摘要
文章提出了一种新的模糊神经网络(FNN:FuzzyNeuralNetwork)控制的变步长盲均衡算法,利用模糊神经网络控制盲均衡算法的迭代步长,以得到更好的均衡性能。该文设计出模糊神经网络控制器的结构并给出状态方程,提出了新的代价函数,推导出控制器参数的迭代公式。计算机仿真表明,该算法与传统恒模(CMA:ConstantModulusAlgorithm)盲均衡算法相比,具有稳定性好的优点。
In this paper,a new variable step-size blind equalization algorithm using the Fuzzy Neuron Network（FNN） to control is proposed.This algorithm uses FNN to control iteration step-size to obtain good equalization performance.In this paper,the structure of FNN controller is designed,state equations are given,a new cost function is proposed and iteration formula are deduced.Results of the simulation show that this algorithm can obtain better stability than traditional Constant Modulus Algorithm（CMA）,
出处
《计算机工程与应用》
CSCD
北大核心
2006年第12期44-46,106,共4页
Computer Engineering and Applications
关键词
盲均衡算法
模糊神经网络
变步长
代价函数
blind equalization algorithm, Fuzzy Neural Network,variable step-size,cost function
