Optimizing Video in the Era of Streaming Wars
(A version of this article originally appeared on Pipeline)
The promise of increased bandwidth and low latency made possible by 5G networks is contributing to a massive increase in the development of new mobile video technologies and streaming services designed specifically for 5G networks. But even as excitement builds among mobile operators over the revenue potential from new 5G video services, they also have to account for the new challenges those services will create. And while video will become even more important in a 5G network, the types of video supported will change dramatically.
For instance, while mobile gaming currently has little impact on existing mobile networks, 5G networks will fully empower cloud gaming services such as Google Stadia, further driving a demand for high-definition video delivered without buffering or stalling. To support this, operator networks will need the ability to support three to four times more bandwidth compared to typical usage today. Crucially, this video will need to be delivered at latencies of a few milliseconds to enable an authentic gaming experience. All in all, mobile video will account for as much as 90 percent of 5G traffic, driven by gaming and increased video streaming as well as more esoteric uses of video such as augmented reality (AR), virtual reality (VR) and mixed reality. We will see the continued battle between Moore’s law (processor capacity doubling every 18 months) with devices getting smarter and consuming more HD, AR, VR videos and Shannon’s law, (bits per hertz over the air) which will allow more data speeds and capacity delivered over 5G networks.
History so far has shown that Moore’s law outpaces Shannon’s law, creating significant challenges for operators in terms of congestion in their networks. To add to this challenge, because mobile data traffic patterns are so irregular, capacity planning—even with cloudification of their networks—will remain a challenge, as total mobile data volumes being carried by these networks continue to double every six to 12 months. As an example, a HD video on average requires up to four times more bandwidth than standard video and encrypted over-the-top (OTT) traffic. An Augmented Reality video, which will become a reality with 5G rollout, has an even bigger appetite for network resources, consuming three to five times the volume of a similar-length HD video.
Another key quality metric is driven by the fact that consumers tend to base their concept of video quality on the length of time it takes for the video to start playing. The majority of consumers won’t wait longer than six seconds for video delivery to their mobile device, yet the average buffering time in a mobile network is seven seconds. As such, operators are having to walk a fine line between video picture quality (definition) and video delivery quality (latency).
The question for mobile operators, then, is how to ensure their 5G networks are optimized to meet the huge demands that video streaming applications will create.
Read the rest of the article on Pipeline