Learning speeds up wireless network design for virtual reality indoors
Learning-Accelerated Joint AP/RIS Optimization for Low-Latency Indoor mmWave XR Networks
Networking and Internet Architecture
Summary
Indoor virtual reality devices need fast and reliable wireless connections. The authors study how to use smart surfaces and multiple access points working together to improve these connections. They create a method that optimizes wireless signals but can be slow to run in real time. To fix this, they use learning to speed up the process while keeping good performance. This makes their approach better suited for virtual reality systems that need quick responses.
What this means in practice
- •For network operators: Deploy low-latency indoor wireless networks enhanced with intelligent surfaces for extended reality services using faster optimization enabled by learning.
- •For wireless system designers: Design millimeter wave indoor networks that jointly optimize access point transmissions and reflective surfaces efficiently for latency-critical applications.
Authors
Ali Elghariani, Mohammed Elbamby, Anis Elgabli
Abstract
This paper studies a reconfigurable intelligent surface (RIS)-assisted indoor architecture for extended reality (XR) applications, where multiple access points (APs) simultaneously serve multiple users over a shared frequency channel under a centralized AP controller. We formulate a non-convex sum-rate maximization problem with coupled AP precoders and RIS phase shifts, and develop a centralized alternating optimization (AO) framework for joint design. We rigorously prove that the proposed AO guarantees monotonic improvement of the sum rate that converges to a coordinate-wise stationary point, despite the non-convexity of the resulting subproblems. Although AO provides reliable performance, its iterative nature limits real-time applicability as the system dimension grows. To overcome this limitation, we propose a supervised learning-assisted approach that leverages AO-generated solutions to significantly accelerate convergence. Simulation results demonstrate that the proposed framework achieves near-optimal performance with substantially fewer AO iterations, which makes it well-suited for latency-critical RIS-assisted XR deployments.