Call Admission Control for Multimedia Cellular Networks Using Neuro-Dynamic Programming
Résumé
We consider, in this paper, the call admission control (CAC) problem in a multimedia cellular network that handles several classes of traffic with different resource requirements. The problem is formulated as a Semi-Markov Decision Process (SMDP) problem. It is too complex to allow for an exact solution for this problem, so, we use a real-time neuro-dynamic programming (NDP) [Reinforcement Learning (RL)] algorithm to construct a dynamic call admission control policy. A broad set of experiments shows the robustness of our policies compared to the classical solutions such as Guard Channel. Channels could be frequencies, time slots or codes depending on the radio access technique.