Gaussian mixture modeling for source localization
2011 (English)In: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2011, 2604-2607 p.Conference paper (Refereed)
Exploiting prior knowledge, we use Bayesian estimation to localize a source heard by a fixed sensor network. The method has two main aspects: Firstly, the probability density function (PDF) of a function of the source location is approximated by a Gaussian mixture model (GMM). This approximation can theoretically be made arbitrarily accurate, and allows a closed form minimum mean square error (MMSE) estimator for that function. Secondly, the source location is retrieved by minimizing the Euclidean distance between the function and its MMSE estimate using a gradient method. Our method avoids the issues of a numerical MMSE estimator but shows comparable accuracy.
Place, publisher, year, edition, pages
2011. 2604-2607 p.
, IEEE International Conference on Acoustics, Speech and Signal Processing. Proceedings, ISSN 1520-6149
Bayesian estimations, Closed form, Euclidean distance, Gaussian Mixture Model, Gaussian mixture modeling, Localization, Minimum mean-square error estimators, Prior knowledge, Probability density function (pdf), Source localization, Source location, Bayesian networks, Communication channels (information theory), Estimation, Gaussian distribution, Gradient methods, Image segmentation, Knowledge based systems, Numerical methods, Object recognition, Sensor networks, Sensors, Signal processing, Speech communication
IdentifiersURN: urn:nbn:se:kth:diva-46325DOI: 10.1109/ICASSP.2011.5947018ScopusID: 2-s2.0-80051607561ISBN: 978-1-4577-0538-0ISBN: 978-1-4577-0537-3OAI: oai:DiVA.org:kth-46325DiVA: diva2:453654
36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011; Prague; 22 May 2011 through 27 May 2011
QC 201111152011-11-032011-11-032012-01-11Bibliographically approved