基于双谱的辐射源个体识别技术

发布时间：2018-02-21 08:51

本文关键词： 辐射源个体识别局部积分双谱矩形积分双谱最大比重区间自适应组合核函数　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：通信辐射源个体识别,又称辐射源指纹识别,是近来通信对抗领域一个重要的研究课题。它是指对接收的通信信号进行特征提取,并根据已有的先验信息确定产生信号的辐射源个体的过程。因此对辐射源信号细微特征进行分析并提取出不同于其他辐射源信号的特征对于辐射源个体识别过程极其重要。本文讨论了不同局部积分双谱特征提取方法,通过实验研究比较了现有的几种局部积分双谱以及选择双谱的优缺点。重点研究了矩形积分双谱方法提取信号特征,由于这种积分双谱不仅具有时移不变性、尺度变换性、相位保持性,而且能够较好的无遗漏和无重复的对双谱值进行采样。在矩形积分双谱的基础上采用最大比重区间的改进方法,剔除了贡献小甚至负作用的双谱值,减少了冗余量及这部分带来的噪声值。实验结果表明,改进方法较改进前识别率有所提高,同时在不同信噪比下改进方法也具有较好的识别率。我们采用自适应组合核函数主成分分析方法,由于该核函数能够较好的兼顾全局特征和局部特征,大幅度的降低了特征矢量维数,实验结果验证该方法用于特征矢量维数的约简,在保证识别率的情况下,运算效率大幅度挺高。
[Abstract]:Individual identification of communication emitter, also called fingerprint recognition of emitter, is an important research topic in communication countermeasure field recently. It refers to the feature extraction of the received communication signal. Based on the prior information available, the process of identifying the individual source of the emitter signal is determined. Therefore, the fine features of the emitter signal are analyzed and the characteristics different from those of the other emitter signals are extracted for the individual identification process of the emitter source. In this paper, different local integral bispectral feature extraction methods are discussed. The advantages and disadvantages of several kinds of local integral bispectrum and selective bispectrum are compared by experiments. The rectangular integral bispectrum method is used to extract the signal features, because this integral bispectrum is not only time-invariant, but also scale-transforming. On the basis of rectangular integral bispectrum, the improved method of maximum specific gravity interval is adopted to eliminate the bispectral value with little contribution or even negative effect. The redundancy and the noise caused by this part are reduced. The experimental results show that the improved method has higher recognition rate than that before the improvement. At the same time, the improved method also has a better recognition rate under different SNR. We adopt the adaptive combined kernel function principal component analysis method, because the kernel function can better take into account the global and local features. The experimental results show that this method is used to reduce the dimension of feature vector, and the efficiency of the algorithm is very high under the condition of guaranteed recognition rate.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN975

【参考文献】