基于递归模糊神经网络理论盲均衡算法的研究-电路与系统专业论文.docxVIP

太原理工大学硕士研究生学位论文关键词：盲均衡，动态递归，模糊神经网络，隶属函数，代价函数 太原理工大学硕士研究生学位论文 关键词：盲均衡，动态递归，模糊神经网络，隶属函数，代价函数 II 太原理工大学硕士研究生学位论文THE 太原理工大学硕士研究生学位论文 THE STUDY 0F BLIND EQUALIZATl0N ALGoRITHM BASED oN RECURSIVE FUZZY NEURAL NETWORK ABSTRACT In digital communication systems，the inter-symbol interference Oso caused by non-ideal character of channel is the main factor which a彘cts communication quality．In order to reduce the ISI，a equalizer has to be used in the receiver to compensate channel characteristics．The conventional equalizer resorts to a training sequence to overcome inter-symbol interference．But this method would influence communication efficiency． Blind equalization techniques rely on solely the received channel output signal to adjust the equalizer weights without a known training available． The major contribution ofthis paper is summarized as follow： 1．This paper formulizes the foundational principles and development of blind equalization techniques based on neural network and analyzes equalization algorithms of combining fuzzy technique and neural network, This paper proposes three kinds of blind equalization algorithm principles and applying ways based on fuzzy neural networks，then put two kinds of III 太原理工大学硕士研究生学位论文fuzzy 太原理工大学硕士研究生学位论文 fuzzy neural networks into equalization algorithms． 2．This paper proposes to put two kinds of dynastic recurrent fuzzy neural networks into equalization algorithms，using them to replace the traditional equalizer．The paper deduces the formula of the two kinds of networks and expands them into plural field． 3．The simulation result shows that the convergence performance ofthe new algorithms is beRer than that of the blind equalization algorithm based on feed．forward neural network． KEY WORDS：Blind Equalization，Dynastic Recurrent，Fuzzy Neural Network，Member-ship Function，Cost Function IV 声尸 声尸 明明 本人郑重声明：所呈交的学位论文，是本人在指导教师的指导下， 独立进行研究所取得的成果。除文中已经注明引用的内容外，本论文 不包含其他个人或集体已经发表或撰写过的科研成果。对本文的研究 做出重要贡献的个人和集体，均已在文中以明确方式标明。本声明的 法律责任由本人承担。 论文作者签名：j魍籼辛卜一 Et