Looking to the future of networks


30 Sep 2004

The stark concrete hulk of Adastral Park dominates the south Suffolk skyline. Were it not for the BT name stamped on a tall central tower, it could be mistaken for some form of nameless government-run institution.

It is a truly vast centre — it is scattered across 100 acres and employs 3,500 researchers and engineers. Its name — derived from the Latin ad astra, meaning to the stars — neatly sums up the pioneering nature of the work taking place there.

Originally an RAF research centre during the war, BT acquired it in the mid-Seventies, turning it into the centre of its enormous research programme with a sharp focus on blue sky or fundamental research.

While this type of research remains a key activity — BT undertakes long-term projects with Cambridge University, University College London and the Berkeley University of California among others — there has been a significant shift in BT’s research and development (R&D) model in recent years.

“We realised we couldn’t exploit our longer-term R&D efforts as our business cycle was too slow,” explains Mick Mehler, a spokesman for BT Exact, the telco’s research arm. “So we invited venture capital firms from the US to come and start up a half dozen ventures to try to exploit the technology and give us licensing fees in return. At the same time, we decided to work with partners such as Cisco because we realised that neither side could do it on its own — the equipment manufacturer needs to work with a carrier, and the carrier needs new products and investment in R&D,” he adds.

At the same time, deregulation in the telecoms market has led to a major shift in BT’s business strategy. As competition in the residential voice market continues to intensify, meeting the data networking needs of big business is becoming an increasingly important part of BT’s business both in Ireland and globally, according to Francois Barrault, who, as head of BT International, is responsible for the telco’s non-UK operations.

Barrault says that BT has successfully reinvented itself as a network-centric IT services provider to large organisations and offers a growing range of ICT services to its customers. These include network services (eg managed wide area network and local area network), IT services and applications (eg customer relationship management, managed desktop and portals) and business services (eg business process outsourcing, systems integration and strategy consulting).

This end-user focus has quickly filtered down to research. While many of the purely technical challenges have now been overcome, new challenges have risen to take their place. “What are the issues now? Making the whole lot work together, giving bandwidth to people that they require and cost effectively, and making it easier for the end user — this is where the focus for 2020 and beyond is,” says Mehler.

Broadband, mobility and improved customer experience are all pivotal research areas now at Adastral Park that were virtually unheard of a decade ago. So, too, is voice-over IP (VoIP), also known as internet protocol telephony (IPT), the much-hyped and long-anticipated convergence of voice and data that promises to transform business communications. BT’s strategy for IPT includes developing hosted services aimed at organisations that want the benefits of the technology without investing in an IP-based telephone switch.

At the CeBit exhibition in March, BT announced, along with its equipment partner Cisco, that it would be extending two hosted IPT services — Multimedia VoIP and VoIP Port — to a number of European countries this year, including Ireland. Already Bank of Ireland (and its UK subsidiary Bristol and West) has signed up to hosted VoIP from BT.

Fraser McNicol, product manager, Hosted IP, BT Global Services, concedes that VoIP has failed to live up to the hype but argues that this is about to change. “Traditionally conservative customers such as hospitals are starting to think about IPT and cost savings are a big driver as customers look to get more from less,” he says.

It is the breadth of applications IPT supports that is the real selling point, McNicol argues. “IPT does a lot more than just voice: it also incorporates unified messaging, desktop information services such as directory enquiries, as well as video telephony. It can also hook up specialist and non-specialist equipment such as analogue and IP phones.”

Over in the IP Network Centre, meanwhile, his colleague, Richard Shortland, is giving a vivid demonstration of what happens when a range of voice and data applications is run on a single IP-based virtual private network. As each application is added — a videoconference running on a large plasma screen, a video downloading on to a laptop, a telephone call being made and so on — the quality of the network starts to degrade.

Shortland then demonstrates how, using a technology called MPLS (multi-protocol label switching) a network manager is able to prioritise certain traffic, giving, for example, the highest preference to telephone calls and the lowest to video downloads or web surfing. His point: there used to be a quality issue with IPT but not any more.

As well as showcasing the latest advances in networking technology, Adastral Park also provides an insight into how various technologies may be applied in the future. One of these, BT Pulse, uses a simulated hospital environment to demonstrate how radio frequency ID technology connected to a centralised patient record database can improve healthcare by providing a shared information system linking patients, medical staff, hospitals and drugs companies.

Another group of projects, gathered under the Virtual Worlds heading, focuses on new forms of broadband applications and ways to access them. The latter include ambient interfaces such as digital plants with colour-coded stems that spring up when their owner has, say, a new email or a friend has suddenly logged on to their instant messaging system.

It might seem a bit bizarre today but no more so perhaps than the idea of spending eight hours a day in front of a beige box did 20 years ago.

Through such futuristic projects at Adastral Park, BT is determined to show that not only has it figured out what its customers need today but it also has a pretty good idea what they will need tomorrow as well.

By Brian Skelly

Intelligent products, such as a ‘digital plant’ whose stems light up to denote a message alert, may be the shape of things to come from research at Adastral Park.