认知无线网络中基于特征向量的协方差盲检测方法

传统的协方差检测算法需要信道条件信息和多个认知节点参与。利用主用户信号与噪声信号协方差矩阵相关性的差异，提出了一种基于特征向量的协方差盲检测算法，推导出了所提算法的虚警概率和门限值的闭合表达式。该算法仅需要2个认知节点，不需要任何先验信息，对不确定噪声有很强的适应能力。仿真结果表明，所提算法性能优于其他协方差盲检测算法。

Covariance Blind Detection Method Based on Eigenvector in Cognitive Radio Network

As the blind detection algorithm has the shortcoming that they need information about the channel and more than two cognitive users to detect the primary user,a new blind detection algorithm based on eigenvector using the difference of correlation between the primary user signal and noise signal covariance matrix was presented,and a closed expression was derived for the probability of false alarm and threshold.The proposed method overcomes the noise uncertainty problem only with two cognitive relays and performs well without information about the channel,primary user and noise.Numerical simulations show that the new detector performs better than other detectors.