Asymptotically Optimal Online Solutions for Energy Harvesting Communication Systems

Energy harvesting (EH) has emerged as a promising technique for green communications and it is a novel technique to prolong the lifetime of the wireless networks with replenishable nodes. In this paper, we investigate the online resource allocation for a large class of objective functions in the EH communication systems, which are asymptotically optimal. It is shown that the solution is obtained by only considering the average EH pattern, irrespective of its stochastic dynamics. This optimal solution neither has the complexity of dynamic programming solutions, nor uses the non-causal knowledge about EH pattern. Also, some practical numerical examples for objective functions, which are utilized in communication systems, are considered and general results are derived for them. Moreover, simulation results validate our theoretical findings and show the accuracy of asymptotic theoretical curves for the transmission periods, which are used in practice.