RVGPO interference method based on random simulation optimization
By modeling the RVGPO interference strategy as a random simulation optimization problem and deducing matching relationships, combined with the IPSO-OCBA algorithm, the RVGPO interference strategy is optimized, and the RVGPO interference strategy optimization problem in the existing technology is solved, significantly improving the interference performance.
Patent Information
- Application Number
- CN202411020707.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-01
AI Technical Summary
The optimization of existing RVGPO interference strategy mainly relies on qualitative analysis and expert experience, and the lack of quantitative optimization models and algorithms leads to limited interference performance.
Model the RVGPO interference strategy as a large-scale random simulation optimization problem, deduce the matching relationship between controlled delay and Doppler shift, design the interference strategy optimization algorithm based on IPSO-OCBA, use the matching relationship to eliminate invalid strategies, and narrow the search range , improve optimization efficiency.
It significantly improves the interference effect of RVGPO, overcomes the problems of high variable dimensions and strong coupling, and improves the interference performance.
Smart Images

Figure CN118884367A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of radar electronic countermeasures and relates to an RVGPO interference method based on random simulation optimization. Background Art
[0002] RVGPO (Range-Velocity Gate Pull-Off) jamming is an efficient deception jamming technology that can jam advanced radars with range-velocity joint tracking capabilities and is one of the research hotspots in the field of electronic countermeasures. In this jamming method, the jammer generates false targets by continuously forwarding radar transmission signals with time-frequency modulation (i.e., controlled time delay and Doppler frequency shift), forcing the radar tracking system to lose the real target and form a false track. However, the performance of RVGPO jamming depends largely on the setting of controlled time delay and Doppler frequency shift (i.e., jamming strategy). Since advanced radars with range-velocity joint tracking capabilities will check the matching degree between the target range and velocity (i.e., a smaller speed can hardly produce a significant range offset), jamming strategies that do not meet the matching requirements cannot deceive the radar's range and velocity trackers at the same time. Therefore, how to generate a reasonable jamming strategy is the key to RVGPO jamming research.
[0003] At present, the literature on the optimization of RVGPO jamming strategy is still at the stage of qualitative analysis. In the literature "Kirubarajan T, IMMPDAF for radar management and tracking benchmark with ECM. IEEE Transactions on Aerospace and Electronic Systems, 1998, 34 (4): 1115-1134." and "Benchmark for radar allocation and tracking in ECM, IEEE Transactions on Aerospace and Electronic Systems, 1998, 34 (4): 1097-1114", the authors proposed uniform speed and uniform acceleration drag jamming methods respectively, so that the false target can be separated from the real target in the form of uniform speed and uniform acceleration. In the literature "Multitarget-multisensor tracking: Applications and advances. Volume iii. Norwood, MA, USA: Artech House, 2000", the authors qualitatively analyzed the impact of RVGPO jamming strategy on jamming effect, and further proposed that the controlled delay and Doppler frequency shift should be consistent with the preset motion model of the false target. However, the jamming strategy settings in the above studies all rely heavily on expert experience, and lack the quantitative optimization model of the RVGPO jamming strategy and the corresponding strategy optimization algorithm. However, the quantitative optimization of the RVGPO jamming strategy is a bottleneck problem that must be overcome to further improve the jamming performance. Summary of the invention
[0004] In view of the gap in the background technology, the present invention models the optimization of the RVGPO interference strategy as a large-scale random simulation optimization problem, derives the matching relationship between the controlled delay and Doppler frequency shift in the RVGPO interference strategy, designs a candidate strategy search mechanism based on the matching relationship, and proposes an RVGPO interference strategy optimization algorithm based on IPSO-OCBA. The algorithm uses the matching relationship to exclude invalid RVGPO interference strategies, greatly reduces the search range of the algorithm, greatly improves the optimization efficiency, effectively alleviates the problems of high variable dimension and strong coupling in the optimization of the RVGPO interference strategy, and significantly improves the interference effect.
[0005] The object of the present invention is achieved through the following technical scheme: a RVGPO interference method based on random simulation optimization, comprising the following steps:
[0006] Step 1: Modeling of typical RVGPO interference scenarios;
[0007] Step 2: Establish an RVGPO interference strategy optimization model;
[0008] Step 3: derive the matching relationship between the controlled time delay and the Doppler frequency shift;
[0009] Step 4: Optimize the RVGPO interference strategy.
[0010] The beneficial effects of the present invention are as follows: the present invention models the optimization of the RVGPO interference strategy as a large-scale random simulation optimization problem, derives the matching relationship between the controlled time delay and the Doppler frequency shift in the RVGPO interference strategy, designs a candidate strategy search mechanism based on the matching relationship, and proposes an RVGPO interference strategy optimization algorithm based on IPSO-OCBA, which overcomes the difficulty of quantitative optimization of the RVGPO interference strategy. The advantages of the present invention are that the matching relationship is used to exclude invalid RVGPO interference strategies, which greatly reduces the search range of the algorithm, greatly improves the optimization efficiency, effectively alleviates the problems of high variable dimension and strong coupling in the optimization of the RVGPO interference strategy, and significantly improves the interference effect. The present invention can be applied to fields such as electronic countermeasures. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 This is a schematic diagram of a typical RVGPO interference scenario;
[0012] Figure 2 is a flow chart of the RVGPO interference method based on random simulation optimization of the present invention;
[0013] Figure 3 It is a flowchart of the IPSO-OCBA algorithm provided by the present invention;
[0014] Figure 4 is an iterative convergence curve under different matching weight settings in a specific implementation manner of the present invention;
[0015] Figure 5 It is the iterative convergence curve of the two algorithms when J=10000 and K=15 in the specific implementation mode of the present invention;
[0016] Figure 6 It is the iterative convergence curve of the two algorithms in the case of J=5000, K=15 in the specific implementation mode of the present invention;
[0017] Figure 7 It is the iterative convergence curve of the two algorithms in the case of J=10000, K=20 in the specific implementation mode of the present invention;
[0018] Figure 8 It is the comprehensive interference performance of the four interference methods at different stages when J=10000 and K=15 in the specific implementation mode of the present invention. DETAILED DESCRIPTION
[0019] The present invention establishes an RVGPO interference strategy optimization model based on random simulation optimization, and proposes an interference strategy optimization algorithm based on IPSO-OCBA (Improved Particle Swarm Optimization-Optimal Computational Budget Allocation), which effectively solves the problem of quantitative optimization of RVGPO interference strategy.
[0020] The solution of the present invention is to first model the typical RVGPO interference scenario, then use the track interruption rate and the average error distance as evaluation criteria, establish the RVGPO interference strategy optimization model based on random simulation optimization, and then derive the matching relationship between the controlled delay and the Doppler frequency shift, and finally optimize the RVGPO interference strategy using the IPSO-OCBA algorithm based on the matching relationship. This method overcomes the problem of quantitative optimization of the RVGPO interference strategy and significantly improves the interference performance.
[0021] In order to facilitate the description of the present invention, the following terms are first explained:
[0022] Term 1: RVGPO interference
[0023] RVGPO jammer is a distance-speed joint drag jammer. After the jammer intercepts the radar transmission signal, it will add time delay and Doppler frequency shift to forward it, generating a false target with false distance and speed information, affecting radar target tracking. Typical RVGPO jammer scenarios include: Figure 1 shown.
[0024] Term 2: Tracking Gate
[0025] The tracking gate is a decision threshold used to determine whether the measurement value originates from the target. It is an area centered on the predicted position (or speed) of the tracked target and is used to determine the possible range of the measurement value of the target at the next moment.
[0026] Term 3: PSO
[0027] PSO is a particle swarm optimization algorithm, which is a population-based heuristic algorithm. The particle swarm optimization algorithm simulates the predation process of bird flocks, fish schools, etc. in nature, and finds the global optimal solution to the problem through collaboration in the group.
[0028] Term 4: OCBA
[0029] OCBA is the Optimal Computational Budget Allocation scheme, which is a computing resource allocation method. The optimal computing budget allocation scheme aims to maximize the probability of the algorithm selecting the optimal candidate solution and allocate limited computing resources to more promising candidate solutions.
[0030] The present invention is mainly verified by simulation experiment, and all steps and conclusions are verified to be correct on Matlab 2021b. The technical solution of the present invention is further explained below in conjunction with the accompanying drawings.
[0031] like Figure 2 As shown, a RVGPO interference method based on random simulation optimization of the present invention comprises the following steps:
[0032] Step 1. Modeling of typical RVGPO jamming scenarios: When discussing the effect of the RVGPO jamming strategy on the radar range and speed tracker, it is necessary to first model the false targets generated by the jamming signal. Consider that a radar with range-speed joint tracking capability is tracking a target equipped with a repeater jammer. When the target detects a threat from the radar, it will implement RVGPO jammers on the radar to disrupt its tracking of itself. At each stage, the jammer intercepts the radar transmit pulse and forwards it to the radar receiver after adding additional time-frequency modulation. Therefore, the radar will detect additional false targets whose distance and speed are different from those of the real target. Assume that the controlled delay and Doppler frequency shift of the kth jamming stage are τ respectively. k and f k , then the distance d between the false target and the real target k and speed v k They are:
[0033]
[0034]
[0035] Where K represents the total number of interference stages, c represents the speed of light, λ represents the signal wavelength, and Δτ i and Δf i They represent the controlled time delay and Doppler shift increment in each interference stage respectively;
[0036] Assume that the state equation of the real target is:
[0037] X(k)=F[k,X(k-1)]+U[k,X(k-1)] (3)
[0038] Among them, F[·] represents the state transition matrix of the real target motion state in adjacent stages; U[·] represents the process noise matrix;
[0039] X(k)=[X e (k),X v (k)] T It represents the motion state of the real target in the kth interference stage, which includes the following two parts: target position X e (k) and the Doppler velocity X v (k);
[0040] Therefore, the motion state of the false target in the kth interference stage is for:
[0041]
[0042] Among them, Θ(·) and Ψ(·) represent the distance d k and speed v k Projection function into state space.
[0043] At this time, the radar will measure the position and Doppler velocity of the target; the measurement of the target position is called kinematic measurement, and the measurement of the Doppler velocity is called Doppler measurement, which is expressed as:
[0044]
[0045] Where Z(k)=[Z e (k),Z v (k)] T represents the k-th jamming stage radar's kinematic and Doppler measurements of the real target; h e [·] and h v [·] represents the measurement function; W e [·] and W v [·] represents the measurement noise matrix;
[0046] The radar's measurement value of the false target is expressed as:
[0047]
[0048] in, It represents the k-th jamming stage radar's kinematic and Doppler measurements of false targets.
[0049] In the kth jamming stage, after the radar obtains the real target and false target measurements, it first predicts the target state:
[0050]
[0051] Among them, X(k|k-1) represents the one-step prediction of the target state by the radar tracking system; represents the estimation of the target state by the radar tracking system in the k-1th jamming stage, and f[·] represents the one-step prediction function;
[0052] Then, select the measurements that are likely to come from the real target:
[0053]
[0054] Among them, g e [·] and g v [·] represents the kinematic and Doppler measurement selection functions; radars are represented by h e (k,X(k|k-1)) and Range and velocity tracking gates are established for the center to filter kinematic and Doppler measurements, where is the temporary target state obtained by applying the standard Kalman filter to the position measurement. and denote the kinematic and Doppler measurements that have been selected, respectively; only the measurements selected by both (8) and (9) are used by the radar to update the target state, so the selected measurement set satisfies Since the center of the velocity tracking gate is calculated based on the center of the range tracking gate and the position measurement, the two tracking gates established by the radar are not independent of each other. Under this tracking gate setting, if the Doppler measurement can fall into the velocity tracking gate, it can be proved that it matches the corresponding position measurement.
[0055] Finally, update the target's state estimate:
[0056]
[0057] Where φ[·] represents the fusion function.
[0058] According to formula (10), only the measurement set Therefore, the key to the effectiveness of RVGPO interference is the false target measurement. It can fall into the distance and speed tracking gates at the same time. If the RVGPO interference strategy is set properly, Participating in the target state estimation, the false motion information it carries will undoubtedly affect the target state estimation This error is not limited to the current stage (i.e., the kth stage). As shown in formula (7), The state prediction of the k+1th stage will continue to propagate the error to X(k+1|k). Compared with the case without interference, the tracking gate of the k+1th stage will shift toward the false target. At this time, if the false target measurement of the k+1th stage If the target continues to fall into the tracking gate, the estimation error of the radar will further increase, causing the tracking gate to gradually lean toward the false target and eventually lose the real target.
[0059] Step 2: Establish an optimization model for RVGPO jamming strategy. The effect of RVGPO jamming can be evaluated by the target tracking results of the radar, which mainly depends on the following two points:
[0060] 1) Whether the track is interrupted: whether the real target measurement finally falls within the tracking gate. If there is no real target measurement within the tracking gate in the final stage, that is, the radar target tracking has been interrupted, it means that the jamming effect is good. On the contrary, the tracking has not been interrupted and the jamming effect is poor.
[0061] 2) Error distance: The degree to which the tracking gate deviates from the real target. Its essence is the Euclidean distance between the radar target position estimate and the real target position. The larger the error distance, the more difficult it is for the radar to recapture the real target, and the safer the real target is.
[0062] Therefore, for a disturbance, the track interruption is defined as And the error distance for:
[0063]
[0064] Among them, r1(·) and r2(·) represent the objective function; represents the true target state of all K interference stages, represents the positional and Doppler measurement noise of all K interference stages, Indicates RVGPO interference strategy; It indicates that the radar's tracking of the real target has not been interrupted; Indicates that the radar's tracking of the real target has been interrupted;
[0065] represents the state estimate of the target in the last interference stage, and ||·||2 is the matrix norm.
[0066] Because the independent variable and The elements in are all random and uncertain. For a certain RVGPO interference strategy The result of each interference and Therefore, in order to eliminate the uncertainty in a single evaluation, two indicators for evaluating the RVGPO interference effect are obtained by taking equations (11) and (12): track interruption rate and the average error distance Where E[· ] represents the expectation. The above two evaluation indicators are weighed by weighted aggregation, and the comprehensive interference performance of the RVGPO interference strategy is obtained as follows:
[0067]
[0068] Among them, w1>0 and w2>0 represent the weighted aggregation weights of the two indicators of track interruption rate and average error distance respectively;
[0069] Taking formula (13) as the optimization objective, the RVGPO interference strategy optimization model is established as follows:
[0070]
[0071] in, and Respectively represent the time delay and frequency shift strategies; Ω τ and Ω f Represents the solution space of the RVGPO interference strategy.
[0072] Step 3: Derive the matching relationship between the controlled delay and the Doppler frequency shift; if the false target wants to fall into the tracking gate, it must satisfy the matching relationship to deceive the distance and speed tracker at the same time. From the perspective of the distance-speed coupling relationship, the distance change rate and speed of the real target are consistent, which can be expressed as:
[0073]
[0074] Where T is the time interval between adjacent interference phases, R k and Rv k They represent the distance and radial velocity of the real target relative to the radar in the kth jamming stage respectively;
[0075] Similarly, the false target should satisfy the same constraints, namely:
[0076]
[0077] in, and They represent the distance and radial velocity of the false target relative to the radar in the kth jamming stage respectively; therefore, equation (15) is further rewritten as:
[0078]
[0079] Substituting equation (15) into equation (17), we get the matching relationship between the controlled delay and Doppler frequency shift:
[0080]
[0081] where m(· ) represents the matching function between the controlled delay and the Doppler shift.
[0082] Step 4, RVGPO interference strategy optimization; for the optimization of RVGPO interference strategy, based on the matching function, the present invention proposes the IPSO-OCBA (Improved Particle Swarm Optimization-Optimal Computational Budget Allocation) algorithm. The IPSO-OCBA algorithm mainly includes three parts: PSO optimizer, particle update based on matching relationship, and computational budget allocation scheme based on OCBA. The PSO algorithm controls a group of particles to explore the entire solution space and generate candidate interference strategies for performance comparison. In order to reduce the impact of random factors on the performance comparison of candidate strategies, the OCBA scheme is used to allocate a limited simulation budget. This method allocates more budget to more promising candidate strategies, thereby maximizing the probability of the algorithm selecting the optimal strategy. The particle update based on the matching relationship uses the matching function to improve the particle update mechanism in the PSO algorithm. The main process of the IPSO-OCBA algorithm proposed in the present invention is as follows:
[0083] First, randomly initialize N particles to obtain the initial position vector (i.e., candidate interference strategy)
[0084] and the initial velocity vector Equation (13) is used to evaluate the comprehensive interference performance of all candidate strategies.
[0085] Secondly, the particle position and velocity are updated based on PSO. The formula is as follows:
[0086]
[0087] in, and denote the position and velocity of the nth particle in the gth iteration, respectively. Gbest represents the individual optimal position of the nth particle at the g-1th iteration, that is, the best candidate interference strategy found by the nth particle. g-1 represents the global optimal position at the g-1th iteration, that is, the optimal candidate interference strategy found by all particles; c1 and c2 are set to 1.49445, δ1 and δ2 are random numbers between [0,1]; α g is the inertia weight for velocity update, which is generally set to decrease linearly from 0.9 to 0.4 as the number of iterations g increases, i.e.:
[0088]
[0089] Among them, g max Indicates the maximum number of iterations;
[0090] Then, the particle position is updated based on the matching relationship, and the candidate interference strategy is updated using the matching function of formula (18), which is as follows:
[0091]
[0092] in, and represents the delay strategy and frequency shift strategy of the nth particle in the gth iteration after updating through the matching function; β g represents the matching weight, which is set to decrease linearly with the number of iterations:
[0093]
[0094] Among them, β max and β min are the maximum and minimum values of the matching weight respectively;
[0095] Next, the OCBA scheme is used to allocate the limited computational budget J to all updated candidate strategies Thus, the comprehensive interference performance of all updated strategies is calculated In each allocation stage, the budget increment is allocated and calculated according to equations (24) and (25):
[0096]
[0097] in, represents the total computational budget allocated to the nth particle until the lth allocation stage; b represents the serial number referring to the best particle, that is, the particle with the highest sample mean; and They represent the sample mean and sample variance of the nth particle at the l-1th allocation stage, respectively. The calculation formula is as follows:
[0098]
[0099] in, Represents candidate strategies The interference result at the jth calculation;
[0100] Finally, the current individual optimal particle Pbest is updated according to the result of OCBA g And the global optimal solution Gbest g ; When the maximum number of iterations g is reached max When , the optimization process stops, and the global optimal solution is It is the optimized optimal RVGPO interference strategy.
[0101] The effect of the present invention is further illustrated by the following simulation comparison test:
[0102] Simulation scenario: Assuming the target moves at a constant speed, the radar uses the extended Kalman filter-Doppler assisted integrated probability data interconnection tracking method, and the tracking model is a constant speed model. The specific parameters of the radar and the target are shown in Table 1.
[0103] Table 1
[0104]
[0105]
[0106] Based on the above simulation scenario, the proposed IPSO-OCBA algorithm is used for optimization, and the controlled delay increment range in each stage is Δτ k ∈(0,1.33μs), the controlled Doppler frequency shift increment range is Δf k ∈(-60,60Hz), the main parameters of the algorithm are as follows: number of particles N = 40, maximum number of iterations g max =200; the initial computational budget of each candidate strategy is J0 = 0.015J, and the additional computational budget of each allocation stage is Δ = 0.1J; the weighted weights of the two indicators are w1 = 0.5, w2 = 0.5 / d K,max , where d K,max =cτ K,max / 2, τ K,max =K· 1.33μs represents the maximum delay of the Kth stage.
[0107] In order to avoid the influence of random factors, in the following simulation experiments, the optimization algorithm was repeated 50 times independently to obtain 50 RVGPO interference strategies. The comprehensive interference performance of each strategy They are estimated from 10,000 independent repeated experiments, and the average interference performance of 50 RVGPO strategies is used as the final performance of the algorithm.
[0108] Figure 4 It is the iterative convergence curve of IPSO-OCBA algorithm under different matching weight settings. At this time, the maximum number of interference stages K = 15, and the total calculation budget J = 10000. In order to determine the optimal setting of matching weight, the following three weight settings are compared: Setting 1) β min = 0.1 and β max =0.9; Setting 2)β min = 0.8 and β max =0.9; Setting 3)β min = 0.1 and β max =0.2. Figure 3It can be seen that the algorithm achieves the best performance when setting 1. This is because the matching relationship in the initial stage of the algorithm is more reliable than the PSO optimizer. As the number of iterations increases, the matching relationship gradually guides the algorithm's search area to the matching strategy space, which continuously shrinks the search area. At this time, the reliability of the PSO optimizer gradually increases. Therefore, in the following simulations, the proposed IPSO-OCBA algorithm adopts the matching weight in setting 1.
[0109] Next, in order to demonstrate the role of the particle update mechanism based on matching relationship, the IPSO-OCBA algorithm proposed in the present invention is compared with the PSO-OCBA (Particle Swarm Optimization-Optimal Computational Budget Allocation) algorithm. Figures 5 to 7 These are the iterative convergence curves of the two algorithms when J = 10000, K = 15, J = 5000, K = 15, and J = 10000, K = 20. Figure 5 and Figure 6 It can be found that the performance of both algorithms will decrease as the total computational budget decreases. When the total computational budget is reduced from 10,000 to 5,000, the performance of the IPSO-OCBA algorithm proposed in the present invention only decreases by 6.3%, while the performance of the conventional PSO-OCBA algorithm decreases by 33.7%. This shows that the IPSO-OCBA algorithm proposed in the present invention can better utilize the limited computational budget through the update mechanism based on the matching relationship.
[0110] contrast Figure 5 and Figure 7 It can be found that the performance of both algorithms will decrease with the increase of the total number of interference stages, because more stages mean a larger solution space and higher optimization difficulty. When the total number of interference stages increases from 15 to 20, the performance of the IPSO-OCBA algorithm proposed in the present invention only decreases by 4.8%, while the performance of the conventional PSO-OCBA algorithm decreases by 83.3%. This is because the update mechanism based on the matching relationship can greatly reduce the exploration scope of the algorithm, so the change in dimension has little effect on the performance of the IPSO-OCBA algorithm proposed in the present invention.
[0111] Finally, in order to better verify the performance of the proposed algorithm, the following four different interference methods are compared: the optimal uniform RVGPO interference strategy obtained after traversal (called "UV-RVGPO (Uniform Velocity Range-Velocity GatePull-Off)"), the optimal uniform acceleration RVGPO interference strategy obtained after traversal (called "UA-RVGPO (UniformAcceleration Range-Velocity Gate Pull-Off)"), the RVGPO interference strategy optimized by the proposed IPSO-OCBA algorithm (called "Opt-RVGPO"), and the RGPO (Range GatePull-Off) interference strategy optimized by the PSO-OCBA algorithm (called "Opt-RGPO"). Among them, the traversal range of the uniform RVGPO interference towing velocity is [1m / s, 100m / s]; the traversal range of the uniform acceleration RVGPO interference towing initial velocity is [1m / s, 30m / s], and the traversal range of the towing acceleration is [1m / s 2 ,40m / s 2 ]. Figure 8 It is the comprehensive interference performance of the above four interference methods at different stages when J=10000 and K=15.
[0112] Depend on Figure 8 It can be found that the performance of the RGPO jamming strategy is close to zero, because it does not take into account the speed information and thus cannot deceive the radar that performs distance-speed joint tracking. Among the other three jamming strategies, the optimized RVGPO jamming strategy performs better than the UV-RVGPO and UA-RVGPO jamming strategies, and its advantages are mainly reflected in the following two aspects:
[0113] 1) Interference takes effect faster: The RVGPO interference strategy optimized by the proposed algorithm can take effect from the fourth stage, which is two stages earlier than the UV-RVGPO interference strategy and takes effect at the same stage as the UA-RVGPO interference strategy.
[0114] 2) Better interference effect: The final comprehensive interference performance of the RVGPO interference strategy optimized by the proposed algorithm is much higher than that of the other two strategies, which is 94.5% higher than that of the UV-RVGPO interference strategy and 73.9% higher than that of the UA-RVGPO interference strategy.
[0115] It can be seen from the above simulation that the present invention can well achieve quantitative optimization of the RVGPO interference strategy.
[0116] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.
Claims
1. A RVGPO interference method based on random simulation optimization, It is characterized in that The steps include: Step 1: Modeling of typical RVGPO interference scenarios; Assume that the controlled delay and Doppler shift of the kth interference stage are τ k and f k , then the distance d between the false target and the real target k and speed v k They are: Where K represents the total number of interference stages, c represents the speed of light, λ represents the signal wavelength, and Δτ i and Δf i They represent the controlled time delay and Doppler shift increment in each interference stage respectively; Assume that the state equation of the real target is: X(k)=F[k,X(k-1)]+U[k,X(k-1)] (3) Among them, F[·] represents the state transition matrix of the real target motion state in adjacent stages; U[·] represents the process noise matrix; X(k)=[X e (k),X v (k)] T It represents the motion state of the real target in the kth interference stage, which includes the following two parts: target position X e (k) and the Doppler velocity X v (k); Therefore, the motion state of the false target in the kth interference stage is for: Among them, Θ(·) and Ψ() represent the distance d k and speed v k Projection function into state space; At this time, the radar will measure the position and Doppler velocity of the target; the measurement of the target position is called kinematic measurement, and the measurement of the Doppler velocity is called Doppler measurement, which is expressed as: Where Z(k)=[Z e (k),Z v (k)] T represents the k-th jamming stage radar's kinematic and Doppler measurements of the real target; h e [·] and h v [] indicates measurement function; W e [·] and W v [] represents the measurement noise matrix; The radar's measurement value of the false target is expressed as: in, represents the k-th jamming stage radar's kinematic and Doppler measurements of false targets; In the kth jamming stage, after the radar obtains the real target and false target measurements, it first predicts the target state: Among them, X(k|k-1) represents the one-step prediction of the target state by the radar tracking system; represents the estimation of the target state by the radar tracking system in the k-1th jamming stage, and f[·] represents the one-step prediction function; Then, select the measurements that are likely to come from the real target: Among them, g e [·] and g v [] represents the kinematic and Doppler measurement selection functions; radars are represented by h e (k,X(kk-1)) and Range and velocity tracking gates are established for the center to filter kinematic and Doppler measurements, where is the temporary target state obtained by applying the standard Kalman filter to the position measurement. and denote the kinematic and Doppler measurements that have been selected, respectively; only the measurements selected by both (8) and (9) are used by the radar to update the target state, so the selected measurement set satisfies Finally, update the target's state estimate: Among them, φ[·] represents the fusion function; Step 2: Establish RVGPO interference strategy optimization model; define track interruption conditions And the error distance for: Among them, r 1 (·) and r 2 () represents the objective function; represents the true target state of all K interference stages, represents the positional and Doppler measurement noise of all K interference stages, Indicates RVGPO interference strategy; It indicates that the radar's tracking of the real target has not been interrupted; Indicates that the radar's tracking of the real target has been interrupted; represents the state estimate of the target in the last interference phase, ||·|| 2 is the matrix two norm; Taking equations (11) and (12) as examples, we can obtain two indicators for evaluating the RVGPO interference effect: track interruption rate and the average error distance Where E[·] represents the expectation. The above two evaluation indicators are weighed by weighted aggregation, and the comprehensive interference performance of the RVGPO interference strategy is obtained as follows: Among them, w 1 >0 and w 2 >0 represents the weighted aggregation weights of the two indicators, track interruption rate and average error distance; Taking formula (13) as the optimization objective, the RVGPO interference strategy optimization model is established as follows: in, and Respectively represent the time delay and frequency shift strategies; Ω τ and Ω f represents the solution space of the RVGPO interference strategy; Step 3: Derive the matching relationship between the controlled delay and the Doppler frequency shift; the distance change rate and speed of the real target are consistent, expressed as: Where T is the time interval between adjacent interference phases, R k and Rv k They represent the distance and radial velocity of the real target relative to the radar in the kth jamming stage respectively; Similarly, the false target should satisfy the same constraints, namely: in, and They represent the distance and radial velocity of the false target relative to the radar in the kth jamming stage respectively; therefore, equation (15) is further rewritten as: Substituting equation (15) into equation (17), we get the matching relationship between the controlled delay and Doppler frequency shift: Where m(·) represents the matching function between the controlled delay and Doppler frequency shift; Step 4: Optimize the RVGPO interference strategy; the specific method is: first, randomly initialize N particles to obtain the initial position vector and the initial velocity vector Equation (13) is used to evaluate the comprehensive interference performance of all candidate strategies. Secondly, the particle position and velocity are updated based on PSO. The formula is as follows: in, and denote the position and velocity of the nth particle in the gth iteration, respectively. Gbest represents the individual optimal position of the nth particle at the g-1th iteration, that is, the best candidate interference strategy found by the nth particle. g-1 represents the global optimal position at the g-1th iteration, that is, the optimal candidate interference strategy found by all particles; c 1 and c 2 Set to 1.49445, δ 1 and δ 2 is a random number between [0,1]; α g is the inertia weight for velocity update, which is generally set to decrease linearly from 0.9 to 0.4 as the number of iterations g increases, i.e.: Among them, g max Indicates the maximum number of iterations; Then, the particle position is updated based on the matching relationship, and the candidate interference strategy is updated using the matching function of formula (18), which is as follows: in, and represents the delay strategy and frequency shift strategy of the nth particle in the gth iteration after updating through the matching function; β g represents the matching weight, which is set to decrease linearly with the number of iterations: Among them, β max and β min are the maximum and minimum values of the matching weight respectively; Next, the OCBA scheme is used to allocate the limited computational budget J to all updated candidate strategies Thus, the comprehensive interference performance of all updated strategies is calculated In each allocation stage, the budget increment is allocated and calculated according to equations (24) and (25): in, represents the total computational budget allocated to the nth particle until the lth allocation stage; b represents the serial number referring to the best particle, that is, the particle with the highest sample mean; and They represent the sample mean and sample variance of the nth particle at the l-1th allocation stage, respectively. The calculation formula is as follows: in, Represents candidate strategies The interference result at the jth calculation; Finally, the current individual optimal particle Pbest is updated according to the result of OCBA g And the global optimal solution Gbest g ; When the maximum number of iterations g is reached max When , the optimization process stops, and the global optimal solution is It is the optimized optimal RVGPO interference strategy.
