一种改进的批处理常模算法

分析了传统批处理常模算法的缺点，提出了一种新的批处理常模算法——基于共轭 梯度方向的批处理常模算法（CG-BPCMA）。新算法采用共轭梯度方向作为下降方向，推导 出共轭梯度方向的解析形式。在平坦瑞利衰落信道环境下的MIMO盲均衡系统中的仿真结果表 明，新算法有效地克服了SD-BPCMA和NT-BPCMA的缺点，不仅获得了较低的算法复杂度，而 且能快速收敛到常模代价函数的最小值点。

A modified batch processing constant modulus algorithm

By analysing the shortages of traditional batch processing algorithm, a new batch processing constant modulus algorithm derived by conjugate gradient direction to minimizing the constant modulus criterion(CG-BPCMA) is presented. The new algorithm uses conjugate gradient direction to choose the search directi on and the analytical form of conjugate gradient direction is deduced. Simulatio n results show that in flat Rayl eigh fading synchronous channel environment blind equalization system,the new al gorithm overcomes the shortcomings of SD-BPCMA and NT-BPCMA effectively, not only has lower complexity, but also can converge to the minima of the constant modulus criterion quickly.