TLDR: A research paper explores how Fiber-To-The-Room (FTTR) networks can enable high-quality immersive Extended Reality (XR) collaborations. It proposes a system that uses predictive bandwidth allocation and seamless handover mechanisms to overcome limitations of current wireless technologies like WiFi, ensuring low latency and high quality for demanding XR applications, even with 8K video. Experimental results show significant improvements in end-to-end latency and wireless transmission efficiency, making high-fidelity XR experiences achievable.
Extended Reality (XR), encompassing Augmented Reality (AR), Mixed Reality (MR), and Virtual Reality (VR), is rapidly transforming how humans interact with digital elements and remote machines. These immersive technologies are finding increasing applications in domestic entertainment and industrial settings. However, delivering a comfortable and high-quality experience for XR collaborations, especially with demanding requirements like 8K or 16K video frames, low end-to-end inter-frame latency (under 20 ms), and minimal jitter (under 15 ms), poses significant challenges for current wireless technologies like WiFi.
Traditional WiFi networks often suffer from low data rates, non-guaranteed bandwidth, and high latency, making them inefficient for the stringent demands of immersive XR. Even advanced systems like Beyond 5G mobile and WiFi-6/6E, while capable of supporting standalone XR devices, struggle in dense in-premise environments. This is where Fiber-To-The-Room (FTTR) and FTTR for Business (FTTR-B) emerge as promising solutions.
FTTR technology extends high-capacity Fiber-To-The-x (FTTx) networks directly into the premise, integrating seamlessly with WiFi access points (WAPs). This creates a robust network infrastructure, often involving an external Passive Optical Network (PON) connecting to main FTTR units (MFs), which then serve as Optical Network Units (ONUs) and Optical Line Terminals (OLTs) for subordinate FTTR units (SFs) supporting the WAPs. This architecture ensures high bandwidth availability closer to the end-user devices.
Despite the benefits of FTTR, challenges remain, particularly with resource allocation and seamless transitions between WAPs. Existing WiFi standards, even with different access classes, cannot always guarantee the Quality of Experience (QoE) for XR collaborations due to wireless channel contentions and uncoordinated resource allocation between the fiber and WiFi segments. This can lead to increased queueing latency and uplink jitter, especially under heavy network loads. Another critical issue arises when an XR device moves across the coverage areas of multiple WAPs, often resulting in the device remaining connected to a weaker signal source for too long, disrupting the immersive experience.
To address these limitations, researchers have proposed a Fiber-WiFi coordinated predictive resource allocation scheme combined with seamless handover for XR traffic. This innovative approach leverages the control interfaces of FTTR networks. It works by collecting data transmission statistics from XR stations (STAs) and WAPs at the SFs, relaying them to the MFs and the main OLT. Using machine learning techniques, the system predicts future bandwidth demands and data arrival times of XR frames. Based on these predictions, the WiFi control unit schedules channel access for WAPs and STAs, ensuring contention-free data transmissions.
For seamless handover, if an XR STA’s packet latency and jitter begin to exceed QoE requirements, or if its resource unit demand increases with lower-order modulation and transmit power, a control message prompts the STA to sense signals from neighboring WAPs. Once the STA reports back, the MF identifies a suitable neighboring WAP with sufficient available bandwidth and good channel conditions. Crucially, a network path between the new WAP and the XR application server is pre-established, ensuring a truly seamless transition without interruption to the XR experience.
Experimental studies have demonstrated the effectiveness of this framework. When using a traditional limited-service dynamic bandwidth allocation (LS-DBA), the QoE requirements for 8K XR quality (requiring 360 Mbps) were violated, although 2K and 4K quality could be maintained. However, with the proposed predictive bandwidth DBA (Pred-DBA), the average end-to-end latency of XR frames consistently remained at or below 15 milliseconds, even under various network loads, satisfying the stringent QoE requirements for high-quality XR. Furthermore, the seamless handover mechanism significantly reduced wireless latency for XR frame transmission. For instance, when a machine moved to a different room, handing over to a WAP with stronger signal strength reduced 8K XR frame latency by approximately 90% compared to remaining connected to a WAP at a distance of 20 meters.
Also Read:
- XRoboToolkit: Bridging Extended Reality and Robotics for Enhanced Teleoperation
- The AI Evolution in Virtual Worlds: How Language Models are Redefining VR Gaming
This research highlights that by intelligently coordinating fiber and WiFi resources with predictive allocation and seamless handover capabilities, FTTR networks can indeed provide the necessary infrastructure to support high-quality, immersive XR collaborations, overcoming the limitations of current wireless technologies. You can find more details in the full research paper: Enabling Immersive XR Collaborations over FTTR Networks.
