Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
451000 | Computer Networks | 2015 | 12 Pages |
In cellular networks, energy provision contributes up a significant fraction of the total operational cost. To address this problem, service providers have started considering the deployment of renewable energy sources. Due to the variability of renewable energy source, conventional grid energy sources is still required to provide steady service for users. In this paper, Base Stations (BSs) equipped with hybrid energy sources and limited energy storages are considered. We investigate a joint power allocation and battery management scheme to cut down electricity cost under electricity markets, which is ignored by previous studies. Considering the random data arrival, link quality, renewable energy and electricity price, a stochastic program which minimizes the time-average expected electricity cost while stabilizing the network is formulated. Based on the Lyapunov optimization technique, we design an online algorithm to approximately obtain the optimal solution. Especially, our algorithm can guarantee the worst delay through introducing virtual queues. Furthermore, theoretical analysis shows that our algorithm offers an explicit tradeoff between cost saving and delay performance. Numerical simulation results demonstrate the effectiveness of our algorithm.