I am currently a Researcher at LMA working with the Sounds team, since October 2013. My current research field is the analysis of real environmental sounds with efficient statistical approaches that allow to synthesize realistic sounds. From October 2012 to September 2013, I have been interested in signal processing in cell biology with the Physics team in Fresnel institute. My project involves estimation methods for Super-resolution Imaging by Stochastic Optical Reconstruction Microscopy. From October 2010 to September 2012, I have been a postdoctoral researcher at IMS with the Signal team in Bordeaux, my research involved applying Markov Chain Monte Carlo (MCMC) methods and Sequential Monte Carlo for data estimation in multi-sensors based systems.

Prior to that, I worked at INRETS (in collaboration with LAGIS) on improving GNSS localisation precision in urban canyon using sequential estimation of signals and non-parametrical Bayesian analysis in a nonlinear dynamic model.

I received my Ph.D. in Electrical Engineering with emphasis in signal processing from The National Engineering School of Tunis in January 2008 (in collaboration with SyNeR team in LAGIS). My research has explored a variety of topics mainly in statistical audio signal processing. My research areas in the past included more specifically analysis and characterization of impulsive sounds, wavelets, signal processing for surveillance applications, pattern recognition, HMMs and SVMs.

• A. Rabaoui, H. Kadri, M. Davy, Nonparametric Bayesian Classification of Functional Data.

• Industrial Reports:
• A. Rabaoui, A. Giremus, A Markov Chain Monte Carlo Based Framework for Joint Estimation of Hidden States and Observation Errors Under equality Constraints, Project Supported by a grant from the TOTAL SA Group. (Draft)

• A. Rabaoui, H. Kadri, M. Davy, Nonparametric Bayesian Supervised Classification of Functional Data, ICASSP 2012. (Paper) (Slides)

• H. Kadri, A. Rabaoui, P. Preux, E. Duflos, A. Rakotomamonjy, Functional Regularized Least Squares Classification with Operator-Valued Kernels, The 28th International Conference on Machine Learning (ICML 2011). (Paper) (Slides)

• A. Rabaoui, H. Kadri, M. Davy, Bayesian nonparametric methods are naturally well suited to functional data analysis: The case of functional classification using DPMs. NIPS Workshops 2011. (Paper) (Poster)

• A. Rabaoui, N. Viandier, J. Marais, E. Duflos, On selecting the hyperparameters of the DPM models for the density estimation of observation errors, ICASSP 2011 (IEEE International Conference on Acoustics, Speech and Signal Processing). (Paper) (Poster)

• J. Marais, E. Duflos, N. Viandier, D. F. Nahimana, A. Rabaoui, Advanced signal processing techniques for multipath mitigation in land transportation environment, ITSC2010 (IEEE Conference on Intelligent Transportation Systems), pages 1480-1485, 19 - 22 Septembre, Madeira Island, Portugal. (Paper)

• N. Viandier, A. Rabaoui, J. Marais and E. Duflos, GNSS pseudorange error density tracking using Dirichlet Process Mixture, 13th International Conference on Information Fusion (FUSION 2010), 26-29 July Edinburgh, UK. (Paper)

• N. Viandier, A. Rabaoui, J. Marais, E. Duflos, Studies on DPM for the Density Estimation of Pseudorange Noises and Evaluations on Real Data, IEEE/ION PLANS 2010 (Position Location and Navigation Symposium), 3-6 Mai, Palm Springs, California. (Paper) (Slides)

• A. Rabaoui, N. Viandier, J. Marais, E. Duflos, Using Dirichlet Process Mixtures for the modelling of GNSS pseudorange errors in urban canyon, 22nd International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS 2009), 22-25 September, Savannah, Georgia. pp 2391-2399. (Paper)

• A. Rabaoui, N. Viandier, J. Marais and E. Duflos, On the Use of Dirichlet Process Mixtures for the Modelling of Pseudorange Errors in Multi-constellation Based Localisation, The 9th IEEE International Conference on ITS Telecommunication, 20-22 October 2009, Lille, France. (Paper) (Slides)

• N. Viandier, A. Rabaoui, J. Marais and E. Duflos, Enhancement of Galileo and multi-constellation accuracy by modeling pseudorange noises, The 9th IEEE International Conference on ITS Telecommunication, 20-22 October 2009, Lille, France. (Paper) (Slides)

• H. Kadri, M. Davy, A. Rabaoui, Z. Lachiri et N. Ellouze, Robust Audio Speaker segmentation Using One-Class SVMs, The 2008 European Signal Processing Conference (EUSIPCO'08), 25-29 Ao\FBt, Lausanne, Switzerland. (Paper)

• A. Rabaoui, H. Kadri, Z. Lachiri et N. Ellouze, New approaches based on one-class SVMs for impulsive sounds recognition tasks, IEEE International Workshop on Machine Learning for Signal Processing (MLSP'08), 16-19 October 2008, Cancun, Mexico. (Paper)

• A. Rabaoui, M. Davy, S. Rossignol, Z. Lachiri et N. Ellouze, Sélection de descripteurs audio pour la classification des sons environnementaux avec des SVMs mono-classe, XXIe Colloque GRETSI: Traitement du signal & des images (GRETSI'07). pp. 173-176, 11-14 Sep, Troyes, France. (Paper) (Slides)

• A. Rabaoui, Z. Lachiri et N. Ellouze, Towards an optimal feature set for robustness improvement of sounds classification in a HMM-based classifier adapted to real world background noise, IEEE Fourth International Multi-Conference on Systems, Signals & Devices (SSD'07), 19-22 Mars, Hammamet, Tunisie. (Paper)

• A. Rabaoui, Z. Lachiri et N. Ellouze, Classification robuste des sons environnementaux: Amélioration des paramètres et adaptation des modèles, Traitement & Analyse de l'Information : Méthodes & Applications (TAIMA'07), pp. 409-415, 21-26 Mai, Hammamet, Tunisie. (Paper) (Slides)

• A. Rabaoui, Z. Lachiri et N. Ellouze, Hidden Markov model environment adaptation for noisy sounds in a supervised recognition system, IEEE International Symposium on Communications, Control and Signal Processing (ISCCSP'06), 13-14 March, Marrakech, Marroc. (Paper) (Slides)

• A. Rabaoui, Z. Lachiri et N. Ellouze, Adaptation \E0 l'environnement par MAP et MLLR des modèles de Markov cachés pour la reconnaissance des sons de la vie courante, GEI'06, pp. 485-492, 23-26 Mars, Hammamet, Tunisie. (Paper) (Slides)

• A. Rabaoui, Z. Lachiri et N. Ellouze, Automatic environmental noise recognition, In Proceedings of the IEEE International Conference on Industrial Technology (ICIT'04), Vol. 3, pp. 1670-1675, 8-10 December, Hammamet, Tunisie. (Paper)

• Ph.D. Dissertation

• A. Rabaoui, Supervised classification of environmental sounds using statistical methods, Laboratory TIC, National Engineering School of Tunis (ENIT) , Tunisia, January. 2008. (voluminous document, pdf copy available upon request)