Call admission control policy for multiclass traffic in heterogeneous wireless networks

Abstract

An increasing demand for wireless access to packet-based services has, in turn, driven the need for call admission control strategies to make network roaming more seamless. This paper investigates the optimal call admission control (CAC) policy for heterogeneous wireless networks based on Markov Decision Process (MDP) and Value Iteration Algorithm (VIA). We concentrate on the optimization problems of when different type of calls (i.e., new calls, vertical handoff calls, and horizontal handoff calls) can be admitted into the system where each call may have single or multiple channel requirements (multiclass). Experimental results confirm that the proposed technique can effectively control call admission of a highly dynamic heterogeneous system, which will be crucial for accommodating vertical handoff decision algorithms.