Joint Access Spectrum and Backhaul Energy Allocation for Green Cognitive Heterogeneous Networks

In this paper, we examine the impact of uplink spectrum allocation on the backhaul energy allocation in a decentralized setting. We consider that two small base stations (SBSs) use the band from two spectrum owners to provide the uplink wireless transmission service and that two SBSs use a common renewable power supplier (RPS) to forward the uplink traffic. The wireless traffic will be forwarded to a gateway or macrobase station via the backhaul link. We study the system under a decentralized scenario, where SBSs (wireless operator), the spectrum owner, and the green energy supplier interact with each other. We formulate the interaction as a Stackelberg leader-followers game. The operator acts as the Stackelberg leader by choosing the energy and spectrum prices that it will pay to the RPS and the owner, respectively. Then, the RPS and the owner simultaneously determine the renewable energy storage and bandwidth allocation. We prove that for any set of resource prices, there exists a unique Pareto-optimal equilibrium in the followers' resource allocation game. We also show that the wireless network operator's optimal resource prices lie in one of two regions-one that leads to energy and spectrum allocation balance and one that does not. Furthermore, in the centralized system, we present a characterization of the unique optimal solution. The proposed scheme enables the SBSs to jointly decide to use green energy over backhaul links while properly allocating spectrum resources for wireless uplinks. Finally, the results show that the proposed scheme achieves a system performance comparable to that of a centralized solution.