Spatial Computing Communications for Multi-User Virtual Reality in Distributed Mobile Edge Computing Network

TL;DR

The paper addresses delivering low-latency, energy-efficient multi-user VR over distributed MEC by introducing spatial computing communications (SCC) that couple physical space (users, BSs) with virtual space (shared VR environments) and formulating a multi-objective optimization (MOCO) between total latency $T(\boldsymbol{x},\boldsymbol{y})$ and energy $E(\boldsymbol{x},\boldsymbol{y})$. A novel MO-CMPO algorithm combines a consistency-model-based supervised learning stage with reinforcement-learning fine-tuning guided by preference weights, using a sparse graph neural network to efficiently generate Pareto-optimal deployment plans. Extensive simulations on real NR base-station datasets show that MO-CMPO achieves superior hypervolume and significantly lower inference latency than baselines, while revealing deployment patterns that favor local MECs for latency and reduced replication for energy. The work provides a scalable, preference-aware framework for resource deployment in immersive distributed MEC environments, with practical insights into how deployment decisions trade off caching, synchronization, and cross-MEC transmissions. These advances enable more scalable and energy-efficient multi-user VR in future 6G/MEC networks.

Abstract

Immersive virtual reality (VR) applications impose stringent requirements on latency, energy efficiency, and computational resources, particularly in multi-user interactive scenarios. To address these challenges, we introduce the concept of spatial computing communications (SCC), a framework designed to meet the latency and energy demands of multi-user VR over distributed mobile edge computing (MEC) networks. SCC jointly represents the physical space, defined by users and base stations, and the virtual space, representing shared immersive environments, using a probabilistic model of user dynamics and resource requirements. The resource deployment task is then formulated as a multi-objective combinatorial optimization (MOCO) problem that simultaneously minimizes system latency and energy consumption across distributed MEC resources. To solve this problem, we propose MO-CMPO, a multi-objective consistency model with policy optimization that integrates supervised learning and reinforcement learning (RL) fine-tuning guided by preference weights. Leveraging a sparse graph neural network (GNN), MO-CMPO efficiently generates Pareto-optimal solutions. Simulations with real-world New Radio base station datasets demonstrate that MO-CMPO achieves superior hypervolume performance and significantly lower inference latency than baseline methods. Furthermore, the analysis reveals practical deployment patterns: latency-oriented solutions favor local MEC execution to reduce transmission delay, while energy-oriented solutions minimize redundant placements to save energy.

Figures (10)

• Figure 1: An illustration of the system model.
• Figure 2:
• Figure 3:
• Figure 4:
• Figure 6: An illustration of proposed multi-objective optimization with consistency model.