Adaptive mobile VR content delivery for industrial 5.0

Abstract

Mobile virtual reality (VR) is expected to be a key component of the next-generation industrial internet-of-things, which uses immersive technologies to boost virtualization and facilitate human-machine collaboration in Industry 5.0. In this paper, we design a VR content delivery scheme to enhance VR content playback quality in mobile edge computing. The proposed scheme schedules computing resources on network edge to satisfy VR content requests from multiple user devices while reducing the likelihood of rebuffering and improving content freshness during VR video playback. With limited computing resources at the edge server, we develop a deep reinforcement learning (DRL) approach to determine which requests should be satisfied first, given the network and the service dynamics. By analyzing the network dynamics using the Whittle index method, the proposed DRL-based scheme can improve VR service quality with minimal communication overhead in computing scheduling. Simulation results demonstrate that the proposed scheme significantly improves the quality of service for VR content delivery.