Game-Theory Based Buffer-Space and Transmission-Rate Allocations for Optimal Energy-Efficiency Over Wireless Virtual Networks
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2015 IEEE. Wireless network virtualization provides an efficient technique to reduce the cost of overall networks by allowing multiple wireless virtual networks to be mapped onto one physical substrate wireless network. One of its important challenges of this technology is how to efficiently allocate the wireless resources of physical wireless networks to multiple virtual network users. To overcome these challenges, we develop the game- theory based schemes to resolve the resources allocation problem on both buffer/storage space and transmission-rate. Mapping the virtual networks onto the physical wireless network is formulated as the process of bidding for the limited wireless resources from a physical substrate wireless network, and competing with the other mobile-user players, which bid for the same wireless resources of a physical substrate wireless network. Our proposed game- theory schemes work as follows. First, virtual network announces its value function to the physical substrate wireless network, to claim for their required resources (buffer/storage space and transmission bandwidth). Then, the physical substrate wireless network allocates the two types of wireless resources to each virtual wireless network and charges for the provided resources according to the claimed value functions for all virtual users and its current network conditions. Finally our proposed schemes feed back conjectural prices for virtual networks' next-time bid. By accepting the conjectural prices, the game of virtual networks converges the Nash Equilibrium where their long term payoffs are optimized. The simulation results obtained validate and evaluate our proposed schemes through numerical solutions and the simulations.
name of conference
2015 IEEE Global Communications Conference (GLOBECOM)
