Harvesting-Throughput Tradeoff for CDMA-Based Underlay Cognitive Radio Networks With Wireless Energy Harvesting

This paper considers a green cognitive radio network wherein each secondary user (SU) is solely powered by an energy harvester which extracts energy from RF signals of a primary user transmitter. Each SU operates in a harvesting鈥搕ransmitting fashion periodically. In each period, all SUs first harvest energy for a fixed duration and then transmit data using the harvested energy to one access point for the rest of the period in the code-division multiple access method. Considering the intrinsic harvesting-throughput tradeoff, we jointly optimize the harvesting time and transmit powers of SUs to maximize the sum throughput of the network, subject to the primary interference constraint and the energy causality constraint. Despite of the nonconvex nature of the formulated optimization problem, we find an equivalent but convex substitution to the original problem. Then, we propose a dual-decomposition-based solution method to solve the problem. Simulations finally demonstrate the efficiency of this paper.