CSIR researcher profile series

Dr Keith Ferguson,who heads up the real-time video coding research group at the Meraka Institute

Streamlining delivery of internet video broadcasting

World Telecommunication and Information Society Day has been celebrated worldwide on 17 May since 1969. It marks the anniversary of the signature of the first International Telegraph Convention and the creation of the International Telecommunication Union. An expert in a specialised field of information and communications technology, Dr Keith Ferguson of the Meraka Institute at the CSIR works on real-time video coding to deliver internet video broadcasting in a low bandwidth environment.

As new media for advertising or information delivery are identified, information and communications technology (ICT) is quick to follow with solutions to facilitate its use. Dr Keith Ferguson, a principal researcher at the Meraka Institute of the CSIR, has developed a novel method of solving the primary technology problem of seamless switching between independent encoded streams at low bit rates: 'channel hopping' without the dead zone or re-buffering.

Ferguson explains, “Real-time live video delivery services offer immense potential for social transformation in developing countries where resources and expertise are scarce and narrowly focused in urban areas. Currently, South Africa and other developing countries are characterised by limited internet access, with broadband available only at high cost to subscribed users. This technology is fully adaptable to these conditions and also has commercial applications for delivery of entertainment content and advertising to provide a competitive edge in the new media market segment.” He points out that convergence of national social development imperatives and the potential commercial value of the product makes South Africa an ideal context in which to develop this technology.

Researching ways to achieve a smooth stream
The aim of Ferguson's research is to deliver a smooth web cast in varying network conditions found in low bandwidth environments and other technical and infrastructural constrained networks.

Ferguson and senior developer Ralf Globisch aim to have a concept demonstrator of the real-time video coding product up and running within three years

As his research group at the Meraka Institute is small – it comprises himself and a senior developer, Ralf Globisch – he is working in collaboration with the University of Cape Town (UCT) and a Durban-based BEE internet service provider (ISP). “My own role is rather like that of a conductor of an orchestra,” he confesses. “Currently, I'm heading the research programme, participating in the commercialisation drive and overseeing the product development. It is a challenging task to run these three activities, traditionally treated as consecutive, in parallel.” Like any good conductor, he is aware of what is required from each group in the team, and harmonises them all into a whole during the process.

Apart from the project management, the research team at the Meraka Institute is responsible for the system design of the core media server software and part of the video research. All web-based market-facing interfaces are the responsibility of the ISP development team. Several postgraduate students at UCT are engaged in research on specialised areas of multimedia broadcasting.

Ferguson enjoys his multi-talented team, “A project such as this needs different people in different roles. It's very important that team members find a niche in which to perform to their highest potential and to the greatest advantage of the project. I enjoy creating opportunities for people and relish the outcome when they rise to the challenge.”

His goal is to complete this product and have an operational pilot site in about three years' time. “By then, we hope to have all four primary systems that interface with each other over internet protocol (IP) networks, notably, the transcoder, the media switching server, the web and advertising server and the client application,” he explains.

Ferguson is already looking beyond this product, “The video broadcasting product is the precursor to the peer-to-peer scenario, as we have it on Skype. As an IP-based communication system, Skype is affordable but its video conferencing features are at present limited in low bandwidth environments. Our current research is helping us discover better ways to do things in the future.”

Combining international experience with local know-how
Ferguson joined the Meraka Institute in 2007 on his return to South Africa after working in the United Kingdom (UK) for several years. “I'm a South African through and through. During the time I spent in the UK, I was always on my way home,” he quips. Born in Durban, he studied locally at the University of Natal (BSc, MSc) and the University of South Africa (Dipl Dat. M). His impressive academic achievements culminated in a Commonwealth Scholarship Award, which took him abroad to the UK, where he completed a PhD in electronic engineering. He subsequently formed a spin-off company, VICS Ltd, from research carried out at the University of Manchester Institute of Science and Technology to implement very fast, very low bit rate video coding algorithms for the mobile phone industry. “I've always enjoyed taking on a new challenge to push at the boundaries of what is achievable,” he notes.

When the mobile video and video advertising markets started to grow in step with the widespread introduction of 3G services, Ferguson was approached to sell his company. “Although I was urged to stay on in the company, my home has always been in South Africa.”

He returned to South Africa and started the search for an environment in which he could pursue his research and development interest. “Directed research is needed to address inherent problems in the industry,” he explains. “My particular interest lies in taking an idea through from the academic domain to the concept demonstrator phase and then right through to commercial grade products.” He admits that this path can be challenging, “It requires taking core ideas through the full value chain for them to become useful products, by which they are either desirable, in that they solve some specific problem, or are commercially viable. I believe I may have found a home at the CSIR where this kind of work is strongly encouraged.”

Ferguson uses an IP register to keep track of ideas

Keeping track of good ideas
Unusual ideas fascinate Ferguson. Not only does he come up with them frequently, he also keeps track of them in a handwritten IP register, “It's similar to a scrapbook, in which I note and date new ideas. This is not a formal process, but it's a good way of recording non-tangible ideas that have value.”

Ferguson uses his IP register as a 'treasure trove' from which to source patents, publications, research topics for theses and other outputs. His proven ability to generate ideas resulted in his previous registers being sold as part of the assets of his former UK company. The course of scientific and technology discovery is, however, not always smooth. He has had failures in the past, which have left him disappointed. “I then tackle things differently,” he reveals. “When the successes come, I really enjoy them. Celebrations become the order of the day!”

Ferguson is married to Professor Jill Bradbury who is at the School of Psychology of the University of KwaZulu-Natal (UKZN), Durban. He is passionate about the Drakensberg and having grown up at the coast, he loves the sea as well. For additional relaxation, he enjoys reading history.