by the radar; construct the following likelihood ratio test statistic

[0051]

[0052] Where Re{·} represents the real part operation;

[0053] S4-2: According to formula (IX), the detector output under the H0 condition is calculated using the samples generated in the distribution P(y; H0), and a distribution curve of the detector output under the H0 condition is generated. The 1-P fa The value of the quantile is used as the detection threshold γ of the detector, where P fa represents the false alarm probability of the radar system;

[0054] S4-3: Again according to equation (IX), the detector output under the H1 condition is calculated using the samples generated in the distribution P(y; H1). The frequency of the detector output exceeding the detection threshold γ under the H1 condition is calculated. This frequency is divided by the total number of samples under the H1 condition to obtain the detection probability under the interference condition.

[0055] S4-4: Adjust the false alarm probability P fa , repeat steps S4-2 and S4-3 to obtain the detection probability under different false alarm probabilities, and draw the receiver operating characteristic curve.

[0056] Figure 3 The following is a schematic diagram of the interference performance evaluation provided by the example of this application. The horizontal axis in the figure represents the false alarm probability P fa The vertical axis represents the value of the corresponding detection probability; it can be seen from the receiver operating characteristic curve in the figure that compared with the current interference designed based on the JSR and MI criteria, the interference designed based on the relative entropy criterion proposed in this application can further reduce the radar detection probability by more than 20%, and has a better interference effect.

[0057] In one embodiment, the electromagnetic signals in the scene are modeled as radar transmission signals, target echo signals, and interference signals, including:

[0058] The electromagnetic signal in the scene is modeled as three main parts: radar transmission signal s, target echo signal r, and interference signal i. The target echo signal r is expressed as the convolution of the radar transmission signal s and the target response t, and the interference signal i is expressed as the convolution of the radar transmission signal s and the radio frequency noise interference j. Gaussian noise n is used to simulate the radar system noise. At this point, the electromagnetic signal received in the radar interference scenario can be expressed as a combination of the target echo r, the interference signal i, and the system noise n.

[0059] In a specific embodiment, the symbol s represents the electromagnetic signal radiated into space by the radar. When the electromagnetic signal irradiates the target, it is reflected by the target and received by the radar. The symbol t represents the response of the target. The target reflection signal received by the radar is s*t, where * represents a convolution operation. The interference signal generated by the jammer is represented by i. After the jammer intercepts the electromagnetic signal transmitted by the radar, it convolves it with the radio frequency noise j and then radiates it to the radar. The interference signal received by the radar is i=s*j.

[0060] In one embodiment, likelihood functions under two assumptions are established, including:

[0061] Under the first assumption, the electromagnetic signal received by the radar only contains interference signals and system noise. The likelihood function under the first assumption is established as

[0062]

[0063] Among them, H0 represents the first hypothesis, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of radio frequency noise interference, exp represents the exponential function, det represents the matrix determinant, I represents the identity matrix, N is the number of discrete samples of the received signal, the superscript H represents the conjugate matrix, and y represents the electromagnetic signal received by the radar.

[0064] In one embodiment, likelihood functions under two assumptions are established, including:

[0065] Under the second assumption, the electromagnetic signal received by the radar contains target echo signal, interference signal and system noise at the same time. The likelihood function under the second assumption is established as

[0066]

[0067] Among them, H1 represents the second hypothesis, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of radio frequency noise interference, exp represents the exponential function, det represents the matrix determinant, I represents the identity matrix, N is the number of discrete samples of the received signal, the superscript H represents the conjugate matrix, y represents the electromagnetic signal received by the radar, t0 represents the expectation of the target response, and R t Represents the covariance matrix of the target response.

[0068] In one embodiment, the expression for calculating the relative entropy between two likelihood functions is:

[0069]

[0070] Among them, log represents the logarithmic function, tr represents the trace function of the matrix, det represents the matrix determinant, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, t0 represents the expectation of the target response, R t represents the covariance matrix of the target response, and N is the number of discrete samples of the received signal.

[0071] In one embodiment, taking radio frequency noise interference as a decision variable and relative entropy as an objective function, an interference signal optimization model based on relative entropy minimization is established under radio frequency noise power constraints, including:

[0072] Taking radio frequency noise interference as the decision variable and relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the constraint of radio frequency noise power.

[0073]

[0074] subject to tr(SRS H )=E j ,R≥0.

[0075] Where log represents the logarithmic function, E j Represents the total power of interference noise, the constraint expression tr(SRS H )=E j Indicates that the total power of the interference noise is determined, the constraint expression R ≥ 0 indicates that the covariance matrix of the RF noise interference should have a semi-positive definite structure, det represents the matrix determinant, tr represents the matrix trace function, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, t0 represents the expectation of the target response, R t Represents the covariance matrix of the target response.

[0076] In one embodiment, the interference signal optimization model is optimized using an upper bound minimization method to obtain optimized radio frequency noise interference, including:

[0077] Step 1: Initialize the number of iterations k = 0 and set R (k) =αI, where E j represents the total power of the interference noise, S represents the augmented transmit signal matrix, R represents the covariance matrix of the RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, and tr represents the trace function of the matrix;

[0078] Step 2: Construct the following proxy function

[0079]

[0080] Where,

[0081]

[0082] M (k) =S H (S(R t +R(k) )S H +σ 2 I) -1 S

[0083] E A =[I 0]

[0084] as well as

[0085]

[0086] Among them, t0 represents the expected target response, R t represents the covariance matrix of the target response, σ 2 Indicates power;

[0087] Step 3: Use the interior point method to solve the convex optimization problem. The convex optimization problem is:

[0088]

[0089] subject to tr(SRS H )=E j ,R≥0

[0090] Among them, the constraint expression tr(SRS H )=E j Indicates that the total power of the interference noise is determined, and the constraint expression R≥0 indicates that the covariance matrix of the RF noise interference should have a semi-positive definite structure;

[0091] The solution to the convex optimization problem is denoted as R (k+1) And calculate the value of relative entropy;

[0092] Step 4: Determine whether the value of relative entropy converges. If so, set R (k+1) As the optimized radio frequency noise interference, if it does not converge, set k=k+1 and return to step 2.

[0093] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0094] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are 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.

[0095] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A radar jamming signal design and optimization method based on relative entropy criterion, characterized in that: The method comprises: Modeling the electromagnetic signals in the radar jamming scenario, modeling the electromagnetic signals in the scenario into radar transmission signals, target echo signals and jamming signals; The radar target detection problem under interference is formulated as a binary hypothesis testing problem. Under the first hypothesis, the electromagnetic signal received by the radar contains only the interference signal and system noise. Under the second hypothesis, the electromagnetic signal received by the radar contains the target echo signal, the interference signal, and system noise. The likelihood functions under the two hypotheses are established and the expression of the relative entropy between the two likelihood functions is calculated, resulting in a functional relationship between the relative entropy and the RF noise interference. Taking radio frequency noise interference as the decision variable and relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the constraint of radio frequency noise power. The interference signal optimization model is optimized using the upper bound minimization method to obtain the optimized radio frequency noise interference. The optimized radio frequency noise interference is convolved with the radar transmission signal to obtain the optimized interference signal.

2. The method according to claim 1, characterized in that The method further comprises: The probability distribution of the optimized interference signal is calculated; based on the probability distribution of the optimized interference signal, a Monte Carlo method is used to generate data samples that satisfy the likelihood function distribution. Under the premise of a given false alarm probability, the detection probability of the Neyman-Pearson detector before and after the interference is compared to obtain the receiver operating characteristic curve to realize the calculation and estimation of the interference performance of the optimized interference signal.

3. The method according to claim 1, characterized in that Model the electromagnetic signals in the scene as radar transmission signals, target echo signals, and interference signals, including: Modeling electromagnetic signals in the scene as radar emissions s , target echo signal r, interference signal i three main parts, where the target echo signal r is represented by the radar transmission signal s The convolution form of the target response t, the interference signal i is represented by the radar transmission signal s and RF noise interference j convolution form; using Gaussian noise n Simulating radar system noise. At this point, the electromagnetic signal received in the radar jamming scenario can be represented as the target echo r, the jamming signal i, and the system noise. n combination.

4. The method according to claim 1, wherein The likelihood functions under two assumptions are established respectively, including: Under the first assumption, the electromagnetic signal received by the radar only contains interference signals and system noise. The likelihood function under the first assumption is established as Among them, H0 represents the first hypothesis, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of radio frequency noise interference, exp represents the exponential function, det represents the matrix determinant, I represents the identity matrix, N is the number of discrete samples of the received signal, the superscript H represents the conjugate matrix, and y represents the electromagnetic signal received by the radar.

5. The method according to claim 1, wherein The likelihood functions under two assumptions are established respectively, including: Under the second assumption, the electromagnetic signal received by the radar contains target echo signal, interference signal and system noise at the same time. The likelihood function under the second assumption is established as Among them, H1 represents the second hypothesis, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of radio frequency noise interference, exp represents the exponential function, det represents the matrix determinant, I represents the identity matrix, N is the number of discrete samples of the received signal, the superscript H represents the conjugate matrix, y represents the electromagnetic signal received by the radar, t0 represents the expectation of the target response, and R t Represents the covariance matrix of the target response.

6. The method according to claim 1, characterized in that The expression for calculating the relative entropy between two likelihood functions is Among them, log represents the logarithmic function, tr represents the trace function of the matrix, det represents the matrix determinant, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, t0 represents the expectation of the target response, R t represents the covariance matrix of the target response, and N is the number of discrete samples of the received signal.

7. The method according to claim 1, characterized in that Taking RF noise interference as the decision variable and relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the RF noise power constraint, including: Taking radio frequency noise interference as the decision variable and relative entropy as the objective function, an interference signal optimization model based on relative entropy minimization is established under the constraint of radio frequency noise power. Where log represents the logarithmic function, E j Represents the total power of interference noise, the constraint expression tr(SRS H )=E j Indicates that the total power of the interference noise is determined, the constraint expression R ≥ 0 indicates that the covariance matrix of the RF noise interference should have a semi-positive definite structure, det represents the matrix determinant, tr represents the matrix trace function, σ 2 represents power, S represents the augmented transmit signal matrix, R represents the covariance matrix of RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, t0 represents the expectation of the target response, R t Represents the covariance matrix of the target response.

8. The method according to claim 1, characterized in that The interference signal optimization model is optimized using the upper bound minimization method to obtain the optimized RF noise interference, including: Step 1: Initialize the number of iterations k = 0 and set R (k) =αI, where E j represents the total power of the interference noise, S represents the augmented transmit signal matrix, R represents the covariance matrix of the RF noise interference, I represents the identity matrix, the superscript H represents the conjugate matrix, and tr represents the trace function of the matrix; Step 2: Construct the following proxy function Where, M (k) =S H (S(R t +R (k) )S H +σ 2 I) -1 S E A =[I 0] as well as Among them, t0 represents the expected target response, R t represents the covariance matrix of the target response, σ 2 Indicates power; Step 3: Use the interior point method to solve the convex optimization problem, which is: Among them, the constraint expression tr(SRS H )=E j Indicates that the total power of the interference noise is determined, and the constraint expression R≥0 indicates that the covariance matrix of the RF noise interference should have a semi-positive definite structure; The solution to the convex optimization problem is denoted as R (k+1) And calculate the value of relative entropy; Step 4: Determine whether the value of relative entropy converges. If so, set R (k+1) As the optimized radio frequency noise interference, if it does not converge, set k=k+1 and return to step 2.

Images