Multivariate Bayesian Statistics: Models for Source Separation and Signal Unmixing

Daniel B. Rowe

ISBN 10: 1584883189 / ISBN 13: 9781584883180

Published by Chapman and Hall/CRC, 2002

About this Item

Abnutzung / Risse - leicht. Of the two primary approaches to the classic source separation problem, only the Bayesian statistical approach avoids imposing unreasonable model and likelihood constraints. Bayesian methods leverage available information about model parameters, enabling estimation of sources and mixing coefficients while allowing for inferences. This comprehensive treatment of the source separation problem begins with an introduction using the "cocktail-party" analogy. Part I covers the necessary statistical background for the Bayesian source separation model. Part II focuses on instantaneous constant mixing models, where observed vectors and unobserved sources are independent over time but can be dependent within each vector. Part III explores more complex models, accommodating delayed sources, time-varying mixing coefficients, and temporal correlations between observation and source vectors. For each model, the author presents two distinct methods for parameter estimation. Real-world source separation challenges span various fields, including engineering, computer science, economics, and image processing, and often prove more complex than they seem. This book equips readers with essential statistical concepts and current research findings to effectively apply Bayesian methods in tackling the diverse "cocktail party" problems they may encounter.

Seller Inventory # 5b277667-46e3-4978-96b8-6ef36d608cc7

Bibliographic Details

• Title: Multivariate Bayesian Statistics: Models for Source Separation and Signal Unmixing
• Publisher: Chapman and Hall/CRC
• Publication Date: 2002
• Language: English
• ISBN 10: 1584883189
• ISBN 13: 9781584883180
• Binding: Hardcover
• Condition: Fine
• Seller catalogs:
  • Mathematics

About this title

Synopsis

Of the two primary approaches to the classic source separation problem, only one does not impose potentially unreasonable model and likelihood constraints: the Bayesian statistical approach. Bayesian methods incorporate the available information regarding the model parameters and not only allow estimation of the sources and mixing coefficients, but also allow inferences to be drawn from them. Multivariate Bayesian Statistics: Models for Source Separation and Signal Unmixing offers a thorough, self-contained treatment of the source separation problem. After an introduction to the problem using the "cocktail-party" analogy, Part I provides the statistical background needed for the Bayesian source separation model. Part II considers the instantaneous constant mixing models, where the observed vectors and unobserved sources are independent over time but allowed to be dependent within each vector. Part III details more general models in which sources can be delayed, mixing coefficients can change over time, and observation and source vectors can be correlated over time. For each model discussed, the author gives two distinct ways to estimate the parameters. Real-world source separation problems, encountered in disciplines from engineering and computer science to economics and image processing, are more difficult than they appear. This book furnishes the fundamental statistical material and up-to-date research results that enable readers to understand and apply Bayesian methods to help solve the many "cocktail party" problems they may confront in practice.

About the Author

Daniel B. Rowe holds a joint appointment as an assistant professor of Biophysics and Biostatistics at the Medical College of Wisconsin, Milwaukee, Wisconsin, USA.