Nearly 10 years on from the world’s first publicly available LTE services, the impact 4G networks have had on mobile video consumption is clear. The ever-growing availability of online video content has changed the way media is distributed and consumed forever. Consumers have greater control and flexibility to watch what they want, when they want, on the device or platform of their choosing, in increasingly high definition. Research from Openwave Mobility showed that in the last five years, video consumption on mobile networks has grown on average 50-60 per cent year-on-year, a figure that is more than doubled when looking at developing markets in isolation.
In addition, at Mobile World Congress this year, we saw the future of video consumption on the go being realised, with manufacturers such as Huawei, Xiaomi and LG unveiling the first 5G phones to hit the market. Over the next few months, carriers like Verizon in the US and Vodafone in the UK will also start to make 5G services available in selective locations.
The 5G chicken and the data egg
At the dawn of next generation networks, one of the biggest challenges for online video content providers today, lies in managing the costs of Content Delivery Networks (CDNs). While the cost per byte has continued to decrease, the bandwidth required to keep consumers happy is increasing at an even faster rate. Online video demand is certain to continue, but is the development of 5G a response to this, or the means by which it will be accelerated? The significant improvements in transmission speed, quality and reliability will enable close to zero latency between a streaming source and the viewer, and could see newer technologies like virtual and augmented reality viewing experiences enter the mainstream.
In the short term though, early 5G deployments will represent “last mile” broadband data connectivity, with backbone networks growing capacity with the current trajectory driven by fibre backbones, overlaid by CDNs and boosted by such techniques as point of presence (POP) peering across global telecom providers.
Across 4G/LTE and 5G networks, digital video demand will more likely be driven by trends like cord cutting, virtual multichannel video programming distributors (MVPDs), 4K video, and global adoption of video-capable mobile devices on WiFi or mobile networks. Cisco’s study projects that globally, IP video traffic will be 82 per cent of all IP traffic by 2022, up from 75 per cent in 2017. Global IP video traffic will grow four-fold from 2017 to 2022. By that time, 5G is only expected to account for 3 per cent of global mobile devices and connection.
As user demand for data and bitrates keep going up without waiting for 5G, new approaches to video compression are an absolute necessity to keep the cost of streaming down, so content providers can spend their resources on creating more compelling content. Regardless of what network content is being delivered over, this can be achieved through three video workflow techniques: multi-codec streaming, per-file encoding and per-scene adaptation.
Multi-codec streaming
This bandwidth-saving technology enables a player to detect what the most efficient codec will be for streaming video on any browser or platform. AV1, H.264, HEVC, VP9, VVC have all been designed to suit a specific environment or type of content. By encoding videos into multiple codecs and configuring a player to make informed decisions about which files to stream and to whom, operators can deliver the highest picture quality on existing bandwidth. Looking further ahead, multi-codec solutions will assist early adopters as new codecs such as AV1 become ready for mass market penetration – which promises even larger bandwidth reductions.
Per-title encoding
This process involves adjusting an encoding configuration to optimise a specific video asset. It leverages the fact that some videos are far less complex than others, and therefore can be encoded at lower bitrates. Cartoons are a classic example of where this is particularly effective. Generally speaking, they contain scenes with large areas of solid colour at a lower complexity level, which can be compressed much more efficiently than the more detailed scenes in action or sci-fi movies.
Per-title encoding relies on performing a complexity analysis on each title before the encoding process begins. That complexity score is used to adjust the bitrate ladder by sending a new encoding profile to the encoder. The result: each video in a library is encoded in a way that best suits the content – minimising the bitrate and greatly reducing bandwidth usage.
Per-scene adaptation
This technique leverages the fact that the human eye is often unable to register a lot of the information delivered in a video stream. In most videos, there are many scenes that can be streamed at a lower bitrate without the viewer noticing. Per-scene adaptation involves supplementing the adaptation logic that governs the viewer’s behaviour with an additional stream of “quality metadata” that contains information about the visual complexity of a particular segment in the video.
In an adaptive streaming scenario with a standard configuration, the player will attempt to download a video file that fits the screen it is playing on. In a player that is configured for per-scene adaptation, the player can be alerted to the fact that an upcoming segment could be played at a lower bitrate without any noticeable loss of quality. In these cases, the player adjusts itself accordingly to reduce the bandwidth consumption – in some cases by as much as 30 per cent or more.
The quality metadata required to control this process is generated by running an analysis on each video as it is encoded by using a variety of perceptual quality measurements – meaning that the encoding process is optimised for the human eye. This metadata is included in the adaptive package and streams to the player in a similar way that subtitles or closed captions would.
Whenever 5G actually arrives, delivering world-class video content will be dependent on optimising video infrastructure. Multi-codec streaming, per-title encoding and per-scene adaptation are all effective methods of reducing CDN costs – by minimising file size and bandwidth usage – and can be implemented into any video workflow, providing consumers with a best-in-class viewing experience regardless of their network capacity.