
Novel Approach for Single Channel Blind Source Separation
Unsupervised Learning Algorithms and Applications
LAP Lambert Academic Publishing
Published on 2. October 2012
Book
Paperback/Softback
192 pages
978-3-659-26000-1 (ISBN)
Description
Single channel blind source separation (SCBSS) is an intensively researched field with numerous important applications. This book proposes a novel method based on variable regularised sparse nonnegative matrix factorization which decomposes an information-bearing matrix into two-dimensional convolution of factor matrices that represent the spectral basis and temporal code of the sources. To further improve the previous work, a new method is developed based on decomposing the mixture into a series of oscillatory components termed as intrinsic mode functions (IMF). It is shown that IMFs have several desirable properties unique to SCBSS and how these properties can be advantaged to relax the constraints posed by the problem. In addition, this book develops a novel method for feature extraction using psycho-acoustic model and a family of Itakura-Saito divergence based novel matrix factorization has been developed. The proposed matrix factorizations have the property of scale invariant which enables lower energy components to be treated with equal importance as the high energy ones. Results show that all the developed algorithms presented in this book outperformed conventional methods
More details
Language
English
Dimensions
Height: 220 mm
Width: 150 mm
Thickness: 13 mm
Weight
304 gr
ISBN-13
978-3-659-26000-1 (9783659260001)
Schweitzer Classification
Persons
Bin Gao obtained PhD degree (2007-2011) from Newcastle University, UK. Currently, He is a research associate at Newcastle University and his research interests include audio and image processing, machine learning, structured probabilistic modeling on audio applications such as audio source separation, feature extraction and denoising.