GSMA Intelligence forecasts that the number of 5G connections globally will reach 1.3 billion by 2025, covering 40 percent of the world’s population or approximately 2.7 billion people. At that time, the Americas region is expected to account for over 260 million 5G connections or 20 percent of the global market.
The question everyone is asking in telco last couple of years is – do we really need 5G? Do we really need that throughput for our voice, video and messaging services? Can we significantly improve real-time communication services so that customers would be willing to pay for it? RCS aka Advanced Messaging is a great example of how difficult it can be to find the right business model for new technologies. EVS supported in 4G is more than what we need for voice calling. Although video calling is possible in 4G, not that many customers are using this option on their mobile devices. More popular than video is desktop-sharing, collaboration and communication in context. Well, lower latency and better throughput can be useful – but is it a reason strong enough to invest into the new 5G infrastructure, when collaboration applies mainly to fixed networks?
Still there are real-time communication applications which require low latency and huge amount of data so that 4G is not enough and more importantly which are really useful. When people hear about 3D video, holograms, Augmented Reality (AR) and Virtual Reality (VR), they mostly think about gaming. And yes, gaming can be a good example and people like to spend money for entertainment. But there are other examples, where AR and VR can make a difference.
The last week in Future Port Prague we could see many real applications. In the same way designers can discuss a new car, agent can show you a house for a rent. It’s completely a different experience to talk with your colleagues over Skype or to virtually sit with them in the same room. And imagine, what it would mean for a remote training! Or you can see a match of your favorite team real time and from any possible angle. Or .. well the possibilities are endless.
AR/VR becomes extremely useful when we need an expert input. Your engineer can analyze an issue with Wind Turbine tower using a drone. She can be sitting at her office and software can support her decision-making with related KPIs as wind speed, rotation, generated power etc. Remote surgery is also already more than just a theory.
Operators are taking AR/VR seriously and it is one of the main reasons they invest into the 5G infrastructure now. For Telenor 5G is a road to telesurgery. Verizon, AT&T, Singtel and T-Mobile are engaged in AR and VR through a number of trials and showcases. E.g. this year Verizon streamed live, 180-degree stereoscopic video from the Super Bowl field in Minneapolis directly to VR headsets in New York City, as well as offering a virtual in-stadium experience, including high-resolution replays on secondary screens, that demanded multiple 4K and HD video streams over 5G.
From telco point of view AR and VR is not just any media. SBC has to be able to identify AR/VR stream, work with new type of payload (multiple streams, new codecs) and make sure we’ll meet advanced metrics of Quality of Experience.
There are many new questions and requirements. We have to be able to synchronize audio and video with the user interaction and (e.g. in case of remote surgery) haptic feedback. Everyone knows that conflicts between the movement of the user and their senses, usually the visual and the vestibular senses (sensory conflict theory) can lead to nausea or motion sickness also known as virtual reality or cyber sickness. With direct haptic feedback, the QoE parameters become more critical, as latency, jitter or packet loss will cause instability in the closed haptic control loop.
Telco Standards and documents for AR, VR and telesurgery
There are not that many standards yet, many of them are still in progress. Some interesting documents are
3GPP TS 22.261 Service requirements for the 5G system
• Requirements on the data rate
• Requirements on motion‐to sound and motion‐to‐photon latencies
• Requirements on audio/video synchronization
3GPP TR 26.918 – Virtual Reality (VR) media services over 3GPP
- Media formats description
- Audio codecs and models
• Channel based
• Scene based
• Object based
- Video codecs
• Viewport dependent and independent coding
- Latency and synchronization aspects
• Motion to sound, motion to photon latency
• Audio-video synchronization
3GPP TR 26.929 – QoE parameters and metrics relevant to the Virtual Reality user experience
- QoE metrics
GSMA just recently held a round table in LA about CLOUD AR/VR, the key speakers were from Verizon, Mobile China, Deutche Telecom (T-Mobile) and MobiledgeX. The goal now is to identify business model and share best practices for the technical solution. To have a better understanding of what are the mobile operators up to, you can check a Huawei white paper. This can also give us some idea how much data is to be sent over 5G.
VR/AR can change our way of communication. We can work remotely, we can fly, while sitting at home, we can travel through our blood-stream or through history. To enable this powerful technology is not for free. We can reuse a great deal of existing network infrastructure. However to support mission-critical applications and mobile devices, we have to enhance both access networks and core network services.
Would you know about any (service) related standards, please let us know in comments.