PhD Course @University of Genoa: "Data Fusion and Bayesian Interaction Modelling for Cognitive Ambient Intelligence" by Prof. Carlo Regazzoni


The course aims at providing PhD Candidates knowledge on basic tools in data fusion domain together with more advanced theories for representing, modelling and automatically interpreting interactions occurring between users, between users and artificial systems etc. within a smart cognitive environment. A Bayesian approach is used as a main methodological track in the course. In particular this module aims at :

 providing a common framework to identify and to describe methodologies and techniques for integrating multisensorial contextual data by using Data Fusion paradigms and techniques

 providing a common framework for defining behavioural artificial models for context based, adaptive and personalized decision steps used by cognitive system to address and react with respect to different contextual working situations.

 Showing examples and applications of specific techniques within cognitive telecommunication systems by means of description of two main case studies: cognitive radio and multisensor/multimodal cognitive human-machine interfaces in smart spaces.



 Data Fusion methodologies and techniques for integrating multisensorial contextual data Data Fusion models: the JDL model and its extensions: signals, objects, situations, threats, processes and cognitive refinement. Alignment, association, state extimations steps in data fusion levels. Alignment techniques: Space, Time, Frequency calibration techniques in video and radio based systems. Multisensor data association techniques: nearest neighbour, PDAF and JPDAF. State estimation techniques: from Kalman filter to non linear and non Gaussian state estimation techniques (extended Kalman Filter, Unscented Kalman Filter, Mean Shift, Particle Filters). Bayesian Networks for scene interpretation. Distributed Data Fusion (DDF): models and techniques. Distributed decision theory.

 Interaction Modeling. Bio-inspired behavioral cognitive artificial models for context based, adaptive and personalized decision. Neural basis of consciousness: the Damasio model (core-self, protoself, autobiographical memory and autobiographical self). The brain, memory and prediction. Adaptive and personalized embodied decision models for analyzing situations and driving actions and re-actions within cognitive systems. Decision space representation. Autobiographical memories and their representation and estimation through Bayesian learning techniques.

 Applications and case studies: Cognitive radio: Behavioral models for interactions between base stations and mobile terminals. Cognitive safety and physical security systems (smart patrolling in cooperative environments, preventive automotive vehicles, smart buidings, etc.)


 Course slides will be provided and made available at

Further reading:

 David L. Hall, James Llinas, “Handbook of Multisensor Data Fusion”, CRC Press, 2001;

 Y.Bar-Shalom, W.D.Blair, “Multitarget-Multisensor Tracking: Applications and Advances”, second edition, Artech House, 2000;

 Pramod Varshney, “Distributed Dtection and Data Fusion”, Springer, 1997

 Joseph Mitola III, “Aware, Adaptive and Cognitive Radio: The Engineering Foundations of Radio XML” Wiley-Interscience, 2006

 Anthonio Damasio, “The feeling of what happens: Body and Emotion in the Making of Consciousness” (1999)”, Harcourt Brace & Company,


Course schedule:

Starting date: Monday, June 17 2013. 

Daily lessons from 9.30 to 13.30 (plus 2 hours of lab in the noon, to be agreed with the teacher during the class).

End date: Friday, June 21 2013.

Mansarda Room -DITEN-  Ex-CNR building (4th floor) Via Opera Pia 11, 16145 Genoa.

For info please contact: