基于随机矩阵特征值差的频谱感知改进算法  

DAM Spectrum Sensing Algorithm Based on Random Matrix

作　　者：高鹏[1] 刘芸江[1] 高维廷[1] 陈娟 李曼[3] GAO Peng;LIU Yun-jiang;GAO Wei-ting;CHEN Juan;LI Man(Institute of Information and Navigation,Airforce Engineering University Xi’an 710077,China;Unit 91917 of PLA,Beijing 100000,China;Xi’an Aeronautical University Xi’an 710077,China)

机构地区：[1]空军工程大学信息与导航学院 [2]解放军91917部队 [3]西安航空学院

出　　处：《火力与指挥控制》2018年第8期17-21,共5页Fire Control & Command Control

基　　金：国家自然科学基金(61571364);中国博士后科学基金资助项目(2016M603044)

摘　　要：针对经典特征值类频谱感知算法在低采样、低信噪比下检测效果不佳的问题,基于最大最小特征值之差算法(DMM),利用最小特征值极限分布以及特征值均值的能量特征,提出了一种基于随机矩阵特征值差的频谱感知改进算法(DAM),该算法以特征值均值与最小特征值为检测量,分析比较了算法在两种计算门限下的检测性能(DAM1与DAM2),理论与仿真结果表明,该算法在不增加运算复杂度的同时,在低采样、低信噪比的情况下,较DMM算法以及现有最大特征值与能量差改进算法(ME-S-ED)提升了检测概率,其中DAM1更适用于低信噪比,DAM2更适用于低采样。Aiming at the problem that the eigenvalue based spectrum sensing algorithms don't perform well in the situation of low SNR and small sample,based on the Difference between the Maximum and Minimum eigenvalue spectrum sensing algorithm(DMM),a Difference between the Mean and Minimum eigenvalue spectrum sensing algorithm(DAM)is proposed via the limiting distribution of minimum eigenvalue and the energy quality of the mean eigenvalue with the Random Matrix Theory(RMT).The algorithm performance(DAM1 and DAM2)with two different thresholds relatively are analyzed,which are deducted in different ways.The simulation results show the DAM has the best performance without increasing algorithm complexity over the DMM and current Difference between the Maximum and average Energy spectrum sensing algorithm(ME-S-ED)in the situation of low SNR and small sample,the DAM1 suits low SNR better and the DAM2 suits small sample better.

关 键 词：认知无线电 频谱感知 随机矩阵理论 特征值极限分布 最大最小特征值之差 

分 类 号：TN92[电子电信—通信与信息系统]