Opportunistic Throughput Optimization in Energy Harvesting Dynamic Spectrum Sharing Wireless Networks

We investigate opportunistic transmissions in a time-slotted wireless network, emphasizing constraints arising from finite durations allocated to various network operations and the availability of energy for these operations. Each time frame (time slot) comprises three sub-frames: sensing, reporting, and either transmission or energy harvesting based on the presence or absence of the primary user. We assume a fixed duration for the transmission sub-frame within each time frame. Utilizing a time division multiple access (TDMA) protocol, we manage local sensing data reporting within each time frame; consequently, the reporting time is contingent on the number of users. As a result, with the total time allocated for sensing and reporting being fixed, a trade-off arises between the number of collaborating users and the number of samples. Additionally, energy limitations and causality lead to two scenarios for wireless network operation: energy-deficit and energy-surplus regimes. To address this complexity, we formulate an optimization problem aimed at maximizing overall network throughput while considering constraints imposed by finite durations for various network operations and energy availability. We provide analytical evidence of the convexity of the optimization problem in both energy-deficit and energy-surplus scenarios. Furthermore, we propose two algorithms designed to achieve optimal throughput for each scenario. The accuracy of our analyses is validated through Monte Carlo simulations. Numerical results demonstrate the effectiveness of our proposed approach.