Intermittent sampling phase modulation jamming waveform method generated by particle swarm algorithm

Abstract

The application discloses an intermittent sampling phase modulation jamming waveform method based on a particle swarm algorithm, and aims to solve the defect that a target function obtained by an intermittent sampling phase modulation repeating technology in the prior art is prone to falling into a local optimal solution, and thus leading to poor jamming parameters obtained when an intermittent sampling waveform is generated. The implementation steps of the application comprise the following steps: generating a sample set composed of three different radar signals; generating a target function of a particle swarm; optimizing the jamming parameter code element width and sampling time of the intermittent sampling jamming signal by using the particle swarm algorithm; and generating the corresponding intermittent sampling phase modulation jamming signal by using the optimized optimal jamming parameter. The application has the advantages that an interference waveform library does not need to be established, and the optimal intermittent sampling phase modulation jamming waveform can be obtained by using the particle swarm algorithm.

Description

Technical Field

[0001] This invention belongs to the field of radar communication technology, and further relates to a particle swarm optimization (PSO) algorithm-based intermittent sampling phase modulation (PMMO) interference waveform method in the field of electronic countermeasures. This invention can be used to obtain a superior phase modulation interference waveform when the interference signal is an intermittently sampled PMMO waveform. Background Technology

[0002] With technological advancements, various new types of radar are constantly emerging, and their intelligence levels are continuously improving. Traditional radar countermeasures primarily rely on pre-programmed countermeasures based on prior target feature libraries. This approach struggles to cope with increasingly sophisticated radar anti-jamming measures, leading to ever-higher demands on radar countermeasure equipment in terms of time and accuracy. Therefore, optimizing the generation of jamming waveforms with low latency and high accuracy is crucial in electronic warfare.

[0003] In their paper "Intelligent Optimization of Phase Modulation Waveform Based on Genetic Algorithm" (Systems Engineering & Electronics, 2022(003):044), Sun Jianwei et al. proposed an intermittent sampling phase modulation jamming waveform method based on an intelligent optimization algorithm for radar countermeasure waveform intelligent optimization model. The implementation steps of this method are as follows: A waveform optimization module is added to the traditional radar countermeasure system. This module can establish a virtual radar within the system based on the received radar signal to simulate the receiving process of the opposing radar, and thereby evaluate the effectiveness of the countermeasure waveform synthesized with the current parameters. Simultaneously, guided by the evaluation results, the intelligent optimization algorithm is used to optimize the waveform parameters, thereby improving the countermeasure performance of the transmitted waveform. The drawback of this method is that it is prone to getting trapped in local optima when generating intermittent sampling jamming waveforms, resulting in poor jamming parameters and a poor jamming effect on the radar.

[0004] Hangzhou Dianzi University proposed an interference waveform generation method in its patent application "Method and Apparatus for Generation and Evaluation of Interference Waveforms with Action Sets through Reinforcement Learning" (Application No.: 202210875262.7; Publication No.: CN115378774A). The implementation scheme of this method is as follows: establishing an interference waveform library; constructing an interference waveform decision network, including a policy network and a value network; using a tracking algorithm to perform interference during the exploration phase to fill the experience pool; randomly selecting a set of data from the experience pool; predicting the corresponding interference action at the next moment based on the interference action of the current state through the policy network; generating discretized interference actions based on the interference actions output by the policy network; selecting interference actions from the interference waveform library and combining them with the interference actions output by the policy network, inputting them together into the value network to obtain the interference action with the highest value as the actual interference action. The drawback of this method is that it can only generate interference signals based on the interference waveform library. If a radar signal that does not exist in the interference waveform library is encountered, the corresponding interference waveform cannot be generated. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for generating intermittent sampling jamming signals for radar countermeasures using a particle swarm optimization algorithm. This solves the problems of low accuracy and poor jamming effect in existing technologies, as well as the inability to generate jamming waveforms outside the existing jamming waveform library.

[0006] The idea behind this invention is to use a particle swarm optimization (PSO) algorithm, replacing constant weights with nonlinear adaptive inertia weights. This allows the PSO algorithm to initially tend towards exploration and then gradually shift towards expansion, thus adjusting the solution in local regions. This avoids the problem of intermittent sampling interference signal generation easily getting trapped in local optima when the inertia weights are too small or too large. By setting different radar signals, such as linear frequency modulated signals and binomial coded signals, and setting other experimental parameters such as false alarm rate and pseudo-random sequences, and then using the PSO algorithm to simultaneously optimize the sampling time and symbol width, the optimal intermittent sampling interference waveform for different radar signals can be obtained.

[0007] The specific steps to achieve the objective of this invention are as follows:

[0008] Step 1, Generate radar signal sample set:

[0009] The sample set consists of at least three types of radar signals, with each type containing 50 samples;

[0010] Step 2, the objective function for generating the particle swarm is as follows:

[0011]

[0012] Where N represents the total number of differences between the average threshold of the detection unit closest to the target and the peak value of the target echo pulse compression, ∑ represents the summation operation, lg represents the logarithmic operation to the base 10, and Z represents the threshold height of the detection unit where the target is located. P fa P represents the false alarm probability, N1 represents the number of single-sided protection units of the radar signal, N2 represents the number of average units of the radar signal, and P represents the false alarm probability. jpc (r) represents the phase modulation signal power, which is the average value of the phase modulation signal power in the r-th rightmost detection unit, with the detection unit centered at t=0. epc (0) represents the output signal of the radar matched filter at t=0, x epc (t)=x e (t)*x e '(-t), x epc (t) represents the radar target echo signal x. e (t) The signal output after the matched filter, x e '(-t) represents x e The complex signal (t) is represented by *, where * indicates the convolution operation;

[0013] Step 3: Optimize the interference parameters of the intermittently sampled interference signal using the particle swarm optimization algorithm:

[0014] Step 3.1: Initialize the particle swarm parameters to c1 and c2, set the position boundary range and velocity boundary range, initialize the particle swarm, and initialize the particle swarm velocity.

[0015] Step 3.2: Calculate the fitness value of each particle in the current iteration and record the historical best value P. best With the global optimal value g best ;

[0016] Step 3.3, use the velocity update formula v i (t+1)=v i (t)+c1r1(P best (t)-x i (t))+c2r2(g best -x i (t) updates the velocity of the particle swarm and constrains velocities that exceed the limit, v i (t+1) represents the update rate of the i-th particle at iteration t+1, v i (t) represents the velocity of the i-th particle at iteration t, x i (t) represents the position of the i-th particle in the t-th iteration, ω represents the inertia factor of the system, and r1 and r2 represent two independent random numbers with values ​​in the range [0,1].

[0017] Step 3.4, use the position update formula x i (t+1)=x i (t)+v i (t+1) Update the positions of the particle swarm and constrain out-of-bounds positions, x i (t+1) represents the update rate of the i-th particle in the (t+1)-th iteration;

[0018] Step 3.5: Update the fitness value of each particle with the updated velocity and position;

[0019] Step 3.6: If the updated fitness value of each particle is greater than its fitness value in the previous iteration, then this fitness value is taken as the historical best value P after the current iteration. best If the fitness value is less than the fitness value of the particle in the previous iteration, then the original fitness value is retained as the historical best value P. best ;

[0020] Step 3.7: If the updated fitness value of each particle is greater than the fitness value experienced by the group in the previous iteration, then update the current fitness value to the global optimum g. best If the fitness value is less than the fitness value experienced by the population in the previous iteration, then the original fitness value is retained as the global optimum g. best ;

[0021] Step 3.8: Determine whether the current iteration count has reached the maximum number of iterations. If yes, output the optimal objective function, use the symbol width and sampling time corresponding to the objective function as the optimal intermittent sampling phase modulation signal interference parameters, and then execute step 4. If no, execute step 3.2.

[0022] Step 4: Generate an interference signal using the optimized interference parameters.

[0023] Compared with the prior art, the present invention has the following advantages:

[0024] First, the particle swarm optimization algorithm used in this invention obtains the optimal interference parameters when generating intermittent sampling interference signals. This solves the problem that existing intermittent sampling interference waveform generation techniques are prone to getting trapped in local optima, which prevents them from obtaining the optimal interference parameters and results in poor interference effects on radar. Therefore, the advantage of this invention is that the generated intermittent sampling interference signals have high accuracy in interfering with radar and good interference effects.

[0025] Secondly, this invention does not require the establishment of an interference waveform library, which solves the problem that radar signals other than those in the interference waveform library cannot be optimized and generated for interference waveforms. It can generate intermittent sampling interference waveforms simply by obtaining radar information from the interfering party, giving this invention the advantage of generating intermittent sampling interference signals without the need to establish an interference waveform library. Attached Figure Description

[0026] Figure 1 This is a flowchart of the present invention;

[0027] Figure 2 This is a graph showing the particle swarm optimization results for symbol width and sampling time. Detailed implementation method:

[0028] The present invention will now be further described with reference to the accompanying drawings and embodiments.

[0029] Reference Figure 1 The implementation steps of the embodiments of the present invention will be described in further detail below.

[0030] Step 1: Generate a radar signal sample set.

[0031] In the embodiments of the present invention, a sample set is composed of three types of radar signals, with 50 samples of each type of radar signal, for a total of 150 radar signals.

[0032] The three types of radar signals are linear frequency modulation (FM) signals, phase-coded signals, and nonlinear frequency modulation (FM) signals.

[0033] Step 2: Optimize the interference parameters of the intermittently sampled interference signal using the particle swarm optimization algorithm.

[0034] Set the symbol width T, the parameter to be optimized for intermittent sampling interference signals. c and sampling time τ, where symbol width T c The value range is 0.1–4 μs with an accuracy of 0.01 μs, and the intermittent sampling time τ ranges from 1–10 μs with an accuracy of 0.5 μs.

[0035] Using the following objective function, select the optimal symbol width T for the intermittently sampled interference signal. c And sampling time τ:

[0036]

[0037] Where N represents the total number of differences between the average threshold of the detection unit closest to the target and the peak value of the target echo pulse compression, ∑ represents the summation operation, lg represents the logarithmic operation to the base 10, and Z represents the threshold height of the detection unit where the target is located. P faP represents the false alarm probability, N1 represents the number of single-sided protection units of the radar signal, N2 represents the number of average units of the radar signal, and P represents the false alarm probability. jpc (r) represents the phase modulation signal power, which is the average value of the phase modulation signal power in the r-th rightmost detection unit, with the detection unit centered at t=0. epc (0) represents the output signal of the radar matched filter at t=0, x epc (t)=x e (t)*x e '(-t), x epc (t) represents the radar target echo signal x. e (t) The signal output after the matched filter, x e '(-t) represents x e The complex signal (t) is represented by *, where * denotes the convolution operation.

[0038] Step 3: Optimize the interference parameters of the intermittently sampled interference signal using the particle swarm optimization algorithm.

[0039] Step 3.1: Initialize the particle swarm parameters to c1 and c2, set the position boundary range and velocity boundary range, initialize the particle swarm, and initialize the particle swarm velocity.

[0040] Step 3.2: Calculate the fitness value of each particle in the current iteration and record the historical best value P. best With the global optimal value g best .

[0041] Step 3.3: Update the velocity of the particle swarm using the following velocity update formula:

[0042] v i (t+1)=v i (t)+c1r1(P best (t)-x i (t))+c2r2(g best -x i (t))

[0043] Among them, v i (t+1) represents the update rate of the i-th particle at iteration t+1, v i (t) represents the velocity of the i-th particle at iteration t, x i (t) represents the position of the i-th particle in the t-th iteration, ω represents the inertia factor of the system, and r1 and r2 represent two independent random numbers with values ​​in the range [0,1].

[0044] Step 3.4: Update the positions of the particle swarm using the position update formula:

[0045] x i (t+1)=x i (t)+v i (t+1)

[0046] Where, x i (t+1) represents the update speed of the i-th particle in the (t+1)-th iteration.

[0047] Step 3.5: Update the fitness value of each particle with the updated velocity and position.

[0048] Step 3.6: If the updated fitness value of each particle is greater than its fitness value in the previous iteration, then this fitness value is taken as the historical best value P after the current iteration. best If the fitness value is less than the fitness value of the particle in the previous iteration, then the original fitness value is retained as the historical best value P. best .

[0049] Step 3.7: If the updated fitness value of each particle is greater than the fitness value experienced by the group in the previous iteration, then update the current fitness value to the global optimum g. best If the fitness value is less than the fitness value experienced by the population in the previous iteration, then the original fitness value is retained as the global optimum g. best .

[0050] Step 3.8: Determine if the current iteration count has reached the maximum number of iterations. If yes, output the optimal objective function, and use the symbol width and sampling time corresponding to the objective function as the optimal intermittent sampling phase modulation signal interference parameters before proceeding to step 4. If not, proceed to step 3.2. The maximum number of iterations refers to the number of iterations corresponding to the convergence of the objective function.

[0051] The following is combined with Figure 2 The results of parameter optimization of the intermittently sampled phase-modulated interference signal using the particle swarm optimization algorithm in step 3 are further described. Figure 2 This is a three-dimensional coordinate graph, where the x-axis represents the symbol width in μs, the y-axis represents the sampling time in μs, and the z-axis represents the objective function value (Loss). Figure 2 It contains the values ​​of the Loss of all objective functions generated within the corresponding intervals of symbol width and sampling time. When the optimization is complete, all the particles can be seen to gather in... Figure 2 The objective function is to maximize the loss. The symbol width and sampling time corresponding to the loss are used as the optimal interference parameters. The optimized symbol width and sampling time values ​​will vary depending on the radar signal being processed.

[0052] Step 4: Generate an interference signal using the optimized interference parameters.

[0053] The optimal interference parameters selected in step 3, symbol width and sampling time, are input into the intermittent sampling interference signal formula to generate the corresponding intermittent sampling interference signal. This signal is the optimal interference signal for the radar corresponding to step 1.

[0054] The effectiveness of this invention can be further demonstrated through the following simulation.

[0055] 1. Simulation experimental conditions.

[0056] The hardware platform for the simulation experiment of this invention is: an Intel i7 7700k CPU with a main frequency of 3.2GHz and 16GB of memory.

[0057] The software platform for the simulation experiment of this invention is Windows 10 operating system and Matlab R2019b.

[0058] 2. Simulation content and result analysis:

[0059] The simulation experiment of this invention uses this invention and two existing technologies (ergodic method and genetic algorithm) to optimize the parameters in the intermittently sampled interference signal waveform and generate the optimized interference waveform.

[0060] The existing ergodic method refers to the traditional mathematical ergodic method. For intermittent sampling interference signals, the ergodic method means that for each symbol width and sampling time value of the signal, a corresponding interference signal will be generated. For example, with 10 symbol widths and 10 sampling times, the ergodic method will generate 10x10 intermittent sampling interference signals.

[0061] Prior art 2 refers to the paper "Intelligent Optimization of Phase Modulation Waveform Based on Genetic Algorithm" published by Sun Jianwei et al. (Systems Engineering and Electronics Technology, 2022(003):044), which proposes to use a genetic algorithm to optimize the parameters of intermittent sampling interference waveform and generate the corresponding interference waveform.

[0062] In the simulation experiment of this invention, the parameters of the particle swarm are set as follows: inertia parameter is 1, acceleration factor is 0.5 and 1.5, number of iterations is 500, particle swarm size is 50, and position and velocity are limited to the interval [-2,2] and [-0.5,0.5].

[0063] In the simulation experiment of this invention, the parameters of the genetic algorithm are set as follows: the size of the initial population is 50, the number of generations to terminate evolution is 500, the mutation probability is 0.5, and the crossover probability is 0.8.

[0064] The simulation experiment of this invention uses the method of this invention and two existing technologies. Since each radar has 50 sample sets, each technology obtains 50 objective function values, objective function calculation times, and running times. The 50 values ​​are averaged to obtain the final objective function value, objective function calculation times, and running time for each radar signal. The final results of the three are plotted in Table 1.

[0065] Table 1. Analysis of Optimization Results of Three Algorithms

[0066]

[0067] As shown in the table above, the ergonomic method, genetic algorithm, and particle swarm optimization algorithm were used to generate intermittent sampling interference waveforms for three different radar signals: linear frequency modulated (LFM) signals, phase-coded signals, and nonlinear frequency modulated (NF) signals. The statistical data includes the optimized objective function (in dB), the number of objective function calculations, and the running time (in seconds). Comparing the optimization results of the three methods for the same radar signal reveals that the method of this invention and the traditional ergonomic method can both obtain the optimal objective function, but the number of objective function calculations and the running time are significantly less than those of the ergonomic method. Although the genetic algorithm also has a short running time, it cannot obtain the optimal objective function compared to the method of this invention, and therefore cannot obtain the optimal interference parameters. Table 1 demonstrates that the method of this invention solves the problem of existing methods easily getting trapped in local optima, thus failing to obtain the optimal interference parameters, and it also has a shorter running time, meaning it can obtain intermittent sampling interference waveforms quickly and efficiently.

Claims

1. A method for generating intermittent sampling phase modulation interference waveforms using a particle swarm optimization algorithm, characterized in that, The optimal parameters of the intermittently sampled interference signal are obtained using the particle swarm optimization algorithm, and an intermittently sampled interference waveform composed of the optimal parameters is generated. The steps of this method are as follows: Step 1, Generate radar signal sample set: The sample set consists of at least three types of radar signals, with each type containing 50 samples; Step 2, the objective function for generating the particle swarm is as follows: ； in, This represents the total difference between the average threshold of the detection unit closest to the target and the peak value of the target echo pulse compression. This represents the summation operation. This represents a logarithmic operation with base 10. Indicates the threshold height of the detection unit where the target is located. , This represents the probability of a false alarm. Indicates the number of single-sided protection units for the radar signal. Indicates the number of units in the radar signal averaging. Let represent the phase modulation signal power. This signal power is calculated by dividing the detection unit into two parts with t=0 as the center position and recording the average phase modulation signal power in the r-th unit to the right. This represents the output signal of the radar matched filter at t=0. , Indicates the target echo signal of the radar The signal output after the matched filter express Complex signal, Indicates the convolution operation; Step 3: Optimize the interference parameters of the intermittently sampled interference signal using the particle swarm optimization algorithm: Step 3.1, initialize the particle swarm parameters as follows: and Set the position and velocity boundary ranges, initialize the particle swarm, and initialize the particle swarm velocity. Step 3.2: Calculate the fitness value of each particle in the current iteration and record the historical best values. With global optimum ; Step 3.3: Update the velocity of the particle swarm using the velocity update formula, and constrain velocities that exceed the limit. ,in, This represents the update speed of the i-th particle at iteration t+1. This represents the velocity of the i-th particle at iteration t. This represents the position of the i-th particle in iteration t. The inertia factor represents the system. and This represents two independent random numbers whose values ​​are in the range [0,1]. Step 3.4, using The position update formula updates the positions of the particle swarm and constrains out-of-bounds positions. This represents the position of the i-th particle at iteration t+1; Step 3.5: Update the fitness value of each particle with the updated velocity and position; Step 3.6: If the updated fitness value of each particle is greater than its fitness value in the previous iteration, then this fitness value is taken as the historical best value after the current iteration. If the fitness value is less than the fitness value of the particle in the previous iteration, then the original fitness value is retained as the historical best value. ; Step 3.7: If the updated fitness value of each particle is greater than the fitness value experienced by the population in the previous iteration, then update the current fitness value to the global optimum. If the fitness value is less than the fitness value experienced by the population in the previous iteration, then the original fitness value is retained as the global optimum. ; Step 3.8: Determine whether the current iteration count has reached the maximum iteration count. If yes, output the optimal objective function, use the symbol width and sampling time corresponding to the objective function as the optimal intermittent sampling phase modulation signal interference parameters, and then execute Step 4. If no, execute Step 3.

2. Step 4: Generate an interference signal using the optimized interference parameters.

2. The method for generating intermittent sampling phase modulation interference waveforms using particle swarm optimization algorithm according to claim 1, characterized in that, The three types of radar signals mentioned in step 1 refer to: linear frequency modulation signal, phase-coded signal, and nonlinear frequency modulation signal.

3. The method for generating intermittent sampling phase modulation interference waveforms using particle swarm optimization algorithm according to claim 1, characterized in that, The maximum number of iterations mentioned in step 3.8 refers to the number of iterations that occur when the objective function converges.

Images