Joint Energy and Throughput Optimization in Energy Harvesting Cognitive Sensor Networks

In this study, we explore the energy efficient power allocation scheme in an energy harvesting cognitive sensor network where some distributed sensors with energy harvesting capability attempted to access the legitimate frequency band under the limit of energy consumption and interference avoidance. Each sensor, as a Secondary User (SU), participate in cooperative spectrum sensing by exploiting energy detection and send the results to the Fusion Center (FC). Then, the FC determines whether PU is present on the channel or not based on obtained energy values. If the channel is detected to be idle, SUs first harvest energy from a dedicated power beam, which is broadcasted by the FC, and then allow the access to the channel and send their information data. The challenge of power allocation and spectrum access in order to reach maximum achievable data rate and minimum energy consumption at the secondary network, is formulated as an optimization problem. It is shown that this problem is nonlinear fractional programming, and the well-known Dinkelbach algorithm to deal with it is exploited. Simulation results showed that the proposed algorithm considerably improves the energy efficiency of the secondary network in comparison to the conventional cognitive network.