ICASSP99 Source and Signal Separation

Source and Signal Separation
Home Full List of Titles 1: Speech Processing CELP Coding Large Vocabulary Recognition Speech Analysis and Enhancement Acoustic Modeling I ASR Systems and Applications Topics in Speech Coding Speech Analysis Low Bit Rate Speech Coding I Robust Speech Recognition in Noisy Environments Speaker Recognition Acoustic Modeling II Speech Production and Synthesis Feature Extraction Robust Speech Recognition and Adaptation Low Bit Rate Speech Coding II Speech Understanding Language Modeling I 2: Speech Processing, Audio and Electroacoustics, and Neural Networks Acoustic Modeling III Lexical Issues/Search Speech Understanding and Systems Speech Analysis and Quantization Utterance Verification/Acoustic Modeling Language Modeling II Adaptation /Normalization Speech Enhancement Topics in Speaker and Language Recognition Echo Cancellation and Noise Control Coding Auditory Modeling, Hearing Aids and Applications of Signal Processing to Audio and Acoustics Spatial Audio Music Applications Application - Pattern Recognition & Speech Processing Theory & Neural Architecture Signal Separation Application - Image & Nonlinear Signal Processing 3: Signal Processing Theory & Methods I Filter Design and Structures Detection Wavelets Adaptive Filtering: Applications and Implementation Nonlinear Signals and Systems Time/Frequency and Time/Scale Analysis Signal Modeling and Representation Filterbank and Wavelet Applications Source and Signal Separation Filterbanks Emerging Applications and Fast Algorithms Frequency and Phase Estimation Spectral Analysis and Higher Order Statistics Signal Reconstruction Adaptive Filter Analysis Transforms and Statistical Estimation Markov and Bayesian Estimation and Classification 4: Signal Processing Theory & Methods II, Design and Implementation of Signal Processing Systems, Special Sessions, and Industry Technology Tracks System Identification, Equalization, and Noise Suppression Parameter Estimation Adaptive Filters: Algorithms and Performance DSP Development Tools VLSI Building Blocks DSP Architectures DSP System Design Education Recent Advances in Sampling Theory and Applications Steganography: Information Embedding, Digital Watermarking, and Data Hiding Speech Under Stress Physics-Based Signal Processing DSP Chips, Architectures and Implementations DSP Tools and Rapid Prototyping Communication Technologies Image and Video Technologies Automotive Applications / Industrial Signal Processing Speech and Audio Technologies Defense and Security Applications Biomedical Applications Voice and Media Processing Adaptive Interference Cancellation 5: Communications, Sensor Array and Multichannel Source Coding and Compression Compression and Modulation Channel Estimation and Equalization Blind Multiuser Communications Signal Processing for Communications I CDMA and Space-Time Processing Time-Varying Channels and Self-Recovering Receivers Signal Processing for Communications II Blind CDMA and Multi-Channel Equalization Multicarrier Communications Detection, Classification, Localization, and Tracking Radar and Sonar Signal Processing Array Processing: Direction Finding Array Processing Applications I Blind Identification, Separation, and Equalization Antenna Arrays for Communications Array Processing Applications II 6: Multimedia Signal Processing, Image and Multidimensional Signal Processing, Digital Signal Processing Education Multimedia Analysis and Retrieval Audio and Video Processing for Multimedia Applications Advanced Techniques in Multimedia Video Compression and Processing Image Coding Transform Techniques Restoration and Estimation Image Analysis Object Identification and Tracking Motion Estimation Medical Imaging Image and Multidimensional Signal Processing Applications I Segmentation Image and Multidimensional Signal Processing Applications II Facial Recognition and Analysis Digital Signal Processing Education Author Index A B C D E F G H I J K L M N O P Q R S T U V W X Y Z	Separation Of A Class Of Convolutive Mixtures: A Contrast Function Approach Authors: Carine Simon, Laboratoire systèmes de communication - UMLV - Champs sur Marne - 5, bvd Descartes - 77454 Marne la Vallée Cedex - FRANCE (France) Philippe Loubaton, Laboratoire systèmes de communication - UMLV - Champs sur Marne - 5, bvd Descartes - 77454 Marne la Vallée Cedex - FRANCE (France) Christophe Vignat, Laboratoire systèmes de communication - UMLV - Champs sur Marne - 5, bvd Descartes - 77454 Marne la Vallée Cedex - FRANCE (France) Christian Jutten, LIS/TIRF - 44,avenue Félix Viallet - 38031 Grenoble Cedex - FRANCE (France) Guy d'Urso, EDF/DER - 6, quai Watier - 78401 Chatou Cedex - FRANCE (France) Page (NA) Paper number 1666 Abstract: In this paper, we address the problem of the separation of convolutive mixtures in the case where the non Gaussian source signals are not necessarily filtered versions of i.i.d. sequences. In this context, we show that the contrast functions, used in the linear process source case, still allow to separate the sources by a deflation approach. Some particular properties of higher order cumulants based contrast functions are also given. IC991666.PDF (From Author) IC991666.PDF (Rasterized) TOP A New Time-Domain Deconvolution Algorithm And Its Applications Authors: T. Engin Tuncer, Middle East Technical Unv., EE. Dept., Ankara, Turkey (U.K.) Page (NA) Paper number 1143 Abstract: Recently a new time-domain method has been presented for deconvolution [1]. This multidimensional method completely eliminates the problems of the previous methods in one dimension and covers a reasonable part of the solutions in multidimensions. In this paper, we present some of the properties of this method. We will especially focus on the frequency domain behaviour of the algorithm as well as the performance under numerical errors and errors due to noise. In addition we will present examples of the applications including deconvolution with a modified NAS-RIF algorithm. IC991143.PDF (From Author) IC991143.PDF (Rasterized) TOP Blind Signal Separation for Convolutive Mixing Environments Using Spatial-Temporal Processing Authors: James P Reilly, Communications Research Laboratory, McMaster University, 1280 Main St. W., Hamilton Ontario, CANADA L8S 4K1 (Canada) Lino E Coria Mendoza, Communications Research Laboratory, McMaster University, 1280 Main St. W, Hamilton, Ontario, CANADA L8S 4K1 (Canada) Page (NA) Paper number 2031 Abstract: In this paper we extend the infomax technique [1] for blind signal separation from the instantaneous mixing case to the convolutive mixing case. Separation in the convolutive case requires an unmixing system which uses present and past values of the observation vector, when the mixing system is causal. Thus, in developing an infomax process, both temporal and spatial dependence of the observations must be considered. We propose a stochastic gradient based structure which accomplishes this task. Performance of the proposed method is verified by subjective listening tests and quantitative measurements. IC992031.PDF (From Author) IC992031.PDF (Rasterized) TOP Dynamic Signal Mixtures and Blind Source Separation Authors: Dragan Obradovic, Page (NA) Paper number 2286 Abstract: Methods for blind source separation (BSS) from linear instantaneous signal mixtures have drawn a significant attention due to their ability to recover original independent non-Gaussian sources without analyzing their temporal statistics. Hence, original voices or images (modulo permutation and linear scaling) are extracted from their mixtures without modeling the dynamics of the signals. The typical methods for performing blind source separation are Linear Independent Component Analysis (ICA) and the InfoMax method. Linear ICA directly penalizes a suitably chosen measure of the statistical dependence between the extracted signals. These measures are either obtained from the Information theoretic postulates such as the mutual information or from the cumulant expansion of the associated probability density functions. The InfoMax method is based on the entropy maximization of the non-linear transformation of the separated signals. This paper analyzes extensions of the instantaneous blind source separation techniques to the case of linear dynamic signal mixtures. Furthermore, the paper introduces a novel method based on combining Time Delayed Decorrelation (TDD) with the minimization of the cumulant cost function. TDD is used to obtain an acceptable initial condition for the cumulant based cost function optimization in order to reduce the numerical complexity of the latter method. This combined approach is illustrated on two examples including a real life cocktail party example. Keywords: higher order statistics, signal reconstruction IC992286.PDF (From Author) IC992286.PDF (Rasterized) TOP On Underdetermined Source Separation Authors: Anisse Taleb, Christian Jutten, Page (NA) Paper number 1737 Abstract: This paper discusses some theoritical results on underdetermined source separation i.e. when the mixing matrix is degenerate, espcially when there is more sources than observations. In this case, we show that the sources can be restored up to an arbitrary additive random vector. In the particular case of discrete sources, very relevant for digital communications, we show that this vector is certain. IC991737.PDF (From Author) IC991737.PDF (Rasterized) TOP Single Channel Signal Separation using Linear Time Varying Filters: Separability of Non-Stationary Stochastic Signals Authors: James R Hopgood, Peter J.W Rayner, Page (NA) Paper number 1682 Abstract: Separability of signal mixtures given only one mixture observation is defined as the identification of the accuracy to which the signals can be separated. The paper shows that when signals are separated using the generalised Wiener filter, the degree of separability can be deduced from the filter structure. To identify this structure, the processes are represented on an arbitrary spectral domain, and a sufficient solution to the Wiener filter is obtained. The filter is composed of a term independent of the signal values, corresponding to regions in the spectral domain where the desired signal components are not distorted by interfering noise components, and a term dependent on the signal correlations, corresponding to the region where components overlap. An example of determining perfect separability of modulated random signals is given. IC991682.PDF (From Author) IC991682.PDF (Rasterized) TOP Blind Source Separation without Optimization Criteria? Authors: Vicente Zarzoso, Asoke K Nandi, Page (NA) Paper number 1144 Abstract: Blind source separation aims to extract a set of independent signals from a set of observed linear mixtures. After whitening the sensor output, the separation is achieved by estimating an orthogonal transformation, which in the real-mixture two-source two-sensor case is entirely characterized by a single rotation angle. This contribution studies an estimator of such an angle. Even though it is derived from geometric notions based on the scatter-plots of the signals involved, it is found, empirically, to exhibit a performance clearly up to the mark of other methods based on optimality criteria and, theoretically, to improve and generalize one of such procedures. The simplicity of the suggested estimator results in a straightforward adaptive version, which converges regardless of the source distribution, for quite mild conditions, and whose asymptotic analysis is easy to carry out. The applicability of the estimator in a full separation system is also illustrated. IC991144.PDF (From Author) IC991144.PDF (Rasterized) TOP Deconvolution of Ultrasonic Nondestructive Evaluation Signals Using Higher-Order Statistics Authors: Lahouari Ghouti, Chi Hau Chen, Page (NA) Paper number 5067 Abstract: In ultrasonic nondestructive evaluation (NDE) of materials, pulse echo measurements are masked by the characteristics of the measuring instruments, the propagation paths taken by the ultrasonic pulses, and are corrupted by addictive noise. Deconvolution operation seeks to undo these masking effects and extract the defect impulse response which is essential for identification. In this contribution, we show that the use of higher-order statistics (HOS)-based deconvolution methods is more suitable to unravel the aforementioned effects. Synthetic and real ultrasonic data obtained from artificial defects is used to show the improved performance of the proposed technique over conventional ones, based on second-order statistics (SOS), commonly used in ultrasonic NDE. IC995067.PDF (From Author) IC995067.PDF (Rasterized) TOP