Computationally efficient method to evaluate the performance of guard-channel-based call admission control in cellular networks

TitleComputationally efficient method to evaluate the performance of guard-channel-based call admission control in cellular networks
Publication TypeJournal Article
Year of Publication2006
AuthorsYavuz, E. A., and V. C. M. Leung
JournalVehicular Technology, IEEE Transactions on
Volume55
Pagination1412 -1424
Date Publishedjul.
ISSN0018-9545
KeywordsCAC, call admission control, call blocking probability, call dropping probability, cellular networks, cellular radio, channel holding times, computational complexity, guard channel, handoff calls, Markov processes, mean channel holding, one-dimensional Markov chain, probability, quality of service, telecommunication congestion control, wireless channels
Abstract

Many guard-channel-based call admission control (CAC) schemes for cellular networks have been proposed to provide the desired quality of service to not only new calls but also ongoing calls when they hand off to neighboring cells. Blocking/dropping probabilities of new/handoff calls are generally analyzed using one-dimensional Markov chain modeling under specific assumptions to avoid solving large sets of flow equations that makes exact analyses of these schemes using multidimensional Markov chain models infeasible. This is the case with the "traditional" approach, which assumes that channel holding times for new and handoff calls have equal mean values, and the "normalized" approach, which relaxes this assumption but is accurate only for the new call bounding CAC scheme. In this paper, we reevaluate the analytical methods for computing new/handoff call blocking/dropping probabilities for several widely known CAC schemes and develop an easy-to-implement method under more general assumptions. Numerical results show that when the mean channel holding times for new and handoff calls are different, the proposed "effective holding time" approach gives more accurate results compared with the traditional and the normalized methods while keeping the computational complexity low. The accuracy of these methods and their levels of computational complexity with the exact solution are also compared

URLhttp://dx.doi.org/10.1109/TVT.2006.877470
DOI10.1109/TVT.2006.877470

a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949
Email:

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia