Reconfigurable Intelligent Surface Assisted Mobile Edge Computing With Heterogeneous Learning Tasks

The ever-growing popularity and rapid development of artificial intelligence (AI) have raised rethinking on the evolution wireless networks. Mobile edge computing (MEC) provides a natural platform for AI applications since has rich computation resources to train machine learning (ML) models, as well low-latency access data generated by mobile Internet things (IoT) devices. In this article, we present an infrastructure perform ML tasks at MEC server with assistance reconfigurable intelligent surface (RIS). contrast conventional communication systems where principal criterions are maximize throughput, aim maximizing performance. Specifically, minimize maximum error all participating users jointly optimizing transmit power users, beamforming vectors base station (BS), phase-shift matrix RIS. An alternating optimization (AO)-based framework is proposed optimize three terms iteratively, successive convex approximation (SCA)-based algorithm developed solve allocation problem, closed-form expressions derived design direction method multipliers (ADMM)-based designed efficiently problem. Simulation results demonstrate significant gains deploying RIS validate advantages our algorithms over various benchmarks. Lastly, unified sensing-communication-learning based CARLA simulator SECOND network, use case (3D object detection in autonomous driving) scheme demonstrated platform.