Unicast vs Multicast

Internet enabled set top boxes and televisions are becoming the norm within UK markets. But what are the pros and cons of unicast and multicast solutions?

Unicast is the most common solution, but doesn't offer the scalability of multicast.

Unicast is the most common solution, but doesn't offer the scalability of multicast.

In the recent Arqiva white paper on ‘Connected Broadcasting’ we highlighted the fact that it is becoming the norm for set top boxes and TVs in the UK to be connected to the web. There is a growing cohort of boxes with a new, extra connection, on top of the normal broadcast receiver via an aerial or satellite. For the pay boxes, this is not remarkable, as the operators are attempting to generate extra revenues from on-demand box sets and movies.

What is remarkable is how ‘connected’ the free to air market is becoming. The arrival of YouView ushered in a generation of boxes that were connected as a default setting (YouView won’t work without the connectivity). However, prior to the arrival of YouView, web connectivity was slowly rolling out as a feature on all free boxes, with access to BBC iPlayer driving connection. In fact, virtually every new Freeview HD box sold since late 2011 is meant to allow quite an advanced level of connectivity (as defined by the DTG in their ‘D Book’ version 6.2.1 onwards). This newly defined standard allows the broadcast signal and the IP stream to connect, allowing interactive services of the kind offered by Arqiva.

Because of the presence of iPlayer, the discussion of these boxes often defaults to on-demand content. However, a more subtle shift is occurring in these boxes, with the arrival of IP streamed linear channels. It is now possible for a channel to secure a listing on the Freeview electronic programme guide (EPG) and deliver their linear signal over IP rather than broadcast DVB. This adds a new option for broadcasters and raises the potential of a range of new, smaller, specialist channels being added to the Freeview line up.

For those broadcasters considering IP distribution, there are at least two options – Unicast and Multicast IP channels. The first of these ‘unicast’ is the most common method for distributing web video streams. Unicast’s advantages are that it is relatively simple to set up and use, and is supported by a global industry of developers and suppliers.  It is the standard way to distribute a linear IP stream, and a TV channel can be run over any part of the internet without specific network configuration. It has limited disadvantages, but the major ones are the cost and complexity in dealing with large numbers of concurrent users. For the TV industry, which is used to broadcast scale viewers this can be an issue. 

A unicast TV channel creates a single, unique stream linking the server to an end viewers’ set top box. So 100k concurrent viewers of a live stream will require 100k unique, concurrent IP streams.  The key thing in terms of cost and scalability is that there are few economies of scale. Each viewer of a unicast channel takes up additional bandwidth. For example, if you have only one viewer playing a 100-kilobits per second (Kbps) stream, only 100 Kbps is being used. If you have 10 viewers all playing 100Kbps channel, then those viewers as a group are taking up 1,000 Kbps. The problem is that TV channels tend to like big audiences and a reasonably small linear channel can challenge the upper limits of unicast.

We expect the continued growth of connected Freeview, Freesat and YouView boxes to drive a related rise in the number of IP delivered linear channels...

The alternative, multicast, is much nearer to being a true broadcast platform. Multicast is relatively new, being pioneered in the UK by BT, who use it to distribute BT Sport in HD to YouView boxes connected via fibre. It is a bandwidth saving mechanism, but the downside is that it can only run over internet networks that have been configured to receive it; this is why BT YouView customers who live in areas not yet enabled for fibre can’t see the BT Sports channels on their box.

In multicast there is no direct link between the broadcaster’s servers and the viewers’ set top boxes. A broadcast channel using multicast relies on multicast-enabled routers to forward the signal as a complete package to the local routers right out at the edge of the internet; as near as possible to a viewer’s homes. The local routers effectively become a multicast ‘station’ - a little like a local radio station.

This means the broadcaster’s server sends out only a single stream to each multicast station. Therefore, the original server experiences the same demand whether only one set top box or 1,000 are viewing. It is more reliable for large audiences and high-definition content, but again there are costs and issues to be considered. The main issue is the restriction on where a channel can be distributed. In the short term, only BT has multicast-enabled networks.

There are also at least two other functional differences that may come into play; advertising and interactivity. Any channel considering addressable dynamic ad insertion into an IP channel may have to veer towards unicast. While there are now dynamic insertion systems for TV ads in multicast, there is no commercial solution for including addressability. This would require both decisioning and ad insertion to occur in a distributed way, out at the ‘multicast station’ routers in the network. At the moment, the solutions to achieve this are conceptual, not commercially ready.

Secondly, interactivity may work differently in unicast and multicast networks. Early innovators with IP channels are citing the ability to offer functionality like split screen, live data and red button interactivity, as some of the benefits of switching to IP. Once again, however, the mechanisms for doing this have been developed for unicast but not yet for multicast networks.

So What?

At Arqiva we expect the continued growth of connected Freeview, Freesat and YouView boxes to drive a related rise in the number of IP delivered linear channels. It is an exciting time for TV; an internet led revolution that is still broadcast-centric has come to fruition. However, there are still some significant questions to answer about network topology and functionality. Our work at Arqiva is focussing on both the technical and commercial implications of this new IP trend. We aim to help our clients understand the cost and functionality trade-offs they must face in this new age. We are interested in the markets response to these issues and would welcome comments about this blog.

1