一种基于MCMA的双模切换变步长的盲均衡算法

在修正常模算法（MCMA）代价函数的基础上，根据QAM信号分布的特点，得到一种误差控制函数在CMA和DD算法间切换的、变步长的盲均衡算法。它充分利用了CMA与DD算法各自的优点，改善了常算法收敛速度慢和收敛后剩余码间干扰大的不足，能够补偿由信道引起的相位误差，同时还具有重新启动的能力。计算机仿真表明：该算法收敛速度较快，收敛后剩余码间干扰很小，输出星座图紧簇，没有相位旋转，而且在信道突变的情况下能快速达到二次均衡。

A variable step size blind equalization algorithm for dual mode switching based on modified constant modulus algorithm

This paper studied on a new cost function of modified constant modulus algorithm, according to the trait of QAM signal, then a variable step size blind equalization algorithm united by constant modulus and decision directed was derived. This derived algorithm makes the best of advantage of CMA and DD algorithm and improves the deficiency of CMA which has a lower convergence rate and a large residual ISI when converged, it can recover the channel phase error and it also has cold start and restart ability at the same time. Computer simulation validates that the derived algorithm which has a faster convergence rate, a quite small residual ISI and a tight constellation without phase rotating when converged, furthermore it can soon reach the convergence again when the channel has a break.