Sensing, probing, and transmitting strategy for energy harvesting cognitive radio

We consider a single channel energy harvesting cognitive radio system, where the joint optimization of spectrum sensing, channel probing and transmission power control is considered with the goal to maximize the throughput. We model this control problem as a two-stage continuous-state Markov decision process with one stage for sensing and probing control, and the other for transmission power control. By utilizing the stochastic structure of the two-stage Markov decision process, we simplify the model via the notion of after-state, which reduces the state space and facilitates decision makings. Finally, the performance of the generated strategy is investigated via simulation.