| In order to solve the problem of repeatable physical deployments and resources couldn’t be dynamic allocation in the traditional wireless networks. Wireless network virtualization makes the management of the network resource and network service separable. It realizes that the network resource abstracts from the traditional wireless network in order to share the resource. So it avoids the waste of resource and improves the network resource utilization.In wireless network virtualization, current research for resource allocation mainly focuses on bandwidth and CPU resource. And the power allocation has not been adequately studied. In wireless network virtualization, power allocation could be considered in a single physical node with multiple coexisting virtual nodes. The reasonable power allocation strategy will improve the utilization of power resources. In order to solve this problem, A Game theory based Two Steps Power Allocation scheme for Wireless Network Virtualization, is proposed, which designs a Stackelberg Equilibrium price strategy between buyer and seller, and then performs auction game to reallocate leisure resource and uses McAfee mechanism to ensure the integrity of the auction. Finally, the experimental simulation shows that the algorithm not only realizes the maximization of the interests, also improves the power resource utilization rate.In addition, because of the dynamic of the wireless environment, coupled with the arrival and departure virtual network request randomness, so that substrate network would be unbalance and result in reduced the rate of acceptance for subsequent requests. Based on this, the paper presents a Two stages Reconfiguration scheme for Wireless Network Virtualization. In the first stage, the algorithm selects virtual networks to remap. Then, the virtual networks with the priorities are managed, and each map is determined by discrete degree. Finally the simulation shows the algorithm improves the acceptance rate and benefit of request at the same time, and achieves the load balance of substrate network. |
