This thesis focus on the theory analysis and algorithm aspects of signal subspace methods used for speech enhancement in digital speech processing The problem is approached by initially performing an analysis of subspace principles applied to speech signals in order to characterize the usefulness of de ning a signal subspace for this application The theory is formulated by means of the singular...

Many signal processing algorithms are based on the computation of the eigenstructure (eigenvalues and eigenvectors) of the covariance matrix of a data matrix. Applications include: direction of arrival estimation in array processing, spectral estimation and CDMA synchronization [1]. The advantages of using the eigenvector-based methods (also called subspace based methods) are well-known. In the...

DISCLAIMER This document describes work undertaken as part of a programme of study at the Faculty of Geo-Information Science and Earth Observation of the University of Twente. All views and opinions expressed therein remain the sole responsibility of the author, and do not necessarily represent those of the Faculty. i ABSTRACT This research work concentrates on understanding the concepts of hyp...

This paper addresses the problem of extracting a time-varying signal subspace from noisy signal measurements for direction-of-arrival (DOA) estimation and tracking. A robust adaptive method for extracting the signal subspace is developed based on robust statistics. The method is more resistant to the presence of outliers than existing adaptive methods. Robust DOA estimation and tracking are ach...

We describe ways to define and calculate L1-norm signal subspaces which are less sensitive to outlying data than L2-calculated subspaces. We focus on the computation of the L1 maximum-projection principal component of a data matrix containing N signal samples of dimension D and conclude that the general problem is formally NP-hard in asymptotically large N , D. We prove, however, that the case ...

A signal-subspace method is derived for the localization and imaging of unknown scatterers using intensity-only wave field data (lacking field phase information). The method is an extension of the time-reversal multiple-signal-classification imaging approach to intensity-only data. Of importance, the derived methodology works within exact scattering theory including multiple scattering.

Signal detection in certain noise environments fits naturally into a statistical hypothesis testing framework. In order to have moderately tractable models, the noise is often assumed to be additive with a multivariate normal distribution. Additionally, computational complexity requirements may demand the assumption of spatial stationarity, particularly in the case when the data is 2-dimensiona...

Abstract We describe ways to define and calculate L1-norm signal subspaces which are less sensitive to outlying data than L2-calculated subspaces. We start with the computation of the L1 maximum-projection principal component of a data matrix containing N signal samples of dimension D. We show that while the general problem is formally NP-hard in asymptotically large N , D, the case of engineer...

Abstract: Gender recognition and age detection are important problems in telephone speech processing to investigate the identity of an individual using voice characteristics. In this paper a new gender and age recognition system is introduced based on generative incoherent models learned using sparse non-negative matrix factorization and atom correction post-processing method. Similar to genera...

A mathematical framework for data representation and for noise reduction is presented in this paper. The basis of the approach lies in the use of wavelets derived from the general theory of frames to construct a subspace capable of representing the original signal excluding the noise. The representation subspace is shown to be efficient in signal modeling and noise reduction, but it may be acco...