ABSTRACT
At a time of increasing demand for video on mobile devices, it is vital to use the resources on existing mobile networks as efficiently as possible.
4G Broadcast (LTE eMBMS) offers the possibility of addressing issues of congestion and peak demand for popular content by sending a single stream once for reception by multiple users within a cell.
This capability could be enabled in localised areas of peak demand, within existing 4G networks, and switched on as required, to allow the network to be dynamically optimised for the current traffic conditions.
BBC Research & Development has been investigating how 4G Broadcast technology might be used to improve the delivery of streamed content to mobile devices.
We have demonstrated two example use cases – firstly as part of an app tailored to a specific event (for example at a sports venue), and secondly by connecting the technology seamlessly to BBC iPlayer (the BBC’s Internet streaming service that offers both live and catch-up content) to allow viewers to continue watching popular content in congested areas, without the experience being spoilt by buffering.
This paper will explain the work BBC Research & Development has been carrying out on 4G Broadcast and present the results of recent trial work.
INTRODUCTION
In recent years, the proliferation of highly capable smartphones means many more people are now interested in watching video on their mobile phones, and there is therefore a corresponding increase in the amount of content available for such devices, both as short-form clips as well as long-form programmes and live streams.
Although much of this viewing is done whilst the device is connected to a WiFi network (either at home or in the office etc.), there is also demand for this content directly over the mobile broadband networks – and the user is probably agnostic to the nature of the connection, being more concerned about both reliability and cost (or limits from data allowances).
Most industry analysts expect that the demand for mobile data will continue to increase significantly, and that a large portion of this will be driven by demand for video.
The concept of the “busy hour” is already well known within the industry when mobile networks are at their most congested at places of peak demand (e.g. busy railway stations during commuting hours), and increasing numbers of people requesting popular video content can only exacerbate this situation.
There is clear evidence that, for example during key sporting events such as the Wimbledon tennis championship, demand for live video content to mobile devices has increased dramatically; on 7th July 2013, when Andy Murray became the first British man to win the Wimbledon title since 1936, 64% of total requests to the BBC Sport site were from handheld devices.
4G Broadcast also has the potential to address spikes in demand over the mobile network when a new show or film is first released.
In all these cases, where large numbers of people are trying to watch the same content, at the same time, within the same geographic area, using a mobile broadband network, currently the network will attempt to stream numerous identical copies of the content, using a unicast mechanism.
4G Broadcast offers the opportunity to replace this with just one single stream for all users in one area, which not only has the potential to be significantly more efficient, but should also allow all users to be assured a consistent good quality experience, without the risk of buffering due to network congestion (in areas of good coverage).
This paper concentrates on the technical aspects of 4G Broadcast and does not address wider business or cost issues (for example the impact of 4G Broadcast on data allowances) or rights considerations.
WHAT IS 4G BROADCAST?
The term ‘4G Broadcast’ is used within this paper to refer to Long Term Evolution (LTE) enhanced Multimedia Broadcast/Multicast Service (eMBMS), the broadcast mode defined by the 3rd Generation Partnership Project (3GPP) in their 4G standards. Use of the term means eMBMS, as currently specified, namely within 3GPP Releases 9, 10 and 11.
Choosing this term was done for two reasons; firstly the aim was to use a name that would be accessible to a wider, non-technical audience; secondly other commonly used terms within industry such as ‘LTE Broadcast’ are not always used consistently and are sometimes used to refer to potential future developments of the specification.
There is typically benefit to be had from the broadcast mode when 2-3 users want to watch the same content concurrently, although the exact point at which it is more efficient to switch from unicast to broadcast will be dependent on the precise propagation conditions and how far the users are from a given cell tower.
No comments yet