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 : 151557
Doc. No : ET23349
Main Entry : Nigel A. Ziyad
Title Proper : A General Approach to Signal Decomposition Using Reinforcement Learning with Custom Signal Dictionaries
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : Herein describes a proposed method to estimate signaIs by an adaptivedecomposition. These signals rtre likely to be nonstationary signds, that is, varying infrequency over time, so it is proposed that a scheme using a set of known functions, referredto as "the dictionary," can be used and adaptively improved by a learning method. Thisscheme allows for greater flexibility in applications that limit the ability for an orthogonalsolution or that require a non-orthogonal description of the signal under analysis becausethe a priori information is incomplete. Other methods that use dictionaries are generallylimited to the type of dictionaries that are valid for the method. A general and adaptivemethod would allow ao flexibility of dictionary selection, and bt the possibility ofclassification and/or compression. Thus, this work makes three main contributions:1e It describes a unique philosophy and proof-ofconcept for utilizing customizabledictionaries: a mechanism to allow user heuristics.2p It demonstrates the possibility of such a mechanism, using a simple learning algorithmand criteria.3a It stimulates further interest in this area of work through.....-....,....-...,..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 : TL46588.pdf
Title and statement of responsibility and : A General Approach to Signal Decomposition Using Reinforcement Learning with Custom Signal Dictionaries [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)