Energy-Efficient Resource Allocation in Multiple UAVs-Assisted Energy Harvesting-Powered Two-Hop Cognitive Radio Network
He Xiao,Chun Wu,Hong Jiang,Li-Ping Deng,Ying Luo,Qiu-Yun Zhang
DOI: https://doi.org/10.1109/jsen.2023.3247436
IF: 4.3
2023-04-05
IEEE Sensors Journal
Abstract:The unmanned aerial vehicle (UAV)-assisted energy harvesting (EH)-based cognitive wireless networks can solve the problems of spectrum scarcity, energy shortage, and challenging layout of communication facilities faced by traditional wireless networks, envisioned as a vital frontier communication technology in the 5G/6G era. In this article, we consider a UAVs-assisted two-hop cognitive radio network (CRN) with EH (UAVs-EH-DHCRN), where multiple UAVs are employed to act as air access points (AAPs) and relay nodes (RNs). Since all EH-powered secondary users (SUs) can simultaneously associate with AAPs by sharing the licensed spectrum of primary users (PUs), intercell interference (ICI) occurs among AAPs. Our goal is to maximize the end-to-end energy efficiency (EE) of the UAVs-EH-DHCRN while considering constraints like ICI, energy, minimum data rate, and interference to PUs. However, the formulated end-to-end EE maximization problem is a typical nonconvex optimization problem with complex constraints that off-the-shelf optimization strategies cannot solve. Therefore, we propose a resource allocation algorithm called approximate convex policy concerning associations, power allocation, and placements (ACP-APP). Specifically, in ACP-APP, we introduce additional parameters to convert the fractional objective function into a polynomial function. Then, the optimal AAPs' placements, optimal associations between SUs and AAPs, and power allocation are obtained alternately by leveraging the proposed circular-based recursive random search (CRRS), 1-D linear programming, and Frank–Wolfe (FW) policy. Simulation results show that our proposed strategy can converge quickly to obtain the approximate optimal solution. Moreover, the ACP-APP can obtain significant end-to-end EE gains compared with benchmark schemes.
engineering, electrical & electronic,instruments & instrumentation,physics, applied