Energy Efficiency Optimization in Energy Harvesting Two-Hop Half-duplex Relay Systems

To prolong the lifetime of satellites when the energy stored in photovoltaic panels is used up, it may need deploy some energy harvesting (EH) relays in the future mobile satellite communication system. In this paper, considering an average throughput constraint within a given time, we investigate the energy efficiency (EE) optimization problem of the fuel-cell-powered satellite in collaboration with a decode-and-forward (DF) half-duplex EH relay. Paying regard to the non-ideal circuit power in the relay system and the randomness of energy arrivals, the sleep mode of both the satellite and the relay is joined into the power allocation strategy. Firstly, we solve this problem by the dynamic programming (DP) algorithm. Due to the high computational complexity, the DP algorithm is not feasible in real satellite systems. We then propose a heuristic algorithm to achieve the suboptimal solution. Numerical results show that the performance of the heuristic algorithm approaches that of the DP algorithm, which indicates the heuristic algorithm can be an good alternative in practice.