基于多维信号特征融合的空间通信目标识别技术  

Space communication target identification based on multi-dimensional signal feature fusion

作　　者：徐书华[1] 徐丽娜[2] 

机构地区：[1]华中科技大学电子与信息工程系,湖北武汉430074 [2]武汉科技大学管理学院,湖北武汉430074

出　　处：《微型机与应用》2011年第9期73-76,共4页Microcomputer & Its Applications

摘　　要：针对空间通信目标个体识别问题,在射频指纹分析的基础上提出了一种多维信号特征融合提取方法。首先分别在时域、频域和高阶谱域对截获的空间通信目标射频信号提取个体多维信号特征,然后对提取的特征进行融合,并应用支撑矢量机对个体进行分类识别,最后采用实测数据对这种识别方案进行了验证。实验表明,通过多维信号特征融合方法可以有效提取空间通信目标的个体信息,并能获得良好的识别效果。A multi-dimensional fusion feature based on radio frequency fingerprinting is proposed to identify individual space communication targets.At first,the multi-dimensional signal features are extracted from time domain,frequency domain and high-order spectra domain.Secondly,the extracted features are interfused and a support vector machine is used for individual identification.Then,the actual samples are used to demonstrate the proposed scheme.The experimental results show that the proposed scheme could extract individual feature effectively and obtain a high recognition rate.

关 键 词：多维特征融合 通信目标识别 支撑矢量机 

分 类 号：TN973.1[电子电信—信号与信息处理]