基于FastICA的雷达信号分选研究

现代战争中新体制雷达的大量涌现,电磁环境变得越来越复杂,对雷达信号分选提出了新的挑战。目前的雷达信号分选领域,多采用基于参数容差的传统分选方法,这些方法受参数误差的影响大,对PDW参数相似的雷达无法分选,已经无法适应复杂电磁环境。在对FastICA算法原理分析的基础上,重点研究了将它应用于PDW参数相近的雷达信号和参差脉冲列的分选,并进行了仿真。仿真结果表明,FastICA是建立在源信号统计独立基础上的处理,对信号相关性敏感,受参数误差的影响小,可以有效解决上述问题,为雷达信号分选提供了一种新的思路。

Radar Signal Sorting Based on FastICA Algorithm

The radar signal sorting faces new challenge with the complicated electromagnetic environment and generation of advanced radars in modern warfare.At present,the traditional methods based on PDW parameter threshold are adopted widely on radar signal sorting.These traditional methods which can not adapt to the complicated electromagnetic environment are affected seriously by parameter error,and can not sort radars which have similar parameters.FastICA algorithm is ana lyzed and chosen to blind separate radar signals which have similar parameters and stagger pulse train.The simulation results show that the algorithm based on the hypothesis of blind source separation also has good recognition effect when the parameter error exists.This method can solve the problem effectively,and provide a new way for radar signal sorting.