STAR-RIS Assisted Offloading Based on Hybrid NOMA: A Time Minimization Perspective

The simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) has emerged as a novel concept and promising technology for future wireless networks. In this article, we propose a STAR-RIS assisted multi-access edge computing (MEC) framework, where the user offloads the tasks to multiple edge nodes, and the STAR-RIS is deployed to enable and enhance the multiple offloading links at the same time. Meanwhile, the hybrid non-orthogonal multiple access (h-NOMA) is adopted to further improve the performance of the STAR-RIS assisted downlink parallel offloading. We aim to minimize the latency by jointly optimizing the offloading order and task division to the edge nodes, the time allocation for the time-division multiple access (TDMA) transmission and the NOMA transmission in the h-NOMA scheme, the transmit power and the decoding orders when employing NOMA, and the transmitting and reflecting coefficients at the STAR-RIS. To solve this highly complex optimization problem, we first derive the optimal solutions of the time allocation and task division in closed forms, and then propose a low-complexity alternative optimizing (AO) algorithm based on successive convex approximation (SCA) technique to optimize the allocation of radio resources. Simulation results show that the proposed STAR-RIS assisted offloading scheme based on h-NOMA can effectively reduce the time latency compared with benchmark schemes, especially when the number of the STAR-RIS elements is small.

I. Introduction

Driven by the rapid development of Internet-of-Things (IoTs), increasing amount of computation-intensive and latency-critical data is generated and processed at the network edge [1]. However, the IoT devices with limited computing resources and stringent battery capacity may suffer from a serious time delay when processing the computation-intensive tasks [2]. To tackle this problem, the multi-access edge computing (MEC) was proposed, which enables the IoT devices to offload their tasks to the edge nodes with rich computation resources and thus offers low-latency computation service. However, when the wireless environments are unfavorable, for example, when the devices locate far from the edge servers (E-Servers) or the transmission links are blocked by obstacles, the wireless offloading still suffers from high transmission latency and low offloading success rate [3], which severely prevent the efficiency of MEC.