Web 5G

by Dominique Hazael-Massieux, W3C, and Dr. Jeff Jaffe, W3C

IEEE 5G Tech Focus: Volume 1, Number 3, September 2017 

Abstract

The advent of fifth generation (5G) networks creates opportunities to leverage network resources in new ways. Application platforms must also evolve to take advantage of very low latency, high throughput, and wide coverage for voice recognition, 3D video, UHD screens, virtual augmented reality, automation, smart cities/buildings, self-driving cars and other powerful applications. In this paper, we examine current and envisioned enhancements to the Open Web Platform that we believe will be essential to realizing the full potential of 5G.

1. Introduction

Fifth generation networks (5G) will provide higher bandwidth, lower latency and better coverage than today’s 4G networks.  Improvements to both the physical network (5G, Low Power Wide Access Network) and its control plane (Multi-access Edge Computing MEC, Software Defined Networking) will make the network more reactive, flexible, and with better performances, and will enable better cooperation among its end points.

In setting the direction for 5G, the International Telecommunication Union (ITU) envisioned [1] ambitious use cases. But while network capabilities are necessary to their realization, the application layer must also be up to the task.

There are many application platforms to choose from, but the most ubiquitous and scalable is the Open Web Platform — the collection of technologies that enables the Web. The Web plays a predominant role in the distribution of content and services: there are a billion websites, most connected devices feature browsers, and software developers identify as “Web developers” far more often than other roles [2].

2. The Web5G Project

Because of this reach, 5G stakeholders are driving enhancements to the Web to ensure a return on their investments in new network infrastructure and applications.

These companies have been doing this type of work at the World Wide Web Consortium (W3C) —which develops standards for the Web— for more than a decade. With 3G networks on the rise in 2005, W3C’s Mobile Web Initiative [3] began to push for Web capabilities to enable the rise of the mobile Internet.

Thus, as 5G standards emerge today, W3C is launching the Web5G project [4] to explore application, platform, and network layer (Figure 1) advances and requirements, including:

• support for new applications that will take advantage of 5G bandwidth and leverage very low latencies.
• increased application/network integration, specifically greater control of the network layer from Web applications.
• future enhancements to maximize the impact of network capabilities.

 

Figure 1: Web 5G accompanying network evolutions up to the application layer. 

Below we discuss current and anticipated W3C activities that will support the 5G vision.

3. Demand for 5G Capabilities

A number of recent advances in Web technology have led to significant increases in network traffic, underlining why it is so important to consider the evolution of the Web and network in tandem. Users will appreciate 5G improvements for network-hungry applications, including:

• Real-time audio and video. The Web Real-Time Communications (WebRTC) protocol enables real-time video and audio in both browsers and native mobile applications, without plug-ins. Analysts predict that 2 billion people will be making use of this technology by 2019 [5], revolutionizing telecommunications.
• In the past few years, video consumption has led to dramatic increases in network traffic, accelerated by the adoption of the HTML5 Web standard. It is easy to imagine the demand skyrocketing as the Web expands to accommodate UHD, HDR, and 360° videos.
• Virtual Reality. Though not yet common, there is a lot of excitement about adding immersive virtual reality experiences to the Web for commerce, communication, entertainment, and much more. A WebVR API [6] would make virtual reality a first class citizen of the Web.
• Web of Things. The Internet had existed for years when the Web was invented in 1989, but it took the harmonized application platform of the Web to usher in the Internet economy. W3C holds a similar view of the “Web of Things” as a uniform application layer that will make it much easier to develop Internet of Things applications that leverage 5G’s low latency and wide coverage.
• Automotive. Connectivity is driving automotive industry interest in platform enhancements for in-vehicle, inter-vehicle, and vehicle-to-infrastructure services. APIs from W3C that provide access to external sensors will benefit from 5G’s low latency.

Although different industries emphasize different use cases, in many cases the required capabilities are common to multiple industries. W3C’s diverse, international community of stakeholders helps build the Web with converging requirements across industries.

4. Integration of the Network Protocol Layer in the Open Web Platform

For many years, the telecommunications industry and other stakeholders at W3C have sought to improve integration of the network protocol layer with Web applications. These enhancements fall into three broad categories:

1. Monitoring of network performance. Recent Web standards enable fine-grain monitoring of network operations (Resource Timing [7] and WebRTC Statistics [8]).
2. Control of network transport characteristics. Recent Web standards support network warming-up (Resource Hints [9]), bidirectional communications (Web Sockets [10]), server-initiated operations (Push [11]), UDP-like transport properties (WebRTC [12]), and client-side proxies (Service Workers [13]).
3. Discovery of network services and interactions with those services. Recent Web standards enable LAN-provided services (Presentation API [14], Remote Playback API [15]), integration with firewall and NAT-traversal protocols (WebRTC), network-provided Push services, and new ways to initiate network operations (Web Background Synchronization [16]).

The Web5G project builds upon the lessons learned at W3C during the past decade, guiding the strategic expansion of the Open Web Platform to prepare for new network capabilities. In particular, we expect that 5G-enabled applications (e.g., automotive, e-health, augmented reality) will require even more fine-grain network customization. We also expect that as more and more network traffic is encrypted, approaches to optimization based on monitoring will lose their effectiveness. These changes call for a more ambitious evolution of the Open Web Platform architecture to address the needs of 5G.

5. An Architecture for Making the Open Web Platform 5G-Ready

To ensure that Web improvements for 5G applications scale globally, we must formulate a flexible architecture ready for customization around monitoring, control and service discovery. The Web5G project brings together telecommunications operators, network equipment providers, content delivery networks, browser vendors, and application developers so the architecture reflects different industry perspectives and constraints. For example:

• At Web scale, it is not practical to customize applications for every network type or operator. The integration mechanism should thus be network agnostic and fall back gracefully when used on legacy networks.
• We assume that, by default, we cannot trust applications and networks. Consequently, we expect that browsers —on behalf of the security and privacy of their users— will negotiate and mediate optimal network interactions.
• As more traffic is encrypted, the potential for reactivity and cooperation from the network can only be achieved if the endpoints of the network traffic (servers and user equipment) take the lead in informing networks about their specific requirements.

Figure 2 shows, at a high level, how browsers of the future could broker cooperation between applications and the network. In this architecture:

1. Upon connecting to a new network interface, the browser obtains the means to connect to a network configuration and monitoring end-point; the exact protocols for this would probably differ from one type of network to another - in networks where only well-known devices can connect, this could be as simple as a predefined browser configuration entry, but obviously a more scalable and robust approach would be required to reach more devices.
2. The browser obtains information about the current and changing networking conditions from that endpoint. Application developers would use this information to tailor application behavior.
3. The browser also communicates to the network via this endpoint an optimal network configuration for a given application. User preferences would shape these requests. User preferences include both manual configurations (by the user or their network/system administrator) or those computed through heuristics such as "frecency" [17].

 

Figure 2: Sequence Diagram of Web & Network integration architecture 

A number of recent projects have adopted similar approaches to brokering, suggesting the Web5G architecture can be achieved:

• Google worked with 3 telecommunication operators in India [18] to make it easier for their users to download YouTube videos during network off-hours under the "Smart Offline" banner. That cooperation is based on a general mechanism for operators to collaborate with Google applications via the Mobile Data Plan API [19] — that API could be generalized. Google also experimented with good results in using MEC Throughput Guidance [20] to enhance video streaming performance.
• The operator-assisted relay service specification [21] proposes an architecture where apps and networks cooperate in setting up optimal relays.
• The collaboration between Cisco and Apple on iOS networking [22] in enterprise networks illustrates how the cooperation of the application layer (mediated by the operating system in this case) with the networking layer (as exposed by Cisco routers) can lead to improved performances for identified applications.

6. Conclusion

Defining and deploying the next generation network involves many technical and business challenges, many known and others not yet known. The Web5G project underway at W3C is conducting a systematic analysis of current performance bottlenecks and how that picture will change with the adoption of 5G. We expect this effort to increase in 2017 with a W3C Workshop (likely a 2-day event) and formally chartered steering committee. All stakeholders are invited to join this effort to realize the full potential of 5G.

References: 

1. “IMT Vision – Framework and overall objectives of the future development of IMT for 2020 and beyond”, Recommendation ITU-R M.2083-0, September 2015. Available: https://www.itu.int/dms_pubrec/itu-r/rec/m/R-REC-M.2083-0-201509-I!!PDF-E.pdf
2. “Developer Roles“ in “Developer Survey Results”, StackOverflow, 2017. Available: https://insights.stackoverflow.com/survey/2017#developer-roles
3. “W3C Launches "Mobile Web Initiative", W3C Press Release, May 2005. Available: https://www.w3.org/2005/05/mwi-pressrelease.html.en
4. “Web5G Roadmap”, May 2017. Available: https://w3c.github.io/media-web-roadmap/web5g/
5. Bubbley, “2015 Q1 Update: WebRTC Market Status & Forecasts Report”, 2015. [Online]. Available: http://disruptivewireless.blogspot.fr/p/blog-page_30.html
6. Vukicevic, B. Jones, K. Gilbert, C. Van Wiemeersch, N. Waliczek, R. Cintron, “WebVR”, Draft W3C Community Group Report, April 2017. Available: https://w3c.github.io/webvr/spec/latest/
7. Jain, T. Reifsteck, J. Mann, Z. Wang, A. Quach, “Resource Timing Level 1”, W3C Candidate Recommendation, March 2017. Available: https://www.w3.org/TR/resource-timing/
8. Alvestrand, V. Singh, “Identifiers for WebRTC's Statistics API”, W3C Working Draft, December 2016. Available: https://www.w3.org/TR/webrtc-stats/
9. Grigorik, “Resource Hints”, W3C Working Draft, March 2017. Available: https://www.w3.org/TR/resource-hints/
10. Hickson, “The Web Socket API”, W3C Candidate Recommendation, September 2012. Available: https://www.w3.org/TR/websockets/
11. Beverloo, M. Thomson, M. van Ouwerkerk, B. Sullivan, E. Fullea, “Push API”, W3C Working Draft, April 2017. Available: https://www.w3.org/TR/push-api/
12. Bergvist, D. Burnett, C. Jennings, T. Brandsetter, B. Aboba, A. Narayanan, “WebRTC 1.0: Real-time Communication Between Browsers“, W3C Working Draft, June 2017. Available: https://www.w3.org/TR/webrtc/
13. Russell, J. Song, J. Archibald, M. Kruisselbrink, “Service Workers 1”, W3C Working Draft, October 2016. Available: https://www.w3.org/TR/service-workers/
14. Foltz, D. Röttsches, “Presentation API”, W3C Candidate Recommendation, June 2017. Available: https://www.w3.org/TR/presentation-api/
15. Vayvod, M. Lamouri, “Remote Playback API”, W3C Working Draft, November 2016. Available: https://www.w3.org/TR/remote-playback/
16. Karlin, M. Kruisselbrink, “Web Background Synchronization”, Draft W3C Community Group Report , August 2016. Available: https://wicg.github.io/BackgroundSync/spec/
17. “Frecency algorithm”, Mozilla Developer Network. Available: https://developer.mozilla.org/en-US/docs/Mozilla/Tech/Places/Frecency_algorithm
18. Jain, “Partnering toward the next generation of mobile networks”, Google Blog, February 2017. Available: https://blog.google/topics/internet-access/partnering-toward-next-generation-mobile-networks/
19. “Google Mobile Data Plan Sharing API”. Available: https://developers.google.com/mobile-data-plan/reference/rest/
20. Jain, A. Terzis, H. Flinck, N. Sprecher, S. Arunachalam, K. Smith, V. Devarapalli, R. Bar Yanai, “Mobile Throughput Guidance Inband Signaling Protocol”, IETF Internet Draft, March 2017. Available: https://tools.ietf.org/html/draft-flinck-mobile-throughput-guidance-04
21. Wang, B.Liu, J.Uberti, P. Ding, “Operator-Assisted Relay Service Architecture (OARS)”, IETF Internet Draft, April 2017. Available: https://tools.ietf.org/html/draft-wang-rtcweb-oars-02
22. “Optimized WiFi Connectivity and Prioritized Business Apps”, Cisco. [Online]. Available: http://www.cisco.com/c/dam/en/us/td/docs/wireless/controller/technotes/8-3/Optimizing_WiFi_Connectivity_and_Prioritizing_Business_Apps.pdf 

 

 Dominique Hazael-Massieux holds an engineering degree from France “Grande Ecole” École Centrale Paris. He leads the Telecommunication Industry relationship effort in W3C, the World Wide Web Consortium, ensuring that the needs of this industry are taken into account in the context of the evolution of the Web. He is also responsible for the Web Real-Time Communications Working Group, the Device and Sensors Working Group and leading the efforts to bring Virtual Reality to the Web.

 

 

Dr. Jeff Jaffe  is Chief Executive Officer of the World Wide Web Consortium. In this role he works with Director Tim Berners-Lee, staff, and membership, and the public to evolve and communicate the W3C's vision. He is responsible for all of W3C's global operations, for maintaining the interests of all of W3C's stakeholders, and for sustaining a culture of cooperation and transparency, so that W3C continues to be the leading forum for the technical development and stewardship of the Web.  After receiving a Ph.D. in computer science from MIT in 1979, Jeff joined IBM's Thomas J. Watson Research Center. During his tenure at IBM, he held a wide variety of technical and management positions, including vice president, Systems and Software Research, corporate vice president of technology, and general manager of IBM's SecureWay business unit.   Jeff then served as president of Bell Labs Research and Advanced Technologies and Executive Vice President and Chief Technology Officer for Novell.  He is a Fellow of ACM and the IEEE. 

Editor: Siming Zhang