MEC中RIS辅助的无人机轨迹和计算策略联合优化

Joint Optimization of RIS-assisted UAV Trajectory and Computing Policy for MEC

在B5G/6G网络中,尽管无人机(Unmanned Aerial Vehicle, UAV)可以作为移动边缘计算(Mobile Edge Computing, MEC)的服务器为地面终端(Ground Terminal, GT)提供通信和计算服务,但仍然面临着因为移动性而导致通信链路被周围障碍物阻挡的挑战。可重构智能表面(Reconfigurable Intelligent Surface, RIS)可以有效地辅助UAV改善与GT的通信链路质量,保证MEC的时延要求。提出了一种RIS辅助的UAV轨迹和计算策略联合优化方案,以最小化MEC的服务能耗为目标,联合优化UAV的三维轨迹、计算任务分发和缓存资源分配。利用连续凸逼近(Successive Convex Approximation, SCA)方法对原始的非凸联合优化问题进行了求解。仿真实验中,选取UAV轨迹固定和计算策略固定的方案为对比依据,验证了所提方案的有效性。结果表明,所提方案在能耗和数据传输速率上均有明显的性能提升。

In B5G/6G networks, although the Unmanned Aerial Vehicle(UAV) can be used as a server for Mobile Edge Computing(MEC) to provide communication and computing services for Ground Terminals(GT), there still exists the challenge of communication link being blocked by surrounding obstacles due to mobility. Reconfigurable Intelligent Surface(RIS) can effectively assist UAV to improve the communication link quality with GT and ensure the delay requirements of MEC service. A joint optimization scheme for RIS-assisted UAV trajectory and computing policy is proposed, which jointly optimize UAV 3D-trajectory, computation task distribution and cache resource allocation to minimize the energy consumption of MEC service. The original non-convex joint optimization problem is solved by Successive Convex Approximation(SCA) method. In the simulation experiment, the scheme with fixed UAV trajectory and the scheme with fixed computing policy are selected as the benchmarks to verify the effectiveness of the proposed scheme. The results show that the proposed scheme has significant performance improvement on energy consumption and data transmission rate.