Over-the-top needs QoS management

Governments around the world are looking at how they can encourage a digital economy. One obstacle is the performance of the Internet. Much like the road system, traffic expands to fill additional capacity, and congestion becomes the limit to traffic flows.

Governments have used various approaches to counter road traffic congestion. One is the toll road, where you pay extra to hopefully avoid congestion. Countries like Singapore use electronic road pricing in urban areas to control road use at peak hours through tolls.

So far, the Internet has the chaos of the open road, but to encourage the digital economy, there has to be a way to provide more control over the quality of service. For media providers, viewers expect the quality of over-the-air or cable if they are paying for content streamed over the Internet. The stuttering delivery of a remote newscast is not acceptable for general entertainment.

It sounds easy: Allow operators and viewers to pay extra to use the fast lane. However, many pressure groups don't like that. The alternative is to build more Internet capacity, but just like roads, traffic expands to meet the new capacity, and who pays for the development? Media content has increased data rates as television evolves from SD to HD. Will compression algorithms keep pace as television adds three dimensions and moves to UHDTV? As consumption via IP increases, it looks like that congestion will be around for a long time.

The proponents of “net neutrality” believe all data traffic should be treated as equal, but this view opposes those who believe in tiered services that can offer better control over quality of service. The Internet has developed from being a simple message carrier to being one of the primary media distribution platforms. The original design goals are stretched to breaking.

Over-the-air has always had QoS control through spectrum management. A broadcaster is given a frequency slot that it knows will be free from co-channel interference within its service area. The Internet has no such management; it's more of a free-for-all where anyone can use the pipe. It is not surprising that it becomes congested.

A 100Mb/s local end is not the whole solution. The capacity of the Internet backbone, the routing and the edge servers all have to be scaled accordingly to deliver a high quality of service, especially for streaming video.

This equality is all very well, but I believe that it leads to a lowest common denominator in the quality of service. There is little incentive to invest in infrastructure if providers cannot enhance revenues through premium offerings. I can pay extra for a high bandwidth local end, so why should a media distributor not be able to pay for priority access across the Internet?

Businesses like public transport are able to offer different classes of service, so why should the Internet be excluded? It is my view that many of those clamoring for net neutrality are those that want free content — the file sharers. It is a fact of life that nothing is free, and premium entertainment delivered over premium networks has to be paid for.

The development of the media business as it embraces delivery over IP, wireless or fiber, needs a fast lane to increase the choices offered to the consumer. Wireless IP access has particular issues over quality of service as the bandwidth available in metropolitan areas is going to be insufficient to meet demand. New consumer devices like the tablet are only exacerbating the problems.

As government regulators ponder on the wisdom of allowing prioritized traffic, the lobby groups on either side of the argument get more vociferous. Managing quality of service does not prevent access to users; it doesn't create a digital divide. But it does allow those who choose to obtain a premium service.

Send comments to: editor@broadcastengineeringworld.com