Brussels, 19 Sep 2003
European scientists have found a reliable method for identifying computer users through the unique way they move their mouse. This new authentication system could help root out online fraud.
Each person has his or her own individual pen stroke that has for centuries been used to authenticate their identity. Now a UK scientist has updated this concept for the online age by developing software that can recognise people's unique mouse strokes as they sign on the dotted line.
It's about using the mouse as a kind of surrogate pen, London University's Peter McOwan, the programme's developer, was quoted as saying. The smart software, he explains, recognises the unique way in which someone writes their name or any other identifying symbol using a mouse. The British scientist believes that his invention could be used to make internet shopping and other online transactions safer against hackers and computer fraudsters, potentially removing one of the biggest obstacles to the growth of this promising sector.
Early trials on computer science students at Queen Mary College, London University, have indicated that the new system is 99% accurate.
Building a better mousetrap
The new system can be used to bolster the security of conventional authentication, such as password verification. "It's an extra biometric layer on top of traditional passwords," McOwan notes.
As opposed to smart cards and other high-tech authentication systems, the beauty of this system is that it requires no specialised equipment. The software works on the basis of a neural network. Neural networks are a form of aritifical intelligence that emulate the 'soft-wired' and flexible way in which our brains work in order to 'learn' how to do things or adapt to changing circumstances.
Recognising the huge potential of neural technology, the European Commission has funded several initiatives and projects in this field. These have included the Neuronet network, which aimed to consolidate European research efforts and increase industrial awareness of the latest research developments. Another EU-backed research project, ASPECT, investigated the use of advanced neural systems in user authentication systems for mobile phone services.