Wireless 4G Cell network with hand off Multiple Class Traffic * Demetres Kouvatsos, Irfan Awan and Yue Li Networks and Performance Engineering Research Group University of Bradford, Bradford BD7 1DP, West Yorkshire, UK Email:(D.D.Kouvatsos, I.Awan)@scm.bradford.ac.uk Y.Li5@bradford.ac.uk Abstract A novel analytic framework is devised for the performance modeling and eval- uation of a wireless 4G cell with bursty multiple class traffic flows consisting of voice calls, streaming media and data packets subject to an efficient medium access control (MAC) protocol with a buffer threshold-based traffic handling generalized partial sharing scheme (GPSS). In this context, an open queueing network model (QNM) is proposed consisting of three interacting multiple class GE-type queueing and delay systems, namely a GSM GE/GE/c 1 /c 1 loss system with c 1 servers, a UMTS GE/GE/c 2 /N 2 / FCFS/CBS queueing system with c2 servers, finite capacity, N 2 and First-Come-First-Served discipline under a com- plete buffer sharing (CBS) management scheme and a GPRS GE/GE/1/N 3 /PS delay system with finite capacity, N 3 , under a discriminatory processor share (PS) transfer rule. The principle of maximum entropy (ME) is used to characterize closed form an- alytic solutions for the joint state probabilities of the aforementioned queueing and delay systems, subject to appropriate GE-type queueing and delay theo- retic constraints, as appropriate. As a consequence, analytic expressions for the aggregate state and blocking probabilities are determined. Typical numerical examples are included to validate the ME performance metrics against Java- based simulation results and also to study the effect of bursty multiple class traffics upon the performance of the wireless 4G cell. Key words: GSM/GPRS, UMTS, generalized partial sharing scheme (GPSS), generalized exponential (GE) distribution, queueing network model (QNM), maximum entropy (ME) principle, performance evaluation, first-come-first-served * This work is partially supported by the EU Network of Excellence (NoE) ’Euro-NGI’ P69/1