【作者】 程海青；

【导师】 张立毅；

【作者基本信息】 太原理工大学 ， 信号与信息处理， 2006， 硕士

【摘要】 盲均衡技术是一种不需要训练序列，能够仅利用接收信号的统计特性对信道特性进行均衡的自适应均衡技术，克服了传统自适应均衡技术的缺陷，成为目前数字通信和信号与信息处理领域的热点研究课题之一。 人工神经网络是理论化的大脑神经网络的数学模型，是基于模仿大脑神经网络结构和功能而建立的一种信息处理系统，能够进行复杂的逻辑操作并实现非线性关系。用神经网络设计均衡器具有重要的理论意义和实用价值。 本文所做的主要工作如下： (1)分析了传统BP算法的缺陷，讨论了其改进形式，提出了基于动量项前馈神经网络的盲均衡算法。通过计算机仿真表明，与BP算法相比，新算法的收敛速度有所提高。 (2)研究了动量项前馈神经网络盲均衡算法中动量项及学习速率对算法性能的影响，提出了变动量项前馈神经网络盲均衡算法和变学习速率动量项前馈神经网络盲均衡算法。仿真结果表明，这两种算法与动量项前馈神经网络盲均衡算法相比，收敛性能明显改善，收敛速度有所提高，稳态剩余误差和误码率减小。 (3)根据前馈神经网络和CMA盲均衡算法的特点，构造出一个新的代价函数，并将其应用于动量项前馈神经网络盲均衡算法中。仿更多还原

【Abstract】 Blind equalization is an adaptive equalization technique without a training sequence; it can equalize the properties of the channel by using the statistic properties of the received signals and overcome the disadvantages of conventional adaptive equalization techniques. Thus, it has become a major focus in the field of digital communication.The artificial neural network is a theoretical mathematical model for the cerebral neural networks; it is an information-processing system that is based on stimulating the structure and function of the cerebral neural networks; it can perform complicated operations and establish nonlinear relationships. The use of neural networks for designing equalizer is of important theoretical significance and practical value.The primary subject of this paper is to(1)Evaluate the disadvantage of conventional back-propagation (BP) algorithms, discuss their improved forms, and propose a blind equalization algorithm using momentum feed-forward neural networks. Computer stimulation shows that the convergence speed of this new algorithm is somewhat increased as compared with BP algorithm.(2) Investigate the effects on the function of algorithm of the momentum in the new algorithm and of the learning rate, and propose two blind equalization algorithms using varying momentum feed-forward更多还原