King's College London

Research portal

Dynamic Configuration and Optimization of WiMAX Networks with Relay Power Saving Modes: Measurement-Based Scenario in a Hilly Region

Research output: Contribution to journalArticle

Daniel Robalo, João R. Oliveira, Fernando J. Velez, Oliver Holland, Hamid Aghvami

Original languageEnglish
Pages (from-to)937-958
Number of pages22
JournalWIRELESS PERSONAL COMMUNICATIONS
Volume85
Issue number3
Early online date24 Jun 2015
DOIs
Publication statusPublished - Dec 2015

King's Authors

Abstract

This paper investigates the performances achievable by WiMAX networks deployed in various sectorization configurations, with and without relay stations (RSs). Further, it studies the dynamic adjustment of the configuration to serve traffic loads at different times of the day while maximising the use of opportunistic sleep modes by relays in conjunction with cell zooming, thereby saving energy. The configuration changes and invocation of opportunistic sleep modes also take into account coverage constraints. This paper first reports extensive propagation measurements that have been undertaken in Covilhã, a hilly area of Portugal which presents a realistic and challenging propagation scenario. Using this scenario as the topographical basis, practical cellular planning results are then obtained and compared, using the dominant path and ray tracing (RT) functionalities of Winprop™. It is shown that without RSs present, the supported throughput is lower in practice because coverage is not 100 %. Further, for the case with omnidirectional cells, coverage reduces to only approximately 60 % if RSs go into sleep mode, and for the tri-sectored cells case coverage drops from 95.75 to 81.90 % (based on RT calculations) if RSs go into sleep mode. There is, however, still a reasonable economic performance in all cases. Additional results demonstrate that savings typically of 47.6 % in RSs’ average power consumption can be achieved. These savings are shown to result in a financial saving for the operator of 10 % of the combined operational and maintenance cost. However, it is observed that such solutions have to be used cautiously in hilly regions due to challenges in maintaining coverage.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454