基于盲源分离结合奇异谱分析的雷达多分量信号识别方法

打开文本图片集

中图分类号：TN971 文献标志码：A

Abstract：The array receives multiple radar signals in the same beam at the same time，which results in time domain aliasing.Itis difficult to perform signal detection and parameter measurement，resulting in difficult in recognizing signal modulation types.Aiming at the above problem，a radar multi-component signal recognition method based on blind source separation（BSS）combined with singular spectrum analysis is proposed. Firstly，the singular spectrum analysis is used to denoise the received array signal，and then the BSS method is used to separate the aliasing multi-component signal. Secondly，the time-frequency transform of the separated signalis caried out to obtain the time-frequency diagram of the signal. Finally，the time-frequency diagram is used as the input of the deep learning network to recognize the signal. Simulation results show that the average recognition rate of multi-component signal received in the same beam reaches 92.67% at 5 dB and the proposed method has a good recognition effect.

Keywords： blind source separation （BSS）；signal recognition； time-frequency analysis；deep learning

0 引言

随着各种新体制雷达的不断发展，电磁环境越来越复杂，侦察天线在同一时刻可能接收到多个脉冲信号，且各脉冲信号的时域、频域调制参数交叠，使得传统的基于脉宽、频率、到达角（directionofarrival，DOA）、脉冲重复间隔的识别方法难以适应复杂电磁环境，识别效果不理想。（剩余12819字）