基于二次协方差矩阵的频谱感知算法  

Spectrum Sensing Algorithm Based on Quadratic Covariance Matrix

作　　者：韩仕鹏 赵知劲[1] 戴绍港[1] HAN Shipeng;ZHAO Zhijin;DAI Shaogang(School of Communication Enginering,Hangehou Dianzi Unicersity,Hangehou Zhejiang 310018,China)

机构地区：[1]杭州电子科技大学通信工程学院,浙江杭州310018

出　　处：《杭州电子科技大学学报（自然科学版）》2019年第3期10-14,20,共6页Journal of Hangzhou Dianzi University：Natural Sciences

基　　金："十二五"国防预研资助项目(41001010401)

摘　　要：协方差矩阵频谱感知算法不需要主用户先验信息,易于实现,但是,低信噪比时,协方差矩阵元素间差异变小,检测性能有待提高。为此,利用噪声的二次协方差矩阵方差大、主用户信号的二次协方差矩阵元素的相关性增强等特点,提出利用二次协方差矩阵方差和对角线元素的判决统计量,推导出判决门限。AWGN信道和Rayleigh信道下的仿真结果表明:新方法在低虚警概率条件下,检测性能有明显提升,同时抗噪声不确定度和抗频偏性能均有改进。Covariance matrix spectrum sensing algorithm does not require primary user prior information.It can be easy to be realized.However,the difference between elements of the covariance matrix is not obvious at low SNR,and its detection performance needs to be improved.Therefore,by using large variance of the covariance of covariance(CoC)matrix of the noise,and the correlation of the elements of the CoC matrix of the main user signal is enhanced.Based on the variance of the CoC matrix and the diagonal elements,the decision statistic is proposed,and the judgment threshold is derived.The simulation experiments under AWGN channel and Rayleigh channel show that the performance of the new method is obviously improved under the low false alarm probability.At the same time,the anti-noise uncertainty and anti-frequency offset performance are improved.

关 键 词：认知无线电 频谱感知 二次协方差矩阵 低虚警概率 

分 类 号：TN925[电子电信—通信与信息系统]