Quality of Service Management Scheme for Adaptive Service in Wireless/Mobile Multimedia Cellular Networks

Sung-Hwan JUNG
Jung-Wan HONG
Chang-Hoon LIE

Publication
IEICE TRANSACTIONS on Communications   Vol.E88-B    No.11    pp.4317-4327
Publication Date: 2005/11/01
Online ISSN: 
DOI: 10.1093/ietcom/e88-b.11.4317
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Wireless Communication Technologies
Keyword: 
adaptive service,  quality of service,  degradation area ratio (DAR),  complete partitioning,  complete sharing,  

Full Text: PDF(1.5MB)>>
Buy this Article


Summary: 
An adaptive service framework is expected to support real-time multimedia services in wirless/mobile cellular networks with various classes of traffic and diverse bandwidth requirements. Quality of service (QoS) provisioning in an adaptive framework is another challenging consideration, such as quantifying the level of bandwidth degradation of an ongoing calls and guaranteeing stable QoS levels. Considering both the period and the depth of degradation, the degradation area ratio (DAR) represents the average ratio of a call's degradation and is one of the meaningful measures for adaptive service in call level analysis. In this paper, analytical models for estimating the DAR and finding the optimal control parameters are presented in multi-class traffic call management situations. In complete partitioning capacity based threshold-type call admission control (CAC), a one-dimensional Markov chain with an absorbing state is proposed for estimating the DAR in each traffic class. We formulate a two-leveled optimization problem minimizing the total blocking probabilities subject to QoS requirements and present the procedures required in finding the optimal capacities and threshold values by using modified dynamic programming. In complete sharing capacity based threshold-type CAC, the multidimensional Markov model is approximately reduced to a one-dimensional model in order to reduce complexity and hence calculation time. The reduced model is compared with multidimensional Markov model in numerical examples. The optimization problem is formulated minimizing the total blocking probabilities subject to QoS requirements and the optimal threshold parameters are found by using a genetic algorithm. Performance of two adopted admission policies in adaptive framework situations is illustrated by numerical results.