ResAll: Energy Efficiency Maximization for Wireless Energy Harvesting Sensor Networks

Energy harvesting is a promising solution to prolong the lifetime of energy-constrained wireless sensor networks. In particular, scavenging energy from ambient radio frequency (RF) signals has drawn a lot of attention recently. In this paper, we apply simultaneous wireless information and power transfer (SWIPT) to a clustered sensor network such that a cluster head node harvests the wireless energy of received RF signals from its cluster members and then employs the harvested energy to compensate the energy consumed by data aggregating and forwarding. In such a network, how to achieve high energy efficiency through trading off between energy harvesting and information decoding is a critical issue. To this end, we formulate the rate and power resource allocation problem in a clustered WSN with SWIPT as a non-convex constrained energy efficiency maximization problem. By exploiting fractional programming and dual decomposition, we further propose a cross-layer resource allocation (ResAll) algorithm consisting of subalgorithms of rate control, power allocation and power splitting to solve the problem efficiently and optimally. Our simulation results reveal that the proposed ResAll algorithm converges within a small number of iterations, and achieves optimal system energy efficiency by balancing energy efficiency, data rate, transmit power and power splitting ratio.