一种基于DFP方法的拟牛顿恒模算法跟踪信道变化能力的研究

Research on the tracking channel changes capacity of a quasi-Newton Constant Modulus Algorithm Based DFP Method

本文根据信道变化快慢程度,把信道模型分成剧变信道和渐变信道。分析研究了恒模盲均衡算法(CMA)在不同的信道模型下跟踪信道变化的能力。针对于原有典型恒模盲均衡算法的性能上的劣势,结合最优化理论,提出了一种线性受限拟牛顿恒模算法,称之为LC-DFPCMA,并将它与传统的线性受限恒模算法(LC-CMA)及最小二乘恒模算法(LSCMA)进行了仿真比较。仿真结果表明,提出的LC-DFPCMA算法比LC-CMA算法能更快地跟踪信道变化且收敛效果较好;与LSCMA算法相比,算法复杂度有明显降低。

According to channel changing rate, the channel model is divided into the acute-change channel and the gradualchange channel. This paper investigates the CMA＇ s capacity of tracking channel changes in different channel model. Aimed at the disadvantages of the original typical CMA blind equalization algorithm, this paper combining the optimization theory, propose a linearly constrained quasi-Newton constant modulus algorithm （LC-DFPCMA）. The simulation results show that the proposed LC-DFPCMA algorithm can track channel changes faster than conventional LC-CMA and make convergence better as well, and its complexity is obviously less than conventional LSCMA.