摘要
为了充分利用样本之间的相关性结构,提出了一种基于协作表示的通信辐射源个体识别方法。首先,为表征通信辐射源个体差异,提取矩形积分双谱特征作为细微特征;其次利用协作表示理论挖掘特征样本之间的相关性结构,克服识别过程中的"小样本"问题;最后利用系数包含的判别性信息构造分类器,实现通信辐射源的个体识别。在实际采集的同厂家同型号FM电台数据集上的实验结果验证了该方法的可行性与有效性。
To make full use of the correlation structure among different samples,？a method of individual com- munication transmitter identification based on collaborative representation is put forward in this paper. Firstly, to characterise the individual differences of communication transmitters, we extract the square integral bispectra features as the fingerprintings of the transmitter. Secondly, using the theory of collaborative representation, we mine the correlation structure among the samples and overcome the ＂small sample size＂ problem in the process of identification. Finally, construct the classifier by using the discrimination information contained in the coefficients, and identify the individual communication transmitters. On the actual collected data set from the FM radios with same manufacturing lot and model, the robust identification results verify the effectiveness of our method.
出处
《通信对抗》
2016年第3期13-17,共5页
Communication Countermeasures
关键词
通信辐射源个体识别
协作表示
稀疏表示
矩形积分双谱
individual communication transmitter identification
collaborative representation
sparse representa-tion
square integral bispectra
