一种基于最大最小特征值的频谱感知改进算法  

Modified algorithm for spectrum sensing based on MME

作　　者：王润亮[1] 李浩[1] 黄焱[1] 张白愚[1] 

机构地区：[1]解放军信息工程大学信息工程学院,郑州450002

出　　处：《计算机应用研究》2012年第7期2638-2641,共4页Application Research of Computers

摘　　要：针对大多数频谱感知算法在低信噪比下性能不佳的问题,借鉴最大最小特征值算法(MME)提出一种改进算法。该算法对接收信号做差分预处理,再求其统计协方差的最大最小特征值比;最后用数值分析方法,结合聂曼—皮尔逊准则(N-P准则)求出恒虚警概率门限进行判决。蒙特卡洛(Monte Carlo)仿真实验对该算法与MME以及能量检测法的性能进行了比较,在极低信噪比下该算法的性能优于MME,而无论信噪比如何均优于能量检测法。实验表明,改进算法适用于低信噪比下频谱感知。The detection capabilities of most spectrum sensing algorithms are badly affected by the local SNR（signal-to-noise ratio）.Aiming at this problem,this paper adopted the advantages of MME algorithm,and proposed a novel algorithm based on MME.Firstly,it employed a difference operation to the received signal.Then,it took the ratio of maximum to minimum eigenvalue as test statistic.Finally,it derived the constant false alarm rate（CFAR） threshold according to numerical analysis methods and Neyman-Pearson criterion.It used all-sided Monte Carlo simulations to show the relationship between detection performance and SNR.All the simulations took MME and energy detection method as contrast.In the case of MME,the novel algorithm evidently improves the detection capability when sensors are deployed in very low SNR scenes.In another case,it’s effectively improved no matter what the SNR is.

关 键 词：认知无线电 频谱感知 最大最小特征值 数值分析 聂曼—皮尔逊准则 恒虚警率门限 

分 类 号：TP393.04[自动化与计算机技术—计算机应用技术]