Signal Analysis and Prediction

This text is based on a selection of topics presented at the First European Conference on Signal Analysis and Prediction that took place in Prague, Czech Republic, 24-27 June 1997.

It consists of four parts: signal analysis; signal identification and modelling; signal prediction; and applications.

The main purpose of the first three parts of the text is to provide current information concerning: methods of time-frequency and time-scale signal analysis; higher order statistics in signal processing; selected methods of signal identification; the Bayesian approach to signal modelling and classification; nonlinear modelling by neural networks, fuzzy-ruled based systems; and methods of signal prediction and noise reduction.

The final section covers signal analysis and modelling for speech recognition in noisy environments, bio-medical signal processing, and environmental signal analysis and modelling.