A learning approach for call admission control with prioritized handoff in mobile multimedia networks

We propose an alternative approach for improving the quality of service in cellular mobile multimedia networks while prioritizing handoff call requests over new call requests. The goal is to reduce the handoff failures while still making efficient use of the network resources. A performance measure is formed as a weighted linear function of new call and handoff call blocking probabilities for each service type. This problem is formulated as a semi-Markov decision process with average cost criterion. A simulation-based learning algorithm is adopted to determine a near-optimal control policy from direct interaction with the network, and without a priori knowledge of the network dynamics or traffic. Extensive simulations are provided to assess the effectiveness of the algorithm under a variety of traffic conditions. Comparisons with some well-known policies are also presented.