Analytical Modeling of Call Admission Control Schemes for Multiclass Traffic Mobile Wireless Networks Salman AlQahtani and Ashraf S. Mahmoud Computer Engineering Department, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia, {salmanq, ashraf}@ccse.kfupm.edu.sa Abstract — This paper studies and analyze two call admission control (CAC) schemes in multimedia cellular wireless networks. These call admission control algorithms are studied for different network configurations. These configurations include, employing the queuing techniques for voice handoff with finite lifetime, differentiating between voice and data calls in terms of the average channel holding time, data bandwidth requirements, and employing queuing techniques for voice handoff and data handoff calls with finite lifetime. The main contribution of this paper is the development of an analytical model for each of the two CAC algorithms specified in this study. In addition to the call blocking and termination probabilities which are usually cited as the performance metrics, in this work we derive and evaluate other metrics that have not be considered by previous work such as the average queue length, the average queue residency, and the time-out probability for handoff calls. We also develop a simulation tool to test and verify our results. Finally, we present numerical examples to demonstrate the performance of the proposed CAC algorithms and we show that analytical and simulation results are in total agreement. Index Terms — Analytical Performance, Call Admission Control, Handoff, QoS, Queuing. I. INTRODUCTION The call admission control (CAC) mechanism is one of the most important components of radio resource management (RRM) affecting the resource management efficiency and QoS guarantees provided to users in current and forthcoming cellular networks, like Universal Mobile Telecommunication System (UMTS). In particular, effective call admission control for wireless sessions with a prioritization for handoff calls is an important research issue. The CAC denotes the process of making a decision whether to admit the new or handoff call taking into account the amount of available resources versus ongoing users’ QoS requirements. Two important quality measures of cellular mobile systems are the probability of blocking for new call requests and the probability of calls blocked when a handoff is attempted due to unavailability of resources. A good CAC scheme has to balance the call blocking and call dropping in order to provide the desired QoS requirements. Recently, a number of call admission control algorithms for cellular mobile systems have been proposed and analyzed in the literature. Different handoff priority- based CAC schemes have been proposed. A simple way of giving priority to handoff requests is to reserve a number of channels as in the priority reservation scheme [1]. An alternative is to support queuing either for handoff voice requests [1] or for both newly originating calls and handoff calls [8][9]. In the queuing based schemes, calls are accepted whenever there are free channels; otherwise they are queued with certain rearrangements in the corresponding queue. Analytical results for these wide ranges of CAC algorithms have been proposed in the literature, for some performance metrics such as call blocking probabilities are obtained, however, for either invalid or at best limiting assumptions. For example, it is observed, that due to the mobility, some of usually used assumptions may not be valid which is the case when the average values of channel holding times for data calls and voice calls are not equal or when the life time of multi-queued handoff calls is finite. Also in case of multimedia traffic, the bandwidth requirements may not be equal for all types of traffic. In this paper, we extend some of the analytical results for two call admission control schemes under more general assumptions and provide some easier-to- compute formulas. In addition to the typically cited blocking probability criterion in most previous studies, in this analytical framework we obtain other important performance criteria such as the average queue length, the average waiting time and the average number of calls in the systems. For comparison amongst the different schemes outlines above, we develop a framework that models the CAC process using Markov chains. II. SYSTEM MODEL The system under consideration is a FDMA cellular network supporting multimedia media traffic. The priority classes of incoming call requests are divided into four types. These types are: 1) real-time (voice) service handoff requests (h1); 2) nonreal-time (data) service handoff requests (h2); 3) newly originating real- time (voice) calls (n1); and 4) newly originating nonreal-time (data) calls (n2).As shown in the generic system model depicted in Fig.1, we consider two classes of calls: voice and data. Additionally, for each class of traffic, handoff calls have priority over new call arrivals.