Throughput Optimization for Energy Harvesting Cognitive Radio Networks with Save-Then-Transmit Protocol

Abstract

In this paper, we consider a time-slotted energy harvesting cognitive radio network with the save-then-transmit protocol, in which the secondary user (SU) is powered by energy harvested from the ambient environment. The primary concern of this study is to design an optimal harvesting-access policy to maximize the throughput of the SU, where the harvesting part of the policy specify the time duration allocated for harvesting energy, and the access part of the policy specify the power level to be used upon transmission. Jointly considering the presence of primary users, diversity of channel quality, time and energy consumption for the sensing process as well as sensing errors, we formulate the above design problem as an infinite-horizon Markov decision process, and propose an algorithm to find the optimal harvesting-access policy using the value iteration. We then investigate the relationship between the throughput, the available energy in the battery and the battery capacity. It is indicated that the achievable throughput is monotonously increases with the available energy in the battery if the available energy is under a threshold, which is determined by the battery capacity. Next, in order to reduce the computational complexity, we propose an optimal myopic policy with a closed-form expression. Finally, the performance of the proposed policies and the impacts of system parameters are evaluated through numerical results.