Research On Radar Signal Reconnaissance Method And Its Simulation Software Design And Implementation

With the rapid development of information technology,the signal forms are more complex and diverse,radar signal reconnaissance is increasingly important in modern warfare,and even determines the success or failure of the war.Radar reconnaissance can obtain the enemy's radar parameters to infer the use of the enemy's radar,provide us with information and gain the initiative of the war and win the war,which also makes the identification and sorting of radar signals more important in radar reconnaissance.Therefore,this thesis studies the radar signal intrapulse identification and signal sorting,and completes the design and implementation of the radar reconnaissance simulation software.The main research work is as follows:(1)Aiming at the problem that the phase characteristic of the phase difference method is greatly affected by noise,the multi-phase difference method is studied to identify the radar intrapulse signal,and the problem of low recognition rate of the low signal-to-noise ratio radar signal is improved.In order to further improve the recognition rate,this thesis adopts the recognition method of deep residual network and aggregated residual network based on migration learning,and uses ResNet34,ResNet50 and ResNext50 network to identify the radar signal time-frequency image,which greatly improves the low signal-to-noise ratio.Compared with the lower signal recognition rate,and finally the complexity of the three networks is measured by the code running time,the ResNext50 network has the best recognition effect and the complexity difference is not large.(2)Aiming at the problem that the traditional pre-sorting and main sorting algorithms cannot sort the repetitive frequency jitter,the variable pulse width and the frequency agile signal,this thesis adopts the fuzzy clustering algorithm,which has high real-time performance,but cannot sort the parameters with relatively large changes.Large frequency agile signal.Then,multi-threshold fuzzy clustering is proposed,adding the angle of arrival parameter,and using the absolute value index method and absolute similarity to construct a fuzzy similarity matrix,and using the first-level arrival time difference check,it can sort repetition frequency jitter,variable pulse width and frequency agility Radar signals can be applied to more complex electromagnetic environments.(3)This thesis completes the design and implementation of radar reconnaissance simulation software.Firstly,the overall scheme is designed according to the simulation requirements,and the module generation and parameter setting are introduced by the method of modularization.Then complete the design and simulation verification of the radar signal generation module,space transmission module,beamforming module,signal detection and estimation module,signal sorting and identification module,and result evaluation module.Based on the radar reconnaissance simulation software,the robustness of the simulation software is improved.In this thesis,the aggregation residual network and multi-threshold fuzzy clustering are used to identify and sort signals,which improves the recognition rate of signals under low signal-to-noise ratio and the ability to sort complex signals.Then,the radar reconnaissance simulation software is designed and implemented.