When the network knows where you look
We are witnessing the advent of 5G. After having already seen early implementations happening in several countries such as the US, South Korea and Australia, 2019 is set to be a pivotal year for the evolution of cellular networks. According to the latest Ericsson Mobility Report, on a global level, major 5G network deployments are anticipated from 2020, bringing the total of 5G subscriptions for enhanced mobile broadband close to 1.5 billion by the end of 2024.
Unsurprisingly, enhanced mobile broadband will be one of the first use cases we will see happening for 5G once the first compatible smartphones become available during Q2 2019. More than ever before, mobile users want better connections and consume more data as they find new content and applications to interact with, as well as games to play. As a result, video traffic in mobile networks is forecast to grow by around 35 percent annually through 2024 to account for 74 percent of all mobile data traffic.
Therefore, it is a big challenge for mobile operators globally to deliver content in an efficient way and, even more, to continue doing so as mobile traffic continues growing unabated.
Video is king
Over the years, improvements in compression and encoding techniques have made video transmission more efficient, which has gone a long way to help deliver high-quality and high-definition (HD) content seamlessly. We know mobile users are not only consuming more video every day, they are also taking higher (and heavier) video quality as it becomes available.
It's become more important nowadays to use more efficient ways to deliver video when live streaming sports and events, higher-quality video (such as Facebook 360) and gaming applications. Live video also requires that content delivery be fast and transmitted with low latency.
4G/LTE has done a good job delivering mobile video. Emerging technologies such as LTE Broadcast will further enhance the end-user experience and network efficiencies. However, there's a limit to how much these techniques be applied without compromising video quality and end-user experience when lots of video is to be delivered to several users, particularly if content is customized and not the same for everyone (as can be the case with live sports, for example).
Eye-tracking technology meets VR and 5G
A very interesting concept we have been recently showing to our customers around the world is the result of a partnership between Ericsson, the Royal Institute of Technology KTH, Stockholm, Sweden, and Tobii, a Swedish high-technology company that develops products for eye control and eye-tracking.
In the demonstration, high-definition video is transmitted from the cloud to an end-user via a VR (Virtual Reality) headset and under two different scenarios:
- In the first scenario, no eye-tracking is being used; like this, video is streamed in real time from the cloud to the user's device. The challenge here is data consumption; if the user is on the move and consuming mobile data, having several users can create strain in the mobile network. Today, almost all VR applications have the users tethered to a fixed device and mobility is minimal, as Augmented Reality (AR) and Mixed Reality applications come to the fore, we can expect more users using data-hungry devices/applications and moving around
- In the second scenario, eye-tracking is activated. Tobii's technology can be currently embedded inside a VR headset which is handy as the device is not bulkier than they already are! Eye-tracking technology allows the headset (and the network) to know exactly what portions of the video the user is looking at. By knowing this, the network can optimise content delivery by just transmitting the bits the user is looking at in HD and leaving the remainder in Standard Definition (SD) which goes unnoticed to the end-user. A fast, reliable and low-latency 5G connection is needed to guarantee seamless, real-time video transmission and for the network to know (also in real-time) the areas in the video that need to be optimised. Bandwidth savings by doing this are around 90 percent which is astonishing
As we move into a 5G-enabled world, one of the big questions that are being asked all the time is about use cases. How will 5G benefit people, businesses and society as a whole?
It might very well be the case that we'll walk around with wearable headsets (way less bulky than today's VR headsets) where AR and VR applications will enhance our surroundings and help us go through our everyday activities. 5G and edge computing will be key in this scenario as the processing of information happens in the cloud and a fast, low-latency connection to our headsets allow them to be minimalistic.
While it might still be too early to say which application will be the most popular under the 5G paradigm, there's one thing we know for certain, the future is exciting.
For a deeper dive into the technology, please check out the Ericsson Research post on this topic from last year.