Time and frequency signals

Dhouha Kbaier Ben Ismail was a post-doctoral researcher at Ifremer from February 2nd, 2014 to July 31st, 2015 (18 months) where she worked with Ingrid Puillat and Pascal Lazure.

The position was offered by the REM/RDT unit in the framework of the in situ monitoring of the sea. The proposed position focused on mathematical methods dedicated to analyze environmental time series such as those acquired by long term monitoring systems, for instance deep sea observatories and coastal buoys, or those obtained from numerical models.

The objective was to review and study time series analysis methods for environmental science in several steps: (1) from a bibliographic review in other scientific domains such as econometrics, health, signal processing, and (2) then to test and to assess some selected methods by carrying out some case studies performed on acquired data. Several categories of methods are tested. Pitfall, advantages and robustness of the methods are assessed according to methods parameterization and data series types.

This led to publications in peer-reviewed journals, and presentations in conferences.A short overview of the outcomes is given hereafter. In addition 2 training workshops “Time series Analysis” were presented by Dhouha Kbaier Ben Ismail and co-organized by Ingrid Puillat in Ifremer Brest the 27-28 May 2015 and 3-4 June 2015. In this training, we focus on advanced spectral analysis adapted to be used for non-stationary time series with missing data and which can be recorded from nonlinear processes. The content of these trainings is gathered in a dedicated textbook, and can be downloaded here after.

 

 

Publications
  • KBAIER BEN ISMAIL Dhouha, Puillat Ingrid and Lazure Pascal, 2015. "Propriétés statistiques de la température, salinité et turbidité mesurées par la station MAREL Carnot dans les eaux côtières de Boulogne-sur-Mer (France)", F.G.Schmitt& A. Lefebvre (Eds.) Mesures haute résolution dans l'environnement marin côtier, Presses du CNRS (in press).
  • KBAIER BEN ISMAIL Dhouha, Lazure Pascal andPuillat Ingrid, 2015. "Advanced spectral analysis and cross correlation based on the empirical mode decomposition: application to the environmental time series", IEEE Geoscience and Remote Sensing Letters (GRSL), Accepted, in Press.
  • KBAIER BEN ISMAIL Dhouha, Puillat Ingrid and Lazure Pascal, 2015 "Statistical properties and time-frequency analysis of temperature, salinity and turbidity measured by the MAREL Carnot station in the coastal waters of Boulogne-sur-Mer (France)", submitted to Journal of Marine Systems (Special Edition for JERICO), Mai 2015.
Presentations and proceedings
  • KBAIER BEN ISMAIL Dhouha, Lazure Pascal andPuillatIngrid, 2014. "Analyse statistique et spectrale de la température, salinité et turbidité du système MAREL", Instrumentation haute fréquence pour l’observation et la surveillance de l’environnement marin, 12-13 Juin 2014, Boulogne-Sur-Mer, France.
  • KBAIER BEN ISMAIL Dhouha, Lazure Pascal andPuillatIngrid, 2015 "Time-frequency analysis of currents data sampled at four stations in the Réunion island: application of EMD and wavelets", International work-conference on Time Series (ITISE), 1–3 July 2015, Granada, Spain.
Recent given trainings and teaching
  • KBAIER BEN ISMAIL Dhouha, 2015. Training in time series analysis, session 1: 27-28 May 2015, Ifremer Brest, France, 10 attendees.
  • KBAIER BEN ISMAIL Dhouha, 2015. Training in time series analysis, session 2: 3-4 June 2015, Ifremer Brest, France, 14 attendees.
  • KBAIER BEN ISMAIL Dhouha and Ingrid Puillat, 2015. Statistical properties and time-frequency analyses of environmental time series, Part I: Stationarity analysis and spectral analysis, Tutorial in MTS/IEEE OCEANS'15, 19-22 October 2015, Washington DC.
  • KBAIER BEN ISMAIL Dhouha and Ingrid Puillat, 2015. Statistical properties and time-frequency analyses of environmental time series, Part II: Advanced time-frequency analysis of environmental time series: application of EMD and comparisons with wavelets, Tutorial in MTS/IEEE OCEANS'15, 19-22 October 2015, Washington DC.