一种结合EWT和成分分析的无线电指纹提取方法

无线电台信号个体识别主要是提取无线电信号中的杂散成分,通过对杂散成分进行分析达到个体识别的效果。针对线电信号杂散成分具有非线性、非平稳性的特点,本文将经验小波变换(EWT)和信号成分分析结合起来,提出了一种新的信号特征提取方法。该方法首先利用EWT对信号进行自适应的分解处理,通过选取部分能够表征个体差异的信号成分进行特征值谱分析,并以信号成分的归一化特征值谱的差异为依据进行信号指纹特征的提取,再根据指纹特征对信号进行识别。仿真结果表明,该方法与HHT和局部积分双谱分析方法相比,具有更加优越的识别性能,并且具有更加优良的特征稳定性,同时受信噪比的影响较小。

A Radio Fingerprint Extraction Method Combining EWT and Component Analysis

Individual identification of radio signals is mainly to extract the stray components of radio signals, and achieve the effect of individual identification by analyzing the stray components. Aiming at the non-linear and non-stationary characteristics of the spurious components of line signals, this paper combines the empirical wavelet transform (EWT) with the signal component analysis, and proposes a new signal feature extraction method. Firstly, the EWT is used to decompose the signal adaptively, and the eigenvalue spectrum is analyzed by selecting some signal components which can represent individual differences. Then the fingerprint feature is extracted based on the difference of normalized eigenvalue spectrum of signal components, and then the signal is identified according to the fingerprint feature. The simulation results show that the proposed method has better recognition performance than HHT and local integral bispectrum analysis method, and has better characteristic stability, and is less affected by signal-to-noise ratio.