A Umist researcher claims to have overcome the cost/bandwidth conundrum.
Broadband faces a fundamental problem: the more successful it is, the more people will use it, and the slower it will become. This has obvious implications for the success of mobile video applications such as those currently being hyped by mobile operator 3.
For the past three years, Keith Ferguson has researching how to make video for low-bandwidth mobile phones acceptable. Now, with backing from BTI's Invest UK DTI/Foreign Office initiative and Umist Ventures, a company called Video & Image Coding Specialists (Vics) has been set up to develop his research into a commercial product.
Ferguson became interested in the concept of video for mobile phones around the time 3G was first being pushed as the next big thing for mobile networks. His main focus was the telecoms problems of the Third World. "While the telco infrastructure in the Third World is sophisticated, coverage and capacity on the networks was low," he explained.
Part of the problem was environmental, with cells for mobile phones having to cover large geographic regions. But there were also economical barriers to its adoption. "In Europe, bandwidth is expensive. Most of the Third World cannot not afford the 3G licences," said Ferguson.
The problem with bandwidth is that the more people that use it, the less bandwidth each user gets. One of the ways European operators have prevented their networks from grinding to halt with too many users is by charging a premium for faster networks.
But fast access is little use without applications, and the killer app touted by the mobile phone industry for 3G and GPRS is streaming video services, such as highlights of the football or news. The problem for the phone operators is the trade-off between picture quality and available bandwidth. At Umist, Ferguson tackled the problem of how to give users acceptable quality video on their handsets for the minimum bandwidth overhead.
"My research project was to look at how to create low bit rate video so that you can provide value for money on a mobile phone network," he said.
Coming from an electrical engineering background, Ferguson had experience of still image compression using programmable hardware designs. Current designs for video compression throw away video frames in order to keep the quality of the video at an acceptable level. While such techniques may go unnoticed when there is little change per frame, throwing away frames when a goal is about to be scored is not going to be acceptable to a user if essential frames relaying the path of the ball from the footballer to the back of the net are missing.
Ferguson's research took him down a different path to video nirvana. "I looked at how the brain perceives images. It is very easy to spot a white ball on a green background, but put the same ball among the crowd and it is impossible."
He began looking at visual patterns that could represent video. To represent every image perfectly would require an infinite number of visual patterns. But it is possible to fool the brain.
If you look at someone moving their hand across their face, for instance, you do not need to see each finger in detail. The same is true of a footballer chasing the ball across a pitch. It is not necessary to depict the footballer's legs moving as clearly as the ball. The brain will still perceive that his legs are moving.
So Ferguson researched patterns that could be used to approximate moving images, allowing him to build an algorithm which he claims is able to represent video using just a small number of visual patterns.
David Stead, managing director at Vics, believes the technology developed from Ferguson's Umist research will offer ordinary users a more acceptable form of video from their handsets than any rival video service. Stead was able to demonstrate acceptable mobile video streaming running over a standard GSM network, with bandwidth of just 6kbps.
The company's unique selling point is that it has circumvented the mobile operators and is speaking directly to content providers. The commercial product, called Mobile Portal, allows end-users who own a Sony Ericcson P800 or Nokia 7650/3650 to watch video streamed from the company's website. Vics has developed a downloadable video player that is currently available for Symbian- based smartphones.
Companies can charge for video clips via "credits" supplied as a premium-rate SMS message that consumers send to Vics.
Stead said, "We would like to see pricing in the region of 20p to 30p per video clip." A price point he said would be far cheaper than the types of video services being touted by some 3G operators. Priced in this way, Stead believes Vics could prove a winner among ringtone firms looking for the next big opportunity.
CV: Keith Ferguson
Keith Ferguson's upbringing in Africa developed his interest in delivering quality video at low prices to offer Third World countries the kind of access to technology enjoyed by First World countries. This became the subject of his PhD.
While working on his PhD, Ferguson met David Stead, who was looking for new technologies to develop commercially. During his time in the UK, Ferguson found that large numbers of people requiring access through a broadband delivery channel faced the same problems as those encountered in the Third World, where fewer people are trying to access technology, but through a much smaller narrowband channel.
The frustrating user experience in both contexts is the same. This is further exaggerated in the delivery of video to mobile phones. Ferguson's research and postdoctoral engineering efforts have culminated in producing a mobile video coding system that is now being brought to market.
What is BTI?
British Trade International (BTI) is the government organisation set up to support the UK's trade and investment strategy. It brings together the work of the Foreign and Commonwealth Office and the DTI on trade development and promotion of inward investment. BTI's two operating units are Trade Partners UK, which helps UK companies trading overseas, and Invest UK, which promotes the UK as an inward investment location.
Getting wired: tell us the future
Research work being undertaken at universities today will change the way we use IT, and Computer Weekly is on a mission to showcase the cutting-edge IT research being conducted in the UK.
Computer Weekly would like to hear from researchers who think they might have made a breakthrough. Each week we will feature innovation in the field of IT, giving a glimpse of how technology will evolve in the coming years.
This was first published in May 2003