Joint estimation decision methods for source localization and restoration in parametric convolution processes. Application to accidental pollutant release

In this paper, the problem of pollutant source localization and flow estimation is addressed. Potential applications of this work include leakage of hazardous chemicals or industrial effluents coming from an accidental situation. It is tackled in a one-dimensional context such as river, tunnel, canal with the aid of a single remote sensor. The pollutant is assumed to be coming from one out of N possible sources. Measurements are the result of a parametric convolution integral. The task may be viewed as a conditional deconvolution which requires a priori knowledge. In order to reduce the set of solutions, a source flow model is considered which introduces time bounds of the accidental spill. A joint estimation decision is derived in a Bayesian framework in both cases: with and without source assumptions. Without source model, the algorithm is unable to recover far sources location. On the contrary, the proposed source model enables to balance decision and take into account near and far sources as well. The benefit for this kind of solution is shown practically in terms of localization quality.

[1]  Arye Nehorai,et al.  A Sequential Detector for Biochemical Release in Realistic Environments , 2007, IEEE Transactions on Signal Processing.

[2]  Jerry M. Mendel,et al.  Maximum likelihood detection and estimation of Bernoulli - Gaussian processes , 1982, IEEE Trans. Inf. Theory.

[3]  Bobby R. Hunt,et al.  Improved Methods of Maximum a Posteriori Restoration , 1979, IEEE Transactions on Computers.

[4]  M. Benjelloun,et al.  Blind Localization by Subspace Method for a Scattering Model , 2000 .

[5]  A. Mohammad-Djafari,et al.  Maximum Entropy and Bayesian Methods , 1993 .

[6]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[7]  M. Benjelloun,et al.  Estimation of Pollutant Source Specifications Using a Conditional Deconvolution Method , 2000 .

[8]  G. Nofsinger,et al.  Distributed chemical plume process detection: MILCOM 2005 #1644 , 2005, MILCOM 2005 - 2005 IEEE Military Communications Conference.

[9]  R. Cristi,et al.  Markov and recursive least squares methods for the estimation of data with discontinuities , 1990, IEEE Trans. Acoust. Speech Signal Process..

[10]  Antonio Artés-Rodríguez,et al.  Sparse deconvolution using adaptive mixed-Gaussian models , 1996, Signal Process..

[11]  Sergio E. Serrano.  Hydrology for engineers, geologists, and environmental professionals , 2010 .

[12]  M. Lavielle.  2-D Bayesian deconvolution , 1991 .

[13]  Michèle Basseville,et al.  Detecting changes in signals and systems - A survey , 1988, Autom..

[14]  Charles A. Bouman,et al.  Inverse problems in atmospheric dispersion with randomly scattered sensors , 2006, Digit. Signal Process..

[15]  Arye Nehorai,et al.  Landmine detection and localization using chemical sensor array processing , 2000, IEEE Trans. Signal Process..

[16]  Donald B. Rubin,et al.  Max-imum Likelihood from Incomplete Data , 1972 .

[17]  D. Commenges.  The deconvolution problem: Fast algorithms including the preconditioned conjugate-gradient to compute a MAP estimator , 1984 .

[18]  Gif-sur-Yvette.,et al.  Scale Invariant Markov Models for Bayesian Inversion of Linear Inverse Problems , 2001, physics/0111124.

[19]  Marc Lavielle,et al.  Bayesian deconvolution of Bernoulli-Gaussian processes , 1993, Signal Process..

[20]  S. Twomey.  The application of numerical filtering to the solution of integral equations encountered in indirect sensing measurements , 1965 .

[21]  Guy Demoment,et al.  Image reconstruction and restoration: overview of common estimation structures and problems , 1989, IEEE Trans. Acoust. Speech Signal Process..

[22]  J. G. De Gooijer,et al.  Methods for Determining the Order of an Autoregressive-Moving Average Process: A , 1985 .

[23]  Michèle Basseville,et al.  Detection of Changes in Signals and Systems , 1987 .

[24]  P. Kathirgamanathan,et al.  Source Term Estimation of Pollution from an Instantaneous Point Source , 2002 .

[25]  Alexander Y. Khapalov,et al.  Localization of unknown sources for parabolic systems on the basis of available observations , 1994 .

[26]  Arye Nehorai,et al.  Biochemical Transport Modeling and Bayesian Source Estimation in Realistic Environments , 2007, IEEE Transactions on Signal Processing.

[27]  Michel Barlaud,et al.  Deterministic edge-preserving regularization in computed imaging , 1997, IEEE Trans. Image Process..

[28]  Kjetil F. Kaaresen,et al.  Deconvolution of sparse spike trains by iterated window maximization , 1997, IEEE Trans. Signal Process..

[29]  Arogyaswami Paulraj,et al.  A subspace approach to blind space-time signal processing for wireless communication systems , 1997, IEEE Trans. Signal Process..