雷达工作模式识别的PSO-DPNN方法

PSO-DPNN method for radar operation modes recognition

针对参数交叠严重环境下的雷达工作模式识别问题,提出一种基于粒子群优化算法(PSO)的离散过程神经网络(process neural network, DPNN)的识别方法。方法依据整个雷达信号脉冲序列的时序变化规律进行识别,首先对雷达信号句法建模并提取雷达短语作为工作模式的特征描述,然后运用PSO有监督训练合适的DPNN网络结构,最后运用训练完成的DPNN识别未知雷达短语的工作模式。对比仿真结果表明,信号参数测量误差10%时识别率为97%,较传统识别方法提高30%,方法在参数交叠严重的情况下的工作模式识别率和抗误差性能提升明显。

Aiming at the problem of radar operation modes recognition under the condition of severe overlap of parameters,a discrete process neural network ( DPNN) based on particle swarm optimization ( PSO) is proposed. The method realize operation modes recognition according to the time series changing law of the whole radar signal pulse sequence. Firstly,radar syntactic modeling method is proposed to extract radar phrases as operation modes character description. Next,the appropriate DPNN structure is built by using the PSO algorithm. Finally,the finished DPNN is used to realize operation modes recognition of unknown radar phrases. The simulation results show that the recognition rate is 97% when the measurement error of signal parameters is 10%,which is 30% higher than the traditional method. It is shown that the recognition rate and anti-error performance of this method are greatly improved when the parameters highly overlap.