Throughput-optimal User Association in Energy Harvesting Relay-Assisted Cellular Networks

Yanzi Song,Ming Zhao,Wuyang Zhou,Hui Han
DOI: https://doi.org/10.1109/wcsp.2014.6992165
2014-01-01
Abstract:We study the user association problem targeting on the throughput optimization for energy harvesting (EH) relay-assisted cellular networks where the base stations (BSs) are powered by grid power and the relay stations (RSs) are powered by renewable energy. Since the harvested energy of RS is stochastic and not always large enough, our challenge is how to match the user association with the energy arrival rate of RS so as to maximize the throughput of the network. For this purpose, we propose a dynamic bias based user association policy which means the user is associated with the BS/RS that provides the strongest-biased-average-received-power, and maximize the throughput by optimizing the bias with different energy constraints. Using tools from stochastic geometry and continuous time Markov chain (CTMC), we first formulate the problem as a throughput-optimal and energy-constrained problem with respect to the bias. Then, by solving the optimization problem, we derive the closed expression of the optimal bias maximizing the throughput. Numerical results show that our dynamic bias based user association policy can always outperform that without bias, especially when the RSs are energy-limited.
What problem does this paper attempt to address?