一种低信噪比下基于时频原子的复杂调制雷达信号分选方法

复杂调制的多种雷达辐射源信号在低信噪比环境下,具有复杂度高、识别效率低的特点.提出一种基于改进的Chirplet时频原子特征的雷达信号识别分选方法.首先分析多种调制方式的雷达信号,然后分析PSO智能优化算法在参数搜选中的原理和优势,并用该算法对时频原子提取过程进行改进,之后提出PSO算法与时频原子概念结合的方法,利用类区分度准则提取信号特征,得到LPI雷达信号的有效表征原子.最后通过仿真实验证明该方法对于3 dB以上多种调制方式的雷达信号,可快速地完成有效识别分选.

A novel method based on time-frequency atom for complex modulated radar signal sorting in low SNR condition

Multiple radar emitter signals with complex modulation in low SNR environments have characteristics such as high complexity and low recognition efficiency.This paper presents a method based on improved Chirplet time-frequency atom characteristics for the recognition and sorting of radar signals.First the paper analyzes radar signals with multiple modulations,and then studies the principle and advantages of the PSO intelligent optimization algorithm in parameter searching and selecting.In the next step the algorithm of time-frequency atom extraction process is improved by the PSO algorithm.After that a combined method of PSO algorithm and time-frequency concept is put forward.This method uses class discrimination criterion to extract the signal features,then obtains characterization atom of LPI radar signal.Finally,simulation results show that this method for complex modulated radar signal in more than 3dB SNR can effectively complete the task of identification and sorting.