Electromagnetic Signal Perception And Design And Simulation Of Intelligent Interference

With the electromagnetic environment of modern electronic warfare becoming more and more complex,the rapid development of intelligent modern radar makes the working mode and transmitted signal of radar become diverse.The traditional radar jamming decision-making method is not enough to make jamming decision quickly for today’s electronic countermeasure task.Therefore,the research demand of radar cognitive jamming is becoming more and more urgent.Aiming at the realization of intelligent jamming of jammer to target radar,this thesis studies and simulates the radar intelligent jamming method based on reinforcement learning.The simulation results show that it is a possible way to combine reinforcement learning algorithm with jammer jamming decision-making process,and verify that reinforcement learning algorithm has a good effect on guiding jammer to make correct decision.The main work and contributions of this thesis are as follows:1.The parameters of the radar signal received by the receiver are estimated,including the frequency,TOA,DOA,PRI and other parameters of the radar signal.At the same time,the received pulse signal is sorted and processed.The estimation of relevant parameters and signal sorting algorithm are simulated,and the required signal information can be obtained correctly according to the algorithm.2.The in pulse modulation information of the received signal is identified.At the same time,the recognition of radar working state is studied based on the estimation results of target radar signal parameters.The main methods used are dynamic clustering method and support vector machine method.The simulation results show that dynamic clustering algorithm and support vector machine algorithm have high accuracy for the recognition of radar working state.3.The typical active jamming of radar is studied and simulated,the radar active jamming style library is established,the possible behavior of radar after jamming is analyzed,the on-line evaluation index and criterion of jamming effect are given,and the on-line evaluation model of jamming effect based on neural network is established.The simulation verifies that the neural network model is used to evaluate the radar signal parameters The working state and interference pattern are regressed to obtain the effectiveness of the interference effect evaluation results.4.Aiming at the problem of radar intelligent jamming decision-making,the reinforcement learning algorithm is introduced,and the radar intelligent jamming decision-making model based on reinforcement learning is established.The radar intelligent jamming decision-making process based on Q-Learning and dqn algorithm is studied.The key components such as action space,state space and reward function in reinforcement learning are designed,and the model is simulated and verified.The results show that the radar intelligent jamming decision-making model based on reinforcement learning can correctly make decisions according to the information such as radar working state and the distance between radar and jammer in different ECM scenarios,The validity of the model is verified.