Towards real-time measurement of information in a scientific setting

Quantifying information has long been the domain of communications, data compression and encryption. However, developments over the last 50 years in the field of non-linear dynamics have shown information interchange to be intrinsic to physical dynamical systems and by implication, biological processes. The practical need to effectively characterise complex dynamical processes serves as motivation for the development of tools capable of measuring information and this paper describes an evolving toolset directed to quantifying information in time-series. Such data series may correspond to signals on a communications channel, or medical signals in diagnostic patient monitoring, neurological pulse trains, seismic data, or even industrial process measurements. Our tools provide a complementary approach to conventional time-series analysis and may be used to sensitively identify state, or changes-in-state in complex systems, without necesitating a detailed understanding of underlying process mechanisms. Our emphasis is on achieving real-time performance in measurement and visualisation, for diagnostic purposes.