Studies On Intelligent Detection And Recognition Technology Of LPI Radar Signals

In the current complex battlefield environment,the LPI radar has been widely used due to its characteristics of low power,frequency agility,large bandwidth and complex modulation signals.Traditional radar reconnaissance methods are difficult to effectively detect and recognize the LPI radar signals.In order to solve the problem of the LPI radar signal detection and recognition,this thesis combines the artificial intelligence to develop an intelligent detection and recognition technology for the LPI radar signals.This thesis analyzes the LPI radar signal characteristics,typical work patterns and waveform parameters.On this basis,the radar signal intelligent detection method based on the BP neural network is studied.Starting from the time-frequency characteristics of the signals,the recognition of the intra-pulse modulation of the LPI radar signals is completed by the convolutional neural network;Finally,based on the SMOTE algorithm,the recognition of the radar work patterns is completed by the BP neural network and the effectiveness of the algorithm is verified.The main work of this thesis are:1.Research on the intelligent detection method of the LPI radar signals.Firstly,the conventional radar signal detection methods and its defects are analyzed.On this basis,the single-point correlation signal detection method is extended.The effects of the correlation accumulation points,the noise sampling points and the signal sampling points on the signal single-point correlation waveform are analyzed.Finally,the BP neural network is used to recognize the single-point correlation waveform to achieve effective detection of LPI radar signals.2.Study on the intelligent intra-pulse modulation recognition of the LPI radar signals.Based on the analysis of the intra-pulse characteristics of radar signals,the shortcomings of the conventional recognition methods are analyzed.The recognition of the intra-pulse modulation of the LPI radar signals is completed by the convolutional neural network;The influence of the convolutional neural network structure and parameters on the recognition performance is studied.Finally,the effectiveness of the method is verified by simulation.3.The LPI radar work pattern intelligent recognition method is studied.Firstly,the typical work patterns and the recognition parameters of LPI radar are analyzed.The radar working pattern recognition algorithm based on the BP neural network is proposed,and the influence of the radar parameters missing on the algorithm is simulated and analyzed.On this basis,the impact of the training samples imbalance caused by the new radar work pattern on the performance of the BP neural network is simulated and analyzed.In order to solve this problem,the BP neural network radar work pattern recognition algorithm based on the SMOTE algorithm is presented and its effectiveness is verified by simulation.