基于HHT与正则化维数的辐射源个体识别

Specific Emitter Identification Based on Hilbert-Huang Transform and Regularization Dimension

信号特征提取是辐射源个体识别技术的关键环节。为提取更加有效的通信辐射源个体特征,提出一种利用希尔伯特-黄变换(Hilbert-Huang Transform,HHT)和正则化维数的辐射源个体识别新方法。首先,通过经验模态分解,在时域将信号分解为多个固有模态函数分量,利用希尔伯特变换获取信号时频能量谱和边缘谱;其次,计算时频能量谱和边缘谱的正则化维数作为辐射源信号的复杂度参数,并联合能量熵组成特征向量;最后,利用支持向量机实现辐射源个体识别。实验结果表明,正则化维数具有良好的类内聚集性和类间可分性,所提方法优于基于HHT能量谱的2种经典方法,并在低信噪比条件下具有较为显著的优势。

The feature extraction of signal is the key step of specific emitter identification(SEI).To extract more discriminating features of communication emitters,a novel approach based on Hilbert-Huang transform(HHT)and regularization dimension(RD)is proposed.Firstly,a signal was decomposed into multiple intrinsic mode function components in the time domain by empirical mode decomposition algorithm to obtain time-frequency energy spectrum and marginal spectrum.Then,RD of time-frequency energy spectrum and marginal spectrum were calculated respectively to characterize complexity of the signal,and energy entropy was combined to form a feature vector.Finally,support vector machine classifier was utilized to identify different emitters.Experimental result shows that RD has good intra-class aggregation and inter-class separability,and the proposed approach outperforms the other two classical approaches based on HHT energy spectrum especially under low SNR conditions.