Optimal and low complexity algorithm for energy efficient power allocation with sensing errors in cognitive radio networks

This paper investigates an energy efficient power allocation scheme in cognitive radio networks (CRNs), where secondary user (SU) may incorrectly sense the unlicensed spectrum and hence transmit data in collision. We aim to maximize the energy efficiency by optimal and low complexity power allocation design. At first, an energy efficient power allocation optimization problem with sensing errors is formulated, under the total power and peak primary users' (PUs') interference constraints. Then, we focus on the analysis of the optimal non-convex power allocation problem, which is of great concern for the energy efficiency in CRNs. At last, utilizing the bisection method and the Lagrange dual decomposition, an optimal power allocation policy with low complexity is proposed. The numerical results show that energy-capacity can be well balanced by the proposed algorithm.