摘要
本文提出了一种基于功放宽带Volterra级数模型的辐射源"指纹"分类方法,以用于电子侦察对宽带信号源的特定辐射源识别.针对传统Volterra系统MIMO建模方法的适应性不足,以及系统辨识过程存在维数爆炸问题,本文推导出一种对编码类和调频类宽带信号都适用的MIMO建模方法,然后直接基于参数子空间比较来实现功放的指纹分类.对本方法的"独立性"和"可测性"进行了分析.数值仿真实验对子空间比较方法进行了验证.
A Power Amplifier （PA） fingerprinting method based on the Volterra series model is proposed for specific emitter identification of wide band emitters. To cover the adaptability and dimension explosion shortcoming of the traditional Volterra identification method,a new MIMO model is derived for the code modulation signal and the linear/nonlinear frequency modulation signal. And the distance of the subspace is adopted to compare different Volterra systems. The independent and measurable properties of the method are given, and the principle of the method is verified by the simulation experiments.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2009年第8期1712-1717,共6页
Acta Electronica Sinica
基金
总装预研(No.41101020501)
