A radar anti-jamming decision-making method combining online preferred domain and intra-domain decision-making
By combining online domain optimization with intra-domain decision-making, and utilizing template matching and Manhattan distance calculation, the radar anti-jamming decision domain is selected and a benefit value sequence is generated. This solves the problem of low efficiency of radar anti-jamming decision-making methods in complex electromagnetic environments and achieves efficient anti-jamming decision-making.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIDIAN UNIV
- Filing Date
- 2023-02-20
- Publication Date
- 2026-05-26
AI Technical Summary
Existing radar anti-jamming decision-making methods are difficult to apply in complex electromagnetic environments, have low decision-making efficiency, and cannot effectively cope with diverse interference signals.
A method combining online domain selection and intra-domain decision-making is adopted. The features of radar and jamming signals are extracted by template matching technology, the similarity is calculated using Manhattan distance to select the decision domain, and an anti-jamming benefit value sequence is generated. The anti-jamming measure with the maximum benefit value is directly selected for decision-making.
It reduces decision-making complexity, improves the efficiency and effectiveness of radar anti-jamming decision-making, and enables it to adapt to complex electromagnetic environments.
Smart Images
Figure CN116299214B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of radar communication technology, and further relates to a radar anti-jamming decision-making method based on a combination of online domain optimization and intra-domain decision-making in the field of electronic countermeasures technology. This invention can be applied to generate corresponding radar anti-jamming schemes against jammers. Background Technology
[0002] The strength of a radar's anti-jamming capability determines its ability to effectively perform target detection, tracking, and guidance functions in jammed environments. Faced with a rapidly changing electromagnetic environment and increasingly diverse jamming patterns, higher demands are placed on radar anti-jamming technology and the selection of anti-jamming strategies. Radar anti-jamming technology is also rapidly advancing with the times. Jamming technology and anti-jamming technology are a unified whole, with the development of jamming technology inevitably leading to the advancement of anti-jamming technology. In recent years, the types of anti-jamming measures have become increasingly diverse.
[0003] Li Ming et al., in their paper "Research on Intelligent Anti-jamming Strategy Based on Finite Zero-Sum Game and Transform Domain Optimization" (Signal Processing, 2020, 36(8):1253-1262.), disclosed a radar anti-jamming decision-making method based on finite zero-sum game and transform domain optimization. The implementation scheme of this method is as follows: First, establish a transform domain anti-jamming game adversarial model, determine the bit error rate (BER) for each combination of anti-jamming and anti-jamming patterns, and the profit matrix contains the BER values corresponding to each pair of anti-jamming measures. Once the BER is determined, the profit matrix can be derived. Second, provide methods for solving the Nash equilibrium for pure and mixed strategies. Third, use the Nash equilibrium theorem in game theory to solve the matrix and obtain the radar's anti-jamming strategy. This method has two shortcomings. First, it relies on the establishment of a profit matrix. In real-world applications, there are numerous types of interference and anti-interference mechanisms. For the radar, there are many unknown types of interference signals, which are complex and highly variable. The radar cannot obtain all the information about the interference, and the relationship between interference and anti-interference cannot be fully constructed. This incomplete information makes it difficult to establish the profit matrix in game theory. Second, the Nash equilibrium provides an equilibrium point for both sides' strategy choices; it represents a balanced strategy acceptable to both sides, rather than the strategy that maximizes profit.
[0004] Xi'an University of Electronic Science and Technology proposed a radar anti-jamming decision-making method in its patent application "An Anti-jamming Decision-Making Method Applied to Cognitive Radar" (Application No.: CN 202110969210.1; Publication No.: CN 113866723 A). The implementation process of this method is as follows: First, acquire the environmental state and identify the type of interference. Second, use the identified environmental state information as the initial state value. Third, execute a reinforcement learning algorithm based on a Markov decision model; the algorithm terminates when the Q-matrix converges. Fourth, based on the converged Q-matrix, select anti-jamming actions using a standard greedy algorithm. The Markov decision process treats the mapping relationship between anti-jamming measures and states as a table; when encountering a certain state, it looks up the corresponding action in the table. The drawback of this method is that in order to generate this table, it is necessary to traverse all interference threat states. When there are too many states and too many radar anti-jamming measures, the mapping relationship table between interference and anti-jamming measures will also become larger, and the amount of data and calculation time in the calculation process will also become longer. In complex electromagnetic environments, there is a problem of low decision-making efficiency. Summary of the Invention
[0005] The purpose of this invention is to address the problems existing in the prior art by providing a radar anti-jamming decision-making method that combines online preferred domain and intra-domain decision-making. This aims to solve the problems that existing anti-jamming decision-making methods are difficult to apply to working scenarios with complex electromagnetic environments and that existing decision-making methods have low returns.
[0006] The present invention aims to achieve its objective by employing a radar anti-jamming decision-making technique that combines online domain optimization with intra-domain decision-making. Features of the interference signal and radar signal are extracted across three sub-domains, and the similarity between the interference signal and the radar transmitted signal in each sub-domain is calculated using the Manhattan distance formula. The domain with the highest similarity is selected as the radar anti-jamming decision domain, eliminating consideration of anti-interference measures in other domains. This reduces the number of anti-jamming measures and significantly lowers the complexity of subsequent decisions, making it adaptable to complex electromagnetic environments. Furthermore, the anti-jamming benefit value of each anti-jamming measure is calculated, and these measures are sorted from largest to smallest based on their anti-jamming benefit value to obtain an anti-jamming benefit value sequence. The anti-jamming measure with the highest anti-jamming benefit value is directly selected during the decision-making process, solving the problems of long computation time and low efficiency. Measures with higher anti-jamming benefit values (i.e., greater returns) have higher priority, thus the establishment of an anti-jamming decision database increases decision-making benefits and addresses the problem of low returns in traditional decision-making methods.
[0007] The specific steps to achieve the objective of this invention are as follows:
[0008] Step 1: Use template matching technology to determine the radar anti-jamming decision domain online.
[0009] Step 1.1: Extract the pulse width, pulse repetition period, and carrier frequency features from the time-frequency domain of the radar transmitted signal, the half-power beamwidth features from the spatial domain, and the polarization angle features from the polarization domain to form a radar signal feature template.
[0010] Step 1.2: Extract the pulse width, pulse repetition period, and carrier frequency characteristics of the jamming signal transmitted by the jammer in the time-frequency domain, the half-power beamwidth characteristics in the spatial domain, and the polarization angle characteristics in the polarization domain. Using the Manhattan distance formula, calculate the similarity between the jamming signal and the radar transmitted signal in the time-frequency domain, spatial domain, and polarization domain, respectively. Select the domain corresponding to the highest similarity as the radar anti-jamming decision domain.
[0011] Step 2: Generate the anti-interference benefit value sequence.
[0012] Seven anti-jamming measures and four jamming patterns across three domains were simulated 28 times under known radar jamming parameters. The anti-jamming benefit value of each anti-jamming measure for each jamming pattern was calculated. The anti-jamming benefit values were sorted from largest to smallest to obtain the benefit sequence of the q-th jamming in the m-domain. Based on the time-frequency domain, spatial domain, and polarization domain, a total of 12 anti-jamming benefit value sequences were generated for the four jamming patterns across three domains.
[0013] Step 3: Generate anti-jamming decision results within the radar anti-jamming decision domain.
[0014] Select the sequence of anti-jamming benefit values corresponding to the jamming pattern of the received jamming signal in the radar anti-jamming decision domain, and select the anti-jamming measure with the maximum anti-jamming benefit value from the sequence as the anti-jamming decision result.
[0015] Compared with the prior art, the present invention has the following advantages:
[0016] First, the radar anti-jamming decision domain is selected online, eliminating the need to consider anti-jamming measures in other domains. This simplifies the state space, reduces the number of anti-jamming measures, and greatly reduces the complexity of subsequent decisions, enabling the radar anti-jamming decision method of this invention to adapt to complex electromagnetic environments.
[0017] Second, this invention uses a radar anti-jamming decision-making method that combines online preferred domain and intra-domain decision-making to generate decision results, directly selecting the anti-jamming measure with the maximum anti-jamming benefit value in the decision. The greater the anti-jamming benefit value, the higher the probability of selecting this anti-jamming measure in the decision. Therefore, the establishment of the anti-jamming decision database solves the problem of low returns in traditional decision-making methods, giving the radar anti-jamming strategy decision-making method generated by this invention the advantage of high returns. Attached Figure Description
[0018] Figure 1 This is a flowchart of the present invention. Detailed Implementation
[0019] The present invention will now be described in further detail with reference to the accompanying drawings.
[0020] Reference Figure 1 The specific implementation steps of the present invention will be described in further detail below.
[0021] Step 1: Use template matching technology to determine the radar anti-jamming decision domain online.
[0022] Step 1.1: Extract the pulse width, pulse repetition period, and carrier frequency features from the time-frequency domain of the radar transmitted signal, the half-power beamwidth features from the spatial domain, and the polarization angle features from the polarization domain to form a radar signal feature template.
[0023] Step 1.2: Extract the pulse width, pulse repetition period, and carrier frequency characteristics of the jamming signal transmitted by the jammer in the time-frequency domain, the half-power beamwidth characteristics in the spatial domain, and the polarization angle characteristics in the polarization domain. Using the Manhattan distance formula, calculate the similarity between the jamming signal and the radar transmitted signal in the time-frequency domain, spatial domain, and polarization domain, respectively. Select the domain corresponding to the highest similarity as the radar anti-jamming decision domain.
[0024] The Manhattan distance formula is as follows:
[0025]
[0026] Where, D(X,Y) m Let D(X,Y) represent the similarity between radar signals and jamming signals in the m-th domain. m The smaller the value, the higher the similarity. X and Y represent radar signal samples and interference signal samples in a certain domain, respectively. m represents the type of domain, m=1 represents the time-frequency domain, m=2 represents the spatial domain, m=3 represents the polarization domain, n is the total number of features, "∑" represents the summation operation, and "||" represents the modulo operation. i m represents the eigenvalue of the i-th feature of the radar signal in the m-th domain, y i m The eigenvalue represents the i-th feature of the interference signal in the m-th domain. i represents the feature index: i=1 represents the pulse width, i=2 represents the pulse repetition period feature, i=3 represents the carrier frequency feature, i=4 represents the half-power beamwidth feature, and i=5 represents the polarization angle feature. Using the Manhattan distance formula, the similarity between the interference signal and the radar transmitted signal is calculated in the time-frequency domain, spatial domain, and polarization domain, respectively. The domain corresponding to the highest similarity is selected as the radar anti-jamming decision domain.
[0027] Step 2: Generate the anti-interference benefit value sequence.
[0028] Seven anti-jamming measures (waveform agility, frequency agility, frequency diversity, sidelobe cancellation, sidelobe concealment, polarization agility, and polarization filtering) and four jamming patterns (frequency sweep jamming, radio frequency noise jamming, dense false target jamming, and agile noise jamming) were simulated 28 times under known radar jamming parameters. The anti-jamming effectiveness value of each anti-jamming measure for each jamming pattern was calculated. The anti-jamming effectiveness value of each anti-jamming measure for each jamming pattern is obtained by the following formula:
[0029]
[0030] in, This represents the anti-interference effectiveness value of the q-th interference pattern corresponding to the p-th anti-interference measure within the m-th domain. p=1 represents waveform agility, p=2 represents frequency agility, p=3 represents frequency diversity, p=4 represents sidelobe cancellation, p=5 represents sidelobe concealment, p=6 represents polarization agility, p=7 represents polarization filtering, q=1 represents radio frequency noise interference, q=2 represents frequency sweep interference, q=3 represents agile noise interference, and q=4 represents dense decoy interference. This represents the signal-to-noise ratio (SNR) of the q-th interference pattern corresponding to the p-th anti-interference measure within the m-th domain after applying the anti-interference measure. This represents the signal-to-noise ratio (SNR) of the qth interference pattern corresponding to the pth anti-interference measure within the mth domain when no anti-interference measure was applied. Using the anti-interference benefit value as the basis for ranking anti-interference measures, the measures are sorted from largest to smallest to obtain the benefit sequence of the qth interference in the mth domain. Based on the time-frequency domain, spatial domain, and polarization domain, a total of 12 anti-interference benefit value sequences are generated across 3 domains for 4 interference patterns.
[0031] For ease of presentation, the anti-interference effectiveness values for the seven anti-interference measures corresponding to four interference patterns are displayed in tabular form. The anti-interference effectiveness value table is as follows:
[0032] Table 1 Anti-interference benefit value
[0033]
[0034] Step 3: Generate anti-jamming decision results within the radar anti-jamming decision domain.
[0035] Select the sequence of anti-jamming benefit values corresponding to the jamming pattern of the received jamming signal in the radar anti-jamming decision domain, and select the anti-jamming measure with the maximum anti-jamming benefit value from the sequence as the anti-jamming decision result.
Claims
1. A radar anti-jamming decision method combining online preferred domain and domain decision, characterized in that, The anti-jamming decision-making result is generated by utilizing the radar anti-jamming decision domain and the anti-jamming benefit value sequence. The specific steps of this decision-making method include the following: Step 1: Determine the radar anti-jamming decision domain online using template matching technology: Step 1.1: Extract the pulse width, pulse repetition period, and carrier frequency features from the time-frequency domain of the radar transmitted signal, the half-power beamwidth features from the spatial domain, and the polarization angle features from the polarization domain to form a radar signal feature template. Step 1.2: Extract the pulse width, pulse repetition period, and carrier frequency features of the jamming signal transmitted by the jammer in the time-frequency domain, the half-power beamwidth features in the spatial domain, and the polarization angle features in the polarization domain. Using the Manhattan distance formula, calculate the similarity between the jamming signal and the radar transmitted signal in the time-frequency domain, spatial domain, and polarization domain, respectively. Select the domain corresponding to the highest similarity as the radar anti-jamming decision domain. The Manhattan distance formula is obtained by calculating the following equation: ； in, For radar signals and jamming signals at the 1st Domain similarity, The smaller the value, the higher the similarity. and These represent radar signal samples and interference signal samples in a certain domain, respectively. Indicates the type of field. Representing the time-frequency domain, Indicates airspace, Represents the polarization domain. For the total characteristic number, "Represents a summation operation," " indicates a modulo operation, Represents the radar signal at the The domain's first The feature values of each feature Represents the interference signal in the m-th domain. The eigenvalues of each feature; Indicates the sequence number of the feature. Indicates pulse width. Indicates the pulse repetition periodicity. Indicates carrier frequency characteristics, Indicates the half-power beamwidth characteristics. Indicates polarization angle characteristics; Step 2, generate the anti-interference benefit value sequence: For seven anti-jamming measures and four jamming patterns across three domains, 28 simulations were performed with known radar jamming parameters. The anti-jamming effectiveness value for each anti-jamming measure and each jamming pattern was calculated. The anti-jamming effectiveness values were sorted from largest to smallest to obtain the [number of simulations]. Such interference The domain benefit sequence is divided into time-frequency domain, spatial domain and polarization domain, generating 12 anti-interference benefit value sequences in 3 domains and 4 interference patterns; The calculation of the anti-interference benefit value of each anti-interference measure for each interference pattern is obtained by the following formula: ； in, Indicates the first Within each domain The first anti-interference measure corresponds to the first The anti-interference benefit value of each interference pattern. Indicates waveform agility. Indicates frequency agility. Indicates frequency diversity. This indicates that the side lobes cancel each other out. This indicates that the side petals are hidden. Indicates polarization agility. Indicates polarization filtering. This indicates radio frequency noise interference. This indicates frequency sweep interference. Indicates clever noise interference. This indicates dense decoy interference. Indicates the first Within each domain The first anti-interference measure corresponds to the first Signal-to-noise ratio after applying anti-interference measures to each interference pattern Indicates the first Within each domain The first anti-interference measure corresponds to the first Signal-to-noise ratio of each interference pattern without anti-interference measures applied; Step 3: Generate anti-jamming decision results within the radar anti-jamming decision domain: Select the sequence of anti-jamming benefit values corresponding to the jamming pattern of the received jamming signal in the radar anti-jamming decision domain, and select the anti-jamming measure with the maximum anti-jamming benefit value from the sequence as the anti-jamming decision result.
2. The radar anti-jamming decision-making method combining online preferred domain and intra-domain decision-making according to claim 1, characterized in that, The seven anti-interference measures in the three domains mentioned in step 2 refer to waveform agility, frequency agility, and frequency diversity in the time-frequency domain; sidelobe cancellation and sidelobe obfuscation in the spatial domain; and polarization agility and polarization filtering in the polarization domain.
3. The radar anti-jamming decision-making method combining online preferred domain and intra-domain decision-making according to claim 1, characterized in that, The four types of interference mentioned in step 2 refer to frequency sweep interference, radio frequency noise interference, dense decoy interference, and agile noise interference.