خط مشی دسترسیدرباره ما
ثبت نامثبت نام
راهنماراهنما
فارسی
ورودورود
صفحه اصلیصفحه اصلی
جستجوی مدارک
تمام متن
منابع دیجیتالی
رکورد قبلیرکورد بعدی
Document Type : Latin Dissertation
Language of Document : English
Record Number : 150993
Doc. No : ET22785
Main Entry : KASHIF MAHMOOD
Title Proper : DECISION FEEDBACK EQUALIZATION USING HYBRID LATTICE-NEURAL NETWORK STRUCTURES
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : The non-linear structure of neural network makes it very suitable for charatelequalization. especially when the channel is hemi& distorted. In h i s rhcsis IVCim~estigate rke performance of three direrent Neural ,Vetwork based Decision FccdhurkEqualization schemes with and without Latrice jilter. The lattice structures arc wellknown for their fast convergence and insensitivity fo the eigen value spread o f 111cchmnel autocorrelarion matrix Firsf, the performance of Radial Basis Funcrior? rtcr~t-or-ktrained through simple Least Mean Square algorithm is investigared for DecisionFeedback Equalization. Second, the Multi Layer Percepnon trained through RecursiveLeast Squares algorithm is used for DFE and its performance is investigated. Finally. fhchybrid neural network based smcture..,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject : Electericl tess
: برق
electronic file name : TL46011.pdf
Title and statement of responsibility and : DECISION FEEDBACK EQUALIZATION USING HYBRID LATTICE-NEURAL NETWORK STRUCTURES [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)