As users in a GPRS/EGPRS network have different Quality of Service (QoS) demands, the network aims to satisfy their demands while maximizing the utilization of the existing resources. A crucial aspect of the resource allocation problem in cellular data networks is the admission control (AC). We propose a new session admission control algorithm that overcomes most of the difficulties encountered by the existing AC algorithms. The novelty of our approach is the use of fuzzy logic (fuzzy inference) for the AC. We demonstrate the efficiency of our AC algorithm by simulations.
The delay characteristics of small IP packets in cellular data networks is investigated through the analysis of the SR-ARQ mechanism. The delay of the radio link controller is taken as the primary delay in the radio access network and all core networking delays are neglected. Results indicate that the average delay may be a misleading measure of system performance for conversational traffic such as Voice over IP packets.
A simple, effective model of a GPRS/GSM cell is developed that can give a clear insight into the behaviour of MAC layer scheduling algorithms for a definable cohort of users. Users are defined in terms of their QoS class, their mobility characteristics and their traffic characteristics. The channel model is a C/I statistical model that has a variable mean value that is generated from a two state Markov model for each user. As an example, a weighted round robin scheduling algorithm is analysed within this simulated environment to predict its effect on traffic flow and user perceived network performance. Results presented show that the scheduling algorithm can simply and effectively implement QoS differentiation and reduce the perceived delay for real time streaming users, thereby improving the user perceived performance of the network.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.