基于特征值矩阵的循环平稳检测算法

摘要

分析循环平稳检测算法,发现循环自相关函数向量在循环统计量计算过程中,信号之间的相关信息未得到充分利用,信息有一定的损失.提出基于特征值矩阵的统计量计算方法,用自相向量的协方差矩阵的特征值矩阵替代协方差矩阵.该方法使循环自相关函数中的信息得到充分利用,计算复杂度降低.仿真结果表明,该算法的检测性能在较低信噪比下优于经典的循环平稳检测算法的检测性能.%There was a certain amount of loss in the correlation of cyclic autocorrelation function vector with cy-clostationary spectrum sensing algorithm when the decision statistics were calculated. The calculation method of cyclic statistics based on eigenvalues matrix was proposed, and therefore, the covariance matrix was replaced by its eigenvalues matrix. The correlation information of the cyclic autocorrelation function was preserved. In addition, the computational complexity was reduced. Simulation results show that the detection performance of the algorithm is better than that of the classical cyclostationary detector at low SNR.