Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/27626
Title: Intelligent adaptive bandwidth provisioning for quality of service in umts core networks
Authors: HUI CHEE KIN, TIMOTHY
Keywords: Reinforcement Learning, adaptive bandwidth provisioning, Quality of Service, UMTS core network, Differentiated Services, 3G
Issue Date: 25-Jun-2004
Source: HUI CHEE KIN, TIMOTHY (2004-06-25). Intelligent adaptive bandwidth provisioning for quality of service in umts core networks. ScholarBank@NUS Repository.
Abstract: The issue of bandwidth provisioning is imperative for Quality of Service (QoS) to be achieved in next-generation Universal Mobile Telecommunications System (UMTS) core networks. The standardized implementation of Differentiated Services (DiffServ) as the service model enables services with a variety of QoS requirements to be transported over the same core network. To ensure that the QoS requirements of all classes can be met efficiently, the proportion of bandwidth provisioned to each class has to be optimized. The task however is made complex by the continuously varying nature of traffic and the constant modification of routes due to the mobility of end nodes. This thesis presents a solution to the bandwidth provisioning problem in both the core backbone network and the radio access network. The solution makes use of a continuous-space, gradient-descent Reinforcement Learning (RL) algorithm to continuously adapt per-aggregate bandwidth provisions to changing conditions to achieve maximum revenue and minimum QoS violations. Extensive simulations of various traffic conditions, QoS requirements and pricing policies show that the algorithm is able to converge towards an optimal solution.
URI: http://scholarbank.nus.edu.sg/handle/10635/27626
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Thesis.pdf1.76 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

208
checked on Dec 18, 2017

Download(s)

159
checked on Dec 18, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.