Edge Intelligence-Based Ultra-Reliable and Low-Latency Communications for Digital Twin-Enabled Metaverse

In this letter, we propose a novel digital twin scheme supported metaverse by jointly considering the integrated model of communications, computing, and storage through the employment of mobile edge computing (MEC) and ultra-reliable and low latency communications (URLLC). The MEC-based URLLC digital twin architecture is proposed to provide powerful computing infrastructure by exploring task offloading, and task caching techniques in nearby edge servers to reduce the latency. In addition, the proposed digital twin scheme can guarantee stringent requirements of reliability and low latency, which are highly applicable for the future networked systems of metaverse. For this first time in the literature, our paper addresses the optimal problem of the latency/reliablity in digital twins-enabled metaverse by optimizing various communication and computation variables, namely, offloading portions, edge caching policies, bandwidth allocation, transmit power, computation resources of user devices and edge servers. The proposed scheme can improve the quality-of-experience of the digital twin in terms of latency and reliability with respect to metaverse applications.

I. Introduction

It is expected that the next generation of wireless communications network will revolutionise human lives in an unprecedented way. With a growing trend in the development of mobile wireless augmented reality (AR) and virtual reality (VR), the manufacturing empowered by industrial Internet-of-Things (IIoT) has been transformed from physical prototypes to virtual and immersive interactions, which eventually will profoundly enhance the efficiency of productivity. Recently, leveraged by real-time optimization theory, artificial intelligence, and digital twin (DT) has been considered as a promising technique to realise the practical implementation of metaverse. For supporting immersive and interoperable metaverse, DT, implemented through computer simulation but completely different from the computer model, will represent digital replicas of physical objects with a real-time two-way interaction. As such, DT provides real-time insight into how the metaverse can be operated and help to optimise the decision-making.