SMDP-Based Downlink Packet Scheduling Scheme for Solar Energy Assisted Heterogeneous Networks

Renewable energy assisted heterogeneous networks (HetNets) can improve system capacity and reduce conventional energy consumption. In this paper, we propose a semi-Markov decision process (SMDP)-based downlink packet scheduling scheme for solar energy assisted HetNets, in which solar radiation is modeled as a continuous-time Markov chain (CTMC) and the arrivals of multi-class downlink packets are modeled as Poisson processes. We derive the state transition probabilities and the discounted state transition probabilities for the composite states of battery, solar radiation. To obtain an asymptotically optimal downlink packet scheduling policy, we solve the semi-Markov decision problem using the relative value iteration algorithm under average criterion and the value iteration algorithm under discounted criterion, respectively. The simulation results show that the average cost of the SMDP-based packet scheduling scheme is less than that of the heuristic packet scheduling scheme.