RESEARCH FILE: Genetic analysis aids faster internet

[ The University of Melbourne Voice Vol. 3, No. 1  14 April - 12 May 2008 ]

By Paul Richiardi

Genome analysis tools have solved a communication industry problem, leading to a patented technology that could speed performance in optical fibres carrying internet communication.

The outcome is a relatively low-cost device to identify and monitor common sources of optical noise that cause signal interference – a job which today would require multiple types of test equipment and major expenditure.

“It’s one of those discoveries that happens when you have interesting mixes of people – such as biological researchers sitting next to IT researchers,” says National ICT Australia Principal researcher Trevor Anderson.

The new technology is the work of Mr Anderson and Dr Adam Kowalczyk with colleagues from NICTA’s Victoria Research Laboratory at the University of Melbourne and the University’s Department of Electrical Engineering.

The idea of genome analysis tools helping to analyse noise in optical fibres came in a chance conversation between Mr Anderson and Dr Kowalczyk.

Dr Kowalczyk was working next door with a NICTA genetics team analysing genetic code to find patterns of gene sequences indicative of tumours.

Mr Anderson’s team had developed a way to present an optical signal as a two-dimensional image. “We thought it would allow us to recognise the ‘fingerprints’ of the various kinds of optical noise. But we didn’t know how to analyse the image,” he says.

“Adam looked at our problem and said: ‘This is easy – we have to identify cancer subtypes using a handful of noisy examples to learn from rather than the thousands available to you. Let’s try our algorithms on your data’. ”

They found the genetic algorithms could be adapted to identify the distinct visual patterns created by common forms of noise and distortion in optical fibres.

The resulting technology offers the potential to confidently increase speed ratings on long haul optical fibres from 10 gigabits per second to 40 gigabits per second or more without losing data due to line noise. It could set a new standard in high speed networks performance monitoring information.

Successful market testing of the technology has seen the team, with entrepreneur David Wright, form a company, Monitoring Division, raise significant venture capital, and report “very positive early customer engagement”.