Showing papers on "Independent component analysis published in 1992"

Independent Component Analysis

Abstract: The Independent Component Analysis (ICA) of a random vector consists of searching for the linear transformation that minimizes the statistical dependence between its components. In order to design a practical optimization criterion, the expression of mutual information is being resorted to, as a function of cumulants. The concept of ICA may be seen as an extension of Principal Component Analysis, which only imposes independence up to second order and consequently defines directions that are orthogonal. Applications of ICA include data compression, detection and localization of sources, or blind identification and deconvolution.

Higher order statistics : proceedings of the International Signal Processing Workshop on Higher Order Statistics, Chamrousse, France, July 10-12, 1991

Abstract: Abbreviated. Basic aspects of higher-order spectra and some of their uses (D.R. Brillinger). Independent component analysis (P. Comon). Blind deconvolution using higher-order statistics (C.L. Nikias). Applications of higher order statistics in image processing (G. Jacovitti). Signal Modeling and Estimation. On the bispectrum of complex signals (I. Jouny). Complex random variables: A tensorial approach (J.L. Lacoume, M. Gaeta). On the degree of variety of the third-order statistics of stationary stochastic processes (F. Sakaguchi). New theoretical results on the bistatistics of 2-D signals (A.T. Erdem, A.M. Tekalp). On a method for noise generation with a specified set of higher order cumulants (A.G. Constantinides et al.). Inverse Problem and Identification. On identifiability of ARMA models of non-Gaussian processing via cumulant matching (J.K. Tugnait). Adaptive ARMA identification using cumulants (J.A. Rodriguez-Fonollosa, J. Vidal). A fast phase determination method by a single cumulant sample (C.-Y. Chi, J.-Y. Kung). Cross-bicepstrum and cross-tricepstrum approaches to multichannel deconvolution (D.H. Brooks, C.L. Nikias). Blind deconvolution method based on resonance model of wave propagation (R. Makowski). On a wide class of polyspectral inverse problems (A. Lannes et al.). Non-Stationary Signal Analysis. Detection of linear periodically time varying processes using higher order spectra (G.R. Wilson et al.). Applications Modified fourth-order moments in texture recognition (G. Ramponi, S. Carrato). Polyspectral analysis of non-stationary signals: System identification, classification and ambiguity functions (A.V. Dandawate, G.B. Giannakis). Application on higher-order spectra to high-resolution radar measurements (R.D. Pierce). Array Processing and Source Separation. Array processing from third order functions (M.A. Lagunas, G. Vazquez). Blind separation of sources: An algorithm for separation of convolutive mixtures (C. Jutten et al.). Nonlinear System Analysis Response computation for discrete-time nonlinear systems with random inputs (R. Ingenbleek, H. Schwarz). Second order volterra array processor mismatched to the fourth order moments of the jammers (P. Chevalier, B. Pincinbono). Polyspectra. Reconstruction of a sampled signal Fourier transform from its bispectrum (J. Le Roux). Analysis of nonlinear phenomena in space plasmas (N. Lounis et al.). Author index. (A complete list of contents is available on request from the Publisher.)