摘要
提出一种基于循环谱切片的通信辐射源个体识别方法。通过计算信号的循环谱密度矩阵,将循环谱密度切片作为初始高维特征,再采用主成分分析方法对其进行降维处理得到指纹特征矢量,最后采取概率神经网络分类器实现辐射源的个体识别。通过对20部手持机的实验表明,使用该方法提取的特征矢量能够较好地反映信号的循环平稳特性,并且特征参数对噪声干扰不敏感,在较低信噪比条件下,系统仍具有较高的正确识别率,说明该方法确实能够较好地解决同型号、同批次、同工作参数通信辐射源的个体识别问题。
A method based on cyclic spectrum density slice for emitter identification is presen- ted. The signal cyclic spectrum density matrix is calculated and its slice is used as the initial high-dimension feature. Then the principal component analysis method is used to descend the dimension and obtain the fingerprint feature vector. Finally, the emitter identification is real- ized by using the neural network classifier. The experimental results based on 20 interphones show that the feature vector extracted by the method can reflect the signal cyclostation charac- teristic and the feature parameter is insensitive to noise and interference. Under the condition of low signal-to-noise ratio （SNR）, the system still has a high correct recognition rate. It shows that the method can deal with the individual identification of emitters with same model and same batch.
出处
《数据采集与处理》
CSCD
北大核心
2013年第3期284-288,共5页
Journal of Data Acquisition and Processing
基金
江苏省自然科学基金(BK2009059)资助项目
关键词
辐射源识别
循环谱
主成分分析
指纹特征
emitter identification
cyclic spectrum
principal component analysis
fingerprint feature
