Bayesian-game based power and spectrum virtualization for maximizing spectrum efficiency over mobile cloud-computing wireless networks

Abstract

Mobile cloud-computing is a wireless network environment that focuses on sharing the publicly available wireless resources. Wireless network virtualization provides an efficient technique to implement the mobile cloud-computing by enabling multiple virtual wireless networks to be mapped onto one physical substrate wireless network. One of the most important challenges of this technique lies in how to efficiently allocate the wireless resources of physical wireless networks to the multiple virtual wireless network users. To overcome these difficulties, in this paper we propose the Bayesian-game based schemes to resolve the wireless resources allocation problem in terms of transmit power and wireless spectrum. We formulate this wireless resources allocation problem as the gaming process where each mobile user bids for the limited wireless resources from physical substrate wireless networks, and competes with the other mobile-user players bidding for the same resources. Since the bidding strategies for other mobile-user players are incomplete information, we propose the Bayesian-game based schemes to dictate the virtual users to request wireless resources based on the probabilities of these incomplete information. Our proposed game is guaranteed to converge to the Bayesian Nash Equilibrium, where each virtual user optimizes their resources request actions with the consideration to actions of the other mobile-user players. The extensive simulation results obtained validate and evaluate our proposed schemes.