Source-linked AI summary
Cellular-Connected Wireless Virtual Reality: Requirements, Challenges, and Solutions
Fenghe Hu, Yansha Deng, Walid Saad, Mehdi Bennis, A. Hamid Aghvami
TL;DR
Cellular-connected VR requires QoS that reflects human perception because immersive services demand ubiquitous connectivity, high rates, reliability, and low interaction latency. The paper maps perceptual requirements to four VR phases, analyzes use cases and enabling technologies, and uses a case study to show that dual connectivity can improve support for VR streaming in cellular networks.
Problem
Cellular-connected VR must provide seamless service despite unstable channels, mobility and handovers, asymmetric traffic, and stringent real-time requirements.
Method
The paper maps human perception to QoS requirements across four VR phases, analyzes four use cases and solutions, and evaluates cellular support through a case study.
Results
Dual connectivity can potentially support VR streaming with new QoS requirements in cellular mmWave networks and improve performance under unstable wireless channels.
Takeaways & Limitations
VR transmission requires joint rate, reliability, and latency support because 6DoF rendering needs real-time low-interaction-latency delivery unlike 3DoF video buffered through jitter.
Abstract
from arXiv · showhide
Cellular-connected wireless connectivity provides new opportunities for virtual reality(VR) to offer seamless user experience from anywhere at anytime. To realize this vision, the quality-of-service (QoS) for wireless VR needs to be carefully defined to reflect human perception requirements. In this paper, we first identify the primary drivers of VR systems, in terms of applications and use cases. We then map the human perception requirements to corresponding QoS requirements for four phases of VR technology development. To shed light on how to provide short/long-range mobility for VR services, we further list four main use cases for cellular-connected wireless VR and identify their unique research challenges along with their corresponding enabling technologies and solutions in 5G systems and beyond. Last but not least, we present a case study to demonstrate the effectiveness of our proposed solution and the unique QoS performance requirements of VR transmission compared with that of traditional video service in cellular networks.
I. INTRODUCTION
The paper frames cellular-connected wireless VR as a way to overcome restricted device mobility and insufficient real-time content support. It defines human-perception-based QoS, analyzes four development phases and four use cases, and evaluates cellular support through simulation.
- Restricted mobility in wired VR devices and limited real-time high-quality content in all-in-one devices hinder mass-market VR applications.
- Cellular connectivity could provide ubiquitous VR experiences from anywhere and at any time while enabling novel applications.
- The paper maps human perceptions to QoS requirements across four VR development phases and corresponding device parameters.
- It examines four cellular-connected VR use cases, their QoS requirements, research challenges, and potential 5G solutions.
- A practical simulation evaluates the challenge of supporting cellular-connected VR with a state-of-the-art cellular network.
II. VR SERVICE REQUIREMENT
The paper defines immersive VR service requirements from human perception and relates them to evolving VR device phases. Relevant perceptual factors include resolution, field of view, refresh rate, and interaction latency.
- VR QoS requirements are derived from human perception requirements and detailed for four phases of VR device evolution.
- Human perception varies with factors such as age, health, and occupation, so the paper reports average values for healthy people aged 12–18.
- Resolution requirements are tied to visual acuity, with more than 720 pixels per degree required to fully satisfy the stated human-eye requirement in an ideal environment.
2) Field-of-View:
Field of view describes the area visible to the eye, while refresh rate governs motion continuity and interaction latency measures response time to user movement. These perceptual factors impose demanding VR display and responsiveness requirements.
- 2) Field-of-View:: The human-eye field of view is specified as 210°×150°, or a 200° diagonal field of view.
- 2) Field-of-View:: Central, peripheral, and monocular vision zones have progressively lower resolution requirements, while commercial headsets reach at most 150° field of view.
- 3) Refresh Rate:: Refresh rate is the number of displayed frames per second; human-perception bounds range from 120 Hz for motion continuity to 1800 Hz for stringent pixel continuity.
- 4) VR Interaction Latency:: VR interaction latency measures the time from a user’s movement until the virtual environment responds, with a target below 20 ms and a common minimum of 10 ms.
B. Mapping Human Perception Requirements to QoS
The paper maps perceptual limits to communication QoS requirements including data rate, error rate, and delay. The resulting requirements are demanding because interaction latency also includes rendering and coding or decoding delays.
- Human perception limits are mapped to concrete VR QoS requirements such as data rate, error rate, and communication delay.
- 50.39 Gbps is the minimum downlink data rate for 360-degree VR transmission at 60 PPD and 120 Hz under 20:1 compression.
- The 10 ms VR interaction-latency requirement includes communication, rendering, and video coding or decoding latency.
- The progressive data-rate expression combines bits per color, pixels per degree, field of view, refresh rate, and compression ratio.
3) Error Rate:
VR development phases define changing device specifications and service requirements, while motion information demands zero error and stream-data errors should remain below 10^-6.
- Motion information requires zero error because inaccuracies can trigger extra processing to regenerate incorrect pixels.
- Stream-data error rates should remain below 10^-6 to avoid human-detectable lost, degraded, or damaged frames.
- VR devices progress through Pre-VR, Entry-level VR, Advanced VR, and Ultimate VR phases with phase-specific technical and service requirements.
III. USE CASES, CHALLENGES, AND POTENTIAL SOLUTIONS
Cellular-connected VR traffic differs from conventional video through stringent, motion-dependent, and asymmetric requirements, motivating distinct use cases and network challenges.
- VR requires seamless low-delay, high-capacity service for real-time rendering and interaction.
- VR content changes with users’ real-time motion, linking transmission demands to user movement.
- Uplink and downlink rates are asymmetric and coupled because both correlate with end-to-end VR performance.
- LTE supports around 15 Mbps with 10–30 ms network delay, below pre-VR requirements, while 5G targets above 50 Mbps everywhere and up to 1 Gbps downlink, reaching Advanced VR data-rate requirements.
- The four cellular-connected VR use cases are automotive video streaming, social sharing at crowded venues, 6 degree-of-freedom streaming, and remote control or tactile Internet.
A. Automotive Video Streaming
Automotive video streaming targets live VR for commuters in high-speed trains and cars, requiring seamless cellular connectivity at 100 Mbps under high mobility.
- VR-AVS enables live VR video streaming for commuters in high-speed trains and cars.
- Supporting VR-AVS requires uniform, seamless connectivity at 100 Mbps for high-mobility users.
- Dual connectivity can maintain seamless service by allowing multiple base stations to transmit and receive data simultaneously.
2) Coordinated Multipoint Transmission and Reception:
For mobility- and latency-sensitive VR, coordinated multipoint transmission, caching, slicing, edge computing, and motion prediction address connectivity and response-time challenges.
- Coordinated Multipoint Transmission and Reception: CoMP serves VR-AVS users through multiple base stations over the same spectrum to reduce data-rate fluctuations caused by blockage during mobility.
- Remote Control and Tactile Internet: VR-RC remotely controls UAVs and robots over long distances, requiring seamless connectivity with latency within 5 ms.
- Remote Control and Tactile Internet: Caching, network slicing, and edge computing are identified as existing approaches for reducing latency in VR-RC.
- Motion Prediction: Motion prediction can pre-fetch users’ fields of view, but systems must tolerate prediction errors and regenerate correct content within the delay threshold.
C. The 6 Degree-of-freedom Content Streaming
6DoF content streaming targets an immersive environment requiring high data rates, low latency, accurate positioning, and substantial computing support. The paper discusses high-frequency transmission and edge computing as enabling approaches, while noting reliability limits for non-line-of-sight channels.
- C. The 6 Degree-of-freedom Content Streaming: 6DoF streaming requires 400–600 Mbps data rates, 5–20 ms latency, and accurate VR-user positioning in the Advanced VR phase.
- C. The 6 Degree-of-freedom Content Streaming: Ultimate VR with 20:1 compression and ultra-low latency requires more than 100 Gbps throughput, motivating mmWave and THz transmission.
- C. The 6 Degree-of-freedom Content Streaming: High-frequency communication is unreliable over non-line-of-sight channels because increasing frequency raises penetration loss through solid materials.
- C. The 6 Degree-of-freedom Content Streaming: Proposed countermeasures include reflectors, proactive mobility management, dynamic frame structures, and integrating mmWave with sub-6 GHz frequencies.
- C. The 6 Degree-of-freedom Content Streaming: Edge computing can reduce computing and end-to-end transmission delay by placing rendering and virtual-environment processing at base stations, edge servers, or VR devices.
D. Social Sharing at Crowded Venues
Social sharing at crowded venues requires very high bidirectional capacity and support for diverse, latency-critical VR services. The case study shows that VR streaming achieves fewer successful deliveries within 7 ms than traditional video, while dual connectivity improves resilience to unstable channels.
- D. Social Sharing at Crowded Venues: VR social sharing at crowded venues requires 12.5 Tbps/km2 in both uplink and downlink directions while integrating applications with diverse QoS requirements.
- D. Social Sharing at Crowded Venues: Massive MIMO can provide multi-gigabit rates over broad coverage and supply angle-of-arrival information useful for positioning VR users.
- D. Social Sharing at Crowded Venues: 6DoF VR streaming requires real-time low-interaction-latency transmission, unlike 3DoF video, which can tolerate unstable QoS through a jitter buffer.
- D. Social Sharing at Crowded Venues: The case study measures successful bitplanes delivered within a 7 ms downlink constraint as user count increases over five minutes.
- D. Social Sharing at Crowded Venues: VR streaming delivers a much lower percentage of successful bitplanes within 7 ms than traditional video, and performance declines as users and interference increase.
- D. Social Sharing at Crowded Venues: Dual connectivity combats unstable wireless channels by connecting each VR device with two base stations simultaneously, improving case-study performance.
V. CONCLUSION
The paper defines QoS requirements for four wireless VR phases from human perception, analyzes four cellular-connected VR use cases, and evaluates dual connectivity for VR streaming. It concludes that cellular-connected VR requires specialized solutions such as motion prediction, massive MIMO, and edge computing.
- V. CONCLUSION: The paper defines QoS requirements for four wireless VR technological phases based on human perception requirements.
- V. CONCLUSION: It presents four cellular-connected VR use cases, quantifies their QoS requirements, and discusses corresponding solutions and trade-offs.
- V. CONCLUSION: The case study shows that cellular mmWave networks can potentially support VR streaming with new QoS requirements using dual connectivity.
- V. CONCLUSION: The paper identifies motion prediction, massive MIMO, and edge computing as technologies for customizing optimization solutions for cellular-connected VR use cases.