基于特征值极限分布的合作频谱感知算法

Cooperative spectrum sensing algorithm based on limiting eigenvalue distribution

采用最新的随机矩阵理论,对多个认知用户接收信号采样协方差矩阵的最小特征值的极限分布进行了分析,提出了一种改进的最大最小特征值合作感知和门限判决方法。该算法不需预知授权用户信号的先验知识,且能有效克服噪声不确定度的影响。与现有算法相比,在给定虚警概率时,仿真结果显示该算法判决门限更低、检测概率更高;而且在认知用户和采样数较少时,也能获得很好的检测性能。

A novel maximum-minimum eigenvalue（NMME） cooperative spectrum sensing algorithm and threshold decision rule are proposed via analyzing minimum eigenvalue limiting distribution of the covariance matrix of the received signals from multiple cognitive users（CU） by means of latest random matrix theory（RMT）. The proposed scheme could not need the prior knowledge of the signal transmitted from primary user（PU） and could effectively overcome the noise uncertainty. At a given probability of false alarm（Pfa）, simulation results show that the proposed scheme can get lower decision threshold and higher probability of detection（Pd） compared with the original algorithm, and it can also get better detection performance with fewer CU and smaller sample numbers.