Introduction Extended Reality (XR) applications have been widely adopted in gaming, healthcare, professional training, and social interaction. Among them, networked multi-user XR (MU-XR) systems are increasingly important. They allow users to communicate and collaborate remotely by streaming real-time data. Moreover, MU-XR systems are becoming increasingly interactive, offering a variety of modes and sensors for user interaction and even allowing users to share their social engagements. Hence, global network conditions increasingly shape user experience (UX). However, current performance engineering remains largely network-centric and predominantly evaluates network capabilities from an individual-user perspective, thereby overlooking how global network resource management (NRM) affects multi-user interaction, multi-user UX, and immersion.

Methods This has led to a lack of a unified framework that can connect multi-user experiences from an application perspective with underlying NRM. To address this gap, this paper conducts a systematic literature review (SLR) of 56 studies published between 2015 and 2025, analyzing UX evaluation methods, multi-user network performance metrics, and NRM-related network functions.

Results The findings reveal that most studies focus on latency minimization to ensure a real-time, immersive collaborative experience, while only a few address network-level support for resource allocation and synchronization for user groups that require immersion. Among the UX evaluation methods, only 6 of the 56 studies conducted user studies, indicating a clear lack of empirical support.

Discussion Furthermore, this SLR proposes a new model for structuring NRM requirements across various MU-XR scenarios, starting at the XR application level, identifying immersion techniques at the XR level, and connecting them to global network performance, aiming to bridge MU-XR applications and their NRM requirements.