非协作条件下的电台个体识别是目前通信侦察领域的一个难题。尽管已有多种方法从电台的发射信号中提取到指纹特征实现个体识别,但均在不同程度上存在着低信噪比条件下电台个体识别准确率不高的问题。在分析信号包络特征可以应用于电台个体识别的基础上,提出了一种基于信号复调制与低通滤波的改进包络提取算法,精确地提取信号包络;并为达到降维的目的,从提取的信号包络中求得一个对信号包络特征具有较强表征能力的新特征参数。仿真结果表明,不同电台个体的新特征参数具有较好的聚类性能,识别准确率在95%以上,有效提高了低信噪比条件下电台个体识别的准确率。%Under conditions of non-cooperation,individual transmitter identification is currently a problem in the field of communi-cations reconnaissance. Although many methods have extracted fingerprint features from the signals emitted by transmitters for individual identification,the accuracy is low at low signal-to-noise ratio (SNR). Based on the analysis that signal envelope features can be used for individual transmitter identification,an improved algorithm of envelope extraction based on complex signal modulation and low pass filter is proposed. The improved algorithm can get more precise signal envelope. Moreover, in order to reduce the dimensionality, a new feature parameter is obtained from the extracted signal envelope which has a stronger representation capability for signal envelope fea-tures. Numerical simulation shows that the proposed parameter has better clustering performance. With this parameter,the identification accuracy can be above 95%,and the accuracy of individual transmitter identification at low SNR is effectively improved.
展开▼