PRU Group Allocation and Dynamic Rate-Splitting Design for Power Minimization in IRS-Assisted UAV MEC Systems With RSMA

Given the rapid development of unmanned aerial vehicles (UAVs) and communication technology, the applications of various low-altitude UAVs have become research hotspots in recent years. However, computation capability and power consumption have been the main reasons restricting its development. For mitigating the power consumption pressure of UAVs, we propose a novel scheme for power minimization in the intelligent reflecting surface (IRS)-assisted UAV mobile edge computing (MEC) system with rate-splitting multiple access (RSMA), where the passive reflecting units (PRUs) of IRS are divided into groups for allocation, and the split proportional indices of RSMA is dynamically designed. Furthermore, we define the power minimization problem and subsequently present a solution, namely passive reflecting group allocation and dynamic rate-splitting design (PGADRD) algorithm. By presenting the analytical solution pertaining to optimal phase manipulation and decoding sequence, the power minimization problem is simplified. Then, we partition the simplified problem into three subproblems employing the block coordinate descent method, subsequently converting these subproblems into convex subproblems utilizing the successive convex approximation technique. The optimal resource allocation and multi-UAV trajectories are obtained by alternately solving these convex subproblems. Lastly, our simulation results show that the PGADRD scheme is beneficial in reducing UAVs’ power consumption, which is more obvious with significant path attenuation and computational burdens.