Publish in OALib Journal

ISSN: 2333-9721

APC: Only $99


Any time

2019 ( 32 )

2018 ( 63 )

2017 ( 48 )

2016 ( 69 )

Custom range...

Search Results: 1 - 10 of 42188 matches for " Steven Van Vaerenbergh "
All listed articles are free for downloading (OA Articles)
Page 1 /42188
Display every page Item
Blind Identification of SIMO Wiener Systems based on Kernel Canonical Correlation Analysis
Steven Van Vaerenbergh,Javier Via,Ignacio Santamaria
Mathematics , 2013, DOI: 10.1109/TSP.2013.2248004
Abstract: We consider the problem of blind identification and equalization of single-input multiple-output (SIMO) nonlinear channels. Specifically, the nonlinear model consists of multiple single-channel Wiener systems that are excited by a common input signal. The proposed approach is based on a well-known blind identification technique for linear SIMO systems. By transforming the output signals into a reproducing kernel Hilbert space (RKHS), a linear identification problem is obtained, which we propose to solve through an iterative procedure that alternates between canonical correlation analysis (CCA) to estimate the linear parts, and kernel canonical correlation (KCCA) to estimate the memoryless nonlinearities. The proposed algorithm is able to operate on systems with as few as two output channels, on relatively small data sets and on colored signals. Simulations are included to demonstrate the effectiveness of the proposed technique.
Nonlinear System Identification using a New Sliding-Window Kernel RLS Algorithm
Steven Van Vaerenbergh,Javier Vía,Ignacio Santamaría
Journal of Communications , 2007, DOI: 10.4304/jcm.2.3.1-8
Abstract: In this paper we discuss in detail a recently proposed kernel-based version of the recursive least-squares (RLS) algorithm for fast adaptive nonlinear filtering. Unlike other previous approaches, the studied method combines a sliding-window approach (to fix the dimensions of the kernel matrix) with conventional ridge regression (to improve generalization). The resulting kernel RLS algorithm is applied to several nonlinear system identification problems. Experiments show that the proposed algorithm is able to operate in a time-varying environment and to adjust to abrupt changes in either the linear filter or the nonlinearity.
Adaptive Kernel Canonical Correlation Analysis Algorithms for Nonparametric Identification of Wiener and Hammerstein Systems
Steven Van Vaerenbergh,Javier Vía,Ignacio Santamaría
EURASIP Journal on Advances in Signal Processing , 2008, DOI: 10.1155/2008/875351
Abstract: This paper treats the identification of nonlinear systems that consist of a cascade of a linear channel and a nonlinearity, such as the well-known Wiener and Hammerstein systems. In particular, we follow a supervised identification approach that simultaneously identifies both parts of the nonlinear system. Given the correct restrictions on the identification problem, we show how kernel canonical correlation analysis (KCCA) emerges as the logical solution to this problem. We then extend the proposed identification algorithm to an adaptive version allowing to deal with time-varying systems. In order to avoid overfitting problems, we discuss and compare three possible regularization techniques for both the batch and the adaptive versions of the proposed algorithm. Simulations are included to demonstrate the effectiveness of the presented algorithm.
Overlapping Mixtures of Gaussian Processes for the Data Association Problem
Miguel Lázaro-Gredilla,Steven Van Vaerenbergh,Neil Lawrence
Computer Science , 2011,
Abstract: In this work we introduce a mixture of GPs to address the data association problem, i.e. to label a group of observations according to the sources that generated them. Unlike several previously proposed GP mixtures, the novel mixture has the distinct characteristic of using no gating function to determine the association of samples and mixture components. Instead, all the GPs in the mixture are global and samples are clustered following "trajectories" across input space. We use a non-standard variational Bayesian algorithm to efficiently recover sample labels and learn the hyperparameters. We show how multi-object tracking problems can be disambiguated and also explore the characteristics of the model in traditional regression settings.
Bayesian Extensions of Kernel Least Mean Squares
Il Memming Park,Sohan Seth,Steven Van Vaerenbergh
Computer Science , 2013,
Abstract: The kernel least mean squares (KLMS) algorithm is a computationally efficient nonlinear adaptive filtering method that "kernelizes" the celebrated (linear) least mean squares algorithm. We demonstrate that the least mean squares algorithm is closely related to the Kalman filtering, and thus, the KLMS can be interpreted as an approximate Bayesian filtering method. This allows us to systematically develop extensions of the KLMS by modifying the underlying state-space and observation models. The resulting extensions introduce many desirable properties such as "forgetting", and the ability to learn from discrete data, while retaining the computational simplicity and time complexity of the original algorithm.
A Probabilistic Least-Mean-Squares Filter
Jesus Fernandez-Bes,Víctor Elvira,Steven Van Vaerenbergh
Statistics , 2015,
Abstract: We introduce a probabilistic approach to the LMS filter. By means of an efficient approximation, this approach provides an adaptable step-size LMS algorithm together with a measure of uncertainty about the estimation. In addition, the proposed approximation preserves the linear complexity of the standard LMS. Numerical results show the improved performance of the algorithm with respect to standard LMS and state-of-the-art algorithms with similar complexity. The goal of this work, therefore, is to open the door to bring some more Bayesian machine learning techniques to adaptive filtering.
Gaussian Processes for Nonlinear Signal Processing
Fernando Pérez-Cruz,Steven Van Vaerenbergh,Juan José Murillo-Fuentes,Miguel Lázaro-Gredilla,Ignacio Santamaria
Computer Science , 2013, DOI: 10.1109/MSP.2013.2250352
Abstract: Gaussian processes (GPs) are versatile tools that have been successfully employed to solve nonlinear estimation problems in machine learning, but that are rarely used in signal processing. In this tutorial, we present GPs for regression as a natural nonlinear extension to optimal Wiener filtering. After establishing their basic formulation, we discuss several important aspects and extensions, including recursive and adaptive algorithms for dealing with non-stationarity, low-complexity solutions, non-Gaussian noise models and classification scenarios. Furthermore, we provide a selection of relevant applications to wireless digital communications.
Towards Trainable Media: Using Waves for Neural Network-Style Training
Michiel Hermans,Thomas Van Vaerenbergh
Computer Science , 2015,
Abstract: In this paper we study the concept of using the interaction between waves and a trainable medium in order to construct a matrix-vector multiplier. In particular we study such a device in the context of the backpropagation algorithm, which is commonly used for training neural networks. Here, the weights of the connections between neurons are trained by multiplying a `forward' signal with a backwards propagating `error' signal. We show that this concept can be extended to trainable media, where the gradient for the local wave number is given by multiplying signal waves and error waves. We provide a numerical example of such a system with waves traveling freely in a trainable medium, and we discuss a potential way to build such a device in an integrated photonics chip.
Nanofluids Thermal Conductivity Measurement in a Bénard Cell
Mohamed Mojahed,Stefan Van Vaerenbergh,Quentin Galand
Advances in Mechanical Engineering , 2013, DOI: 10.1155/2013/498124
Abstract: Thermal conductivity measurements of nanofluids were the subject of a considerable amount of published research works. Up to now, the experimental results reported in the current literature are still scarce and show many discrepancies. In this paper we propose measurements of this parameter using another experimental set-up. Because of very good thermal controls and big aspect ratio, the Bénard set-up is particularly well suited to determine the thermal conductivity. The aim of this paper is to detail the experimental measurement protocol. The investigated liquid is composed of single walled carbon nanotubes dispersed in water. The effect of liquid temperature on thermal conductivity was investigated. Obtained results confirm the potential of nanofluids in enhancing thermal conductivity and also show that the thermal conductivity temperature dependence is nonlinear, which is different from the results for metal/metal oxide nanofluids. 1. Introduction The nanofluids properties are far from being fully explored but one of them that has attracted much interest in the last decades is their potential to increase heat transfer. Many researchers have identified change in thermophysical properties of solutions when nanoparticles are dispersed [1] and the most important fluid property to be investigated for heat transfer is thermal conductivity. Discrepancy exists in nanofluid thermal conductivity data in the literature and enhancement mechanisms have not been fully understood yet. Many parameters modify the physicochemical properties of the nanofluid: nanoparticules concentration [2–4], nanoparticles dimensions [5], and thermal conduction of the basic fluid [6, 7], and probably other physicochemical parameters are to be considered. From the other side, different experimental techniques have been used to determine thermal conductivity of nanofluids, the transient hot wire method [8], the steady-state parallel-plate technique [2], the temperature oscillation technique [9], the optical beam deflection technique [10], and transient optical technique [11]. Unfortunately, the values of thermal conductivity obtained by those techniques on similar nanofluid do not appear to be consistent. So, it will be interesting to examine carefully the measurement technique and details when they are available in literature giving measurements. For nanofluids some factors can deeply modify the heat transfer and consequently the estimated thermal conductivity: convection, mass transfer, and agglomeration of nanoparticules. To provide experimental advance in the topic the Bénard cell
Sequence Diversity in the Dickeya fliC Gene: Phylogeny of the Dickeya Genus and TaqMan? PCR for 'D. solani', New Biovar 3 Variant on Potato in Europe
Johan Van Vaerenbergh, Steve Baeyen, Paul De Vos, Martine Maes
PLOS ONE , 2012, DOI: 10.1371/journal.pone.0035738
Abstract: Worldwide, Dickeya (formerly Erwinia chrysanthemi) is causing soft rot diseases on a large diversity of crops and ornamental plants. Strains affecting potato are mainly found in D. dadantii, D. dianthicola and D. zeae, which appear to have a marked geographical distribution. Furthermore, a few Dickeya isolates from potato are attributed to D. chrysanthemi and D. dieffenbachiae. In Europe, isolates of Erwinia chrysanthemi biovar 1 and biovar 7 from potato are now classified in D. dianthicola. However, in the past few years, a new Dickeya biovar 3 variant, tentatively named ‘Dickeya solani’, has emerged as a common major threat, in particular in seed potatoes. Sequences of a fliC gene fragment were used to generate a phylogeny of Dickeya reference strains from culture collections and with this reference backbone, to classify pectinolytic isolates, i.e. Dickeya spp. from potato and ornamental plants. The reference strains of the currently recognized Dickeya species and ‘D. solani’ were unambiguously delineated in the fliC phylogram. D. dadantii, D. dianthicola and ‘D. solani’ displayed unbranched clades, while D. chrysanthemi, D. zeae and D. dieffenbachiae branched into subclades and lineages. Moreover, Dickeya isolates from diagnostic samples, in particular biovar 3 isolates from greenhouse ornamentals, formed several new lineages. Most of these isolates were positioned between the clade of ‘D. solani’ and D. dadantii as transition variants. New lineages also appeared in D. dieffenbachiae and in D. zeae. The strains and isolates of D. dianthicola and ‘D. solani’ were differentiated by a fliC sequence useful for barcode identification. A fliC TaqMan?real-time PCR was developed for ‘D. solani’ and the assay was provisionally evaluated in direct analysis of diagnostic potato samples. This molecular tool can support the efforts to control this particular phytopathogen in seed potato certification.
Page 1 /42188
Display every page Item

Copyright © 2008-2017 Open Access Library. All rights reserved.