Nature-based algorithms applied to prediction, decision and communications

This was a coordinated project in which 10 spanish universities were involved. Its goal was to apply techniques such as neural networks, nonlinear dynamical models and genetic/evolutionary algorithms, including their combinations with advanced signal processing methods (based on higher-order statistics, for instance) to a set of selected problems in the areas of medical imaging, bussiness, energy and communications. The work of the GTAS group was mainly directed towards the application of neural networks to electric load prediction and fraud detection techniques in cellular networks; and the development of new nonlinear models applied to microwave amplifiers. Also, we started a research line on dynamic nonlinear models and chaos.