Throughput Analysis for Energy Harvesting Cognitive Radio Networks with Unslotted Users

Considering a cognitive radio network (CRN) with the energy harvesting (EH) capability, we design a sensing-based flexible timeslot structure for a secondary transmitter (ST). This structure focuses on an unslotted transmission mode between two primary users (PUs). In this structure, the ST can decide whether to transmit data or to harvest energy based on the sensing results. Aiming to maximize the long-term average achievable throughput of the secondary system, we study an optimal policy, including the optimal energy harvesting time as well as the optimal transmit power. To reduce the computational complexity, we also derive an effective suboptimal policy by maximizing the upper bound on the throughput. Finally, simulation results demonstrate that the proposed flexible timeslot structure outperforms the conventional fixed timeslot structure in terms of average achievable throughput.