基于熵值图的多功能雷达工作模式识别

Multi-Function Radar Working Mode Recognition Based on Entropy Map

针对多功能雷达工作模式多样且信号波形复杂多变,导致常规工作模式识别方法性能不佳的问题,提出基于小波变换的多功能雷达工作状态切换点检测方法,将雷达脉冲序列进行分割,得到单一工作模式脉冲序列样本。其次,基于近似熵、样本熵、模糊熵和排列熵设计了新的熵值图特征,并结合卷积深度神经网络模型,实现了多功能雷达工作模式的智能识别。仿真结果表明,该算法在虚假脉冲率或漏脉冲率为25%时,切换点检测正确率达85%;工作模式识别正确率在虚假脉冲率、漏脉冲率为20%和参数误差为8%时,识别正确率均在83%以上,识别性能皆优于两种对比文献方法,验证了该算法的有效性和优越性。

In order to address the problem that traditional work mode recognition methods cannot adapt to the complex and varied waveform characteristics of multi-functional radar,a detection method of the pulse signal sequence of multi-functional radar state switching points was proposed in this paper,and the pulse signal sequence was segmented into single work mode samples based on wavelet transform.Then,from the perspective of time series analysis and data mining,a multi-functional radar work mode recognition method based on entropy map was realized.This method extracted approximate entropy,permutation entropy,sample entropy,and fuzzy entropy of different inter-pulse parameter sequences to form an entropy map,and used a convolutional deep neural network model to achieve intelligent recognition of multi-functional radar work modes.Simulation results showed that the proposed state switching point detection approach achieved a switching point detection accuracy of 85%when the false pulse rate or the missing pulse rate was 25%,and the operating mode recognition accuracy was above 83%when the false pulse rate,the missing pulse rate and the parameter error were 20%and 8%.Respectively,the recognition performance of the proposed method was superior to the two comparative literature methods,which verified the effectiveness of the algorithm.