MCA-CMA次分量分析恒模算法

Minor Component Analysis-Constant Modulus Algorithm

提出了一种次分量分析恒模算法(Minor component analysis-constant modulus algorithm,MCA-CMA)。该算法从恒模代价函数出发,推导出一个基于Rayleigh熵形式的代价函数,根据Rayleigh熵的性质,可得出最优权矢量就是协方差矩阵的最小特征值所对应的特征矢量,从而引入次分量分析(MCA)寻找最优权矢量,因此该恒模算法称为MCA-CMA算法。仿真结果充分验证该算法的有效性。

xA new constant modulus algorithm （CMA） is proposed and the error cost function is deduced from CMA2-1 one. According to the Rayleigh properties, it is known that an optimal weight vector is eigenvectors of the covariance matrix corresponding to the minimum eigenvalues. Minor component analysis （MCA） is adopted to find out the optimal weight vector. So the constant modulus algorithm is called the MCA-CMA. Simulation results confirm its validity.