Statistical-QoS based gaming for optimal power allocations over virtualized wireless relay networks supporting multimedia services

Abstract

Cooperative communication is one of the efficient techniques to support statistical delay-bounded quality-of-service (QoS) multimedia (audio/video) services over mobile wireless networks by employing relay nodes, which forward the signal received from source node to the destination. Wireless network virtualization provides another efficient technique to implement the data transmissions by enabling multiple virtual wireless networks (VWNs) to be mapped onto one physical substrate wireless network (SWN), increasing the utilization efficiency of SWN. Applying the cooperative communication to virtual wireless networks provides a significant improvement on time-sensitive multimedia services in order to support the diverse quality of service requirements over mobile wireless networks. One of the most important challenges of this application lies in how to efficiently allocate the wireless resources of physical wireless networks to the multiple virtual wireless network users, taking into account the statistical delay-bounded QoS for the multimedia information transmission. To overcome this difficulty, in this paper we propose a novel game-theory based scheme to resolve the wireless resources allocation problem in terms of transmit power, aiming at maximizing the effective-capacity to guarantee the statistical delay-bounded QoS for wireless multimedia communications. We formulate this wireless resources allocation problem as follows. A mobile user starts with asking for the cooperation from the other virtual users in the same virtual network to be as its relay nodes. When at least one virtual user agrees to cooperate, the mobile user executes the following two gaming processes. First, the mobile user bids for the source-node-transmit-power resources from SWN. Second, at the same time the mobile user bids for the relay-transmit-power from those cooperated virtual users, coordinating with the first bid strategies to maximize its payoff (performance-gain minus cost) in each bid round. The extensive simulation results obtained validate and evaluate our proposed schemes.