Research And Implementation Of Radar Active Jamming Recognition Method

Radar plays a vital role in modern electronic warfare,and as the battle between electronic countermeasures and electronic resistance intensifies,the problem of difficult manual extraction of features for active radar interference in complex electromagnetic environments and low recognition rates under low JNR needs to be addressed urgently.In order to take effective countermeasures against radar active interference signals,it is necessary to accurately identify the category to which the interference signal belongs first,and then take effective anti-interference measures accordingly.To this end,this paper designs two interference recognition algorithms with good noise immunity and high recognition rate based on the pattern recognition algorithm of variational modal decomposition and the deep learning algorithm based on convolutional network and gated recurrent network to solve the above problems,in which the traditional pattern recognition methods generally have strong interpretability while the deep learning methods have relatively weak interpretability.The details of the study are as follows:(1)Several common radar active suppression and spoofing interferences are modeled and analyzed,and then several classical interference identification algorithms are introduced and the advantages and disadvantages of each algorithm are analyzed.(2)A radar active interference identification algorithm(SVMD＿PE＿Non Zeros)based on the sparrow search algorithm(SSA)with optimized variational modal decomposition(VMD)parameters and signal feature enhancement is proposed to address the problem of low identification rate due to unreasonable number of decomposition layers of the signal by the variational modal decomposition(VMD)algorithm.The experimental results show that the improved method can effectively improve the recognition rate compared with the unimproved variational modal decomposition(VMD)algorithm,and the signal enhancement method with non-zero processing also has good engineering application value in the field of radar signal processing,and the recognition accuracy of the algorithm reaches 98.18% in the full JNR range of-10～10d B.(3)To address the problem that traditional pattern recognition methods require complex preprocessing of data and have low recognition rate,an interference recognition algorithm is proposed based on Multiscale and Fully Convolutional Neural Network(MFCN)and Gated Recurrent Unit(GRU)in parallel with the interference recognition algorithm.This is an endto-end deep neural network model,in which the original time-domain sequence of the interference signal is input,and the fused features of the signal in time and space can be extracted without complex preprocessing of the data to achieve classification and recognition of the signal.The experimental results show that the recognition accuracy of the network gradually increases as the dry noise ratio gradually increases;the overall recognition rate of the network is 99.4% in the full JNR range of-10 to 10 d B,and the recognition accuracy is close to 100% when the JNR is above-6 d B.(4)The system is developed based on the interference model and deep learning interference identification algorithm for radar active interference simulation and identification.The development environment of the system is windows 10,and the development tools use the Appdesigner software design toolbox that comes with Matlab R2021 b.This toolbox can help users to design a beautiful and simple GUI more easily with the graphics system.The main functions of the system include: the simulation function of radar active interference signals,the identification function of interference signals and the identification function of dataset signals.The experimental results show that all the relevant functions of this system can operate normally.