An-MDP based Dynamic Pricing and Revenue Maximization in Wireless Networks

In this paper, we propose a user centric based general framework for revenue maximization in highly competitive wireless networks. By applying dynamic pricing, each service provider operates an optimal policy that aims at maximizing its revenue. The problem is formulated using the Markov Decision Process (MDP) framework and Qlearning is applied to determine an optimal policy which maximizes the long term revenue of the service provider. Simulation results show that the service provider can maximize its revenue by applying dynamic pricing while supporting an appropriate level of user satisfaction in terms of price and call level QoS.