Revisiting assumptions in backoff process modeling and queueing analysis of wireless local area networks (WLANs)
Faculty / School
Faculty of Computer Sciences (FCS)
Department of Computer Science
Was this content written or created while at IBA?
In this paper, we revisit the modeling assumptions of some of the most widely accepted models in existing literature on the IEEE 802.11 distributed coordination function (DCF) with respect to their accuracy. Ultimately, our objective is to verify the precision of the analytical models with simulation results. First, we outline a number of issues and criticalities in underlying assumptions of existing approaches to backoff process modeling for unsaturated and saturated scenarios. Then, by revisiting the assumptions in existing backoff process models, we reassess them for the more complicated unsaturated case under more realistic assumptions. We argue that under unsaturated load conditions the backoff process exhibits a ρ-persistent behavior. Our claims are supported by observing the simulated behavior of a wireless local area network (WLAN) under varying load conditions. Also, the principle of 'measures of uncertainty' provides theoretical basis for our arguments. Comparison with known models in different simulation environments confirms the increased accuracy of the proposed model over a wide range of settings.
HJRS - X Category, Scopus, Web of Science - Science Citation Index Expanded (SCI)
Iradat, F., Andreev, S., Ghani, S., Nabi, S. I., & Arain, W. (2014). Revisiting assumptions in backoff process modeling and queueing analysis of wireless local area networks (WLANs). Computer Journal, 57 (6), 924-938. Retrieved from https://ir.iba.edu.pk/faculty-research-articles/114