Based On Neural Network Effectiveness Evaluation Of Radar Jamming

Nowadays,with the rapid development of radar technology,radar countermeasure technology has also developed.As an important part of cognitive electronic warfare,the method for evaluating interference effects must have rapid and intelligent capabilities.Compared with the introduction of too many human factors defects in the traditional radar jamming effect evaluation method,the intelligent evaluation method has become the main research direction of radar jamming effect evaluation.For the evaluation of radar interference effect,this thesis mainly studies the evaluation method based on Radial Basis Function(RBF)neural network algorithm and the evaluation method based on Particle Swarm Optimization Back Propagation(PSO-BP)neural network algorithm.The main contents are as follows.Firstly,introduce the radar countermeasure technology,which includes radar reconnaissance technology,radar jamming technology and radar anti-jamming technology.By analyzing the radar jamming pattern and radar anti-jamming method,the radar data characteristics are obtained.Secondly,aiming at the problem of selecting the evaluation index,combining the interference effect evaluation construction criterion and the interference effect evaluation combat skill index,selecting the radar parameter set that can reflect the change of the radar working mode,so as to analyze and obtain the interference evaluation index vector.Then,in the case of insufficient radar parameter information,a radar interference effect evaluation method based on the RBF neural network algorithm is proposed.The RBF neural network is used to process the regression problem with low time and complex algorithm performance,and the index set and the final evaluation result are mapped.Finally,under the condition of sufficient radar processing parameters,a radar interference effect evaluation method based on PSO-BP neural network algorithm is proposed.The PSO algorithm is used to update the weights and thresholds of the BP neural network.According to the updated BP neural network,the evaluation vector is used to evaluate the interference effect.By simulating the evaluation method based on RBF neural network algorithm,the effectiveness of the method is verified under the conditions that the amount of radar parameter information data is small and the error is certain.By simulating the evaluation method of PSO-BP neural network algorithm,it is verified that the algorithm performs well on the evaluation problem when the amount of radar information data is sufficient.The simulation compares the relatively single BP neural network with the PSO-BP neural network.It is verified that after the PSO algorithm optimizes the weights and threshold initialization of the BP neural network,the performance of the neural network has been significantly improved,and the evaluation results are more stable and accurate.The simulation compares the RBF neural network algorithm and the PSO-BP neural network algorithm,and verifies that the evaluation result of the PSO-BP neural network algorithm is more accurate when the amount of data is sufficient,and the performance performance is stable,which is conducive to processing offline.Radar information is sufficient.in the case of a small amount of data,the RBF neural network’s evaluation results obtained within a certain range of error conditions are shorter,which is beneficial to deal with the lack of radar information in real time.