A matching game approach for resource allocation in wireless network virtualization

Abstract

Wireless network virtualization is an effective and economical solution that can significantly enhance the performance of existing radio access networks by decoupling of physical infrastructure from the services it provides. This enables sharing the resources (e.g., subchannels, power, and antennas) of physical infrastructure among multiple network operators. However, the infrastructure provider is required to employ an efficient resource allocation technique to dynamically allocate the resources for the users associated with different network operators. Service contracts with different network operators are crucial for the success of the virtulization scheme deployed by the infrastructure provider. In this paper, a matching game based solution is proposed for resource allocation in OFDMA virtualized wireless network. We considers a market model consisting of an infrastructure provider and multiple network operators. The matching based solution takes into account the objectives of both the infrastructure provider and the network operators. The infrastructure provider aims at maximizing its revenue and while the network operators wants to serve their users at the best performance while paying the minimum amount. Furthermore, we prove that the proposal achieves a pareto optimal solution from the infrastructure provider point of view. Simulation results illustrates the effectiveness of our proposal in terms of spectrum efficiency, quality of service and convergence.