The Experts below are selected from a list of 309 Experts worldwide ranked by ideXlab platform
Marie-christine Roubaud - One of the best experts on this subject based on the ideXlab platform.
-
A Bayesian model for microarray datasets merging
2012Co-Authors: Marie-christine Roubaud, Bruno TorrésaniAbstract:The aggregation of microarray datasets originating from different studies is still a difficult open problem. Currently, best results are generally obtained by the so-called meta-analysis approach, which aggregates results from individual datasets, instead of analyzing aggre-gated datasets. In order to tackle such aggregation problems, it is necessary to correct for interstudy variability prior to aggregation. The goal of this paper is to present a new approach for microarray datasets merging, based upon explicit modeling of interstudy variability and gene variability. We develop and demonstrate a new algorithm for microarray datasets merging. The underlying model assumes normally distributed intrinsic gene expressions, distorted by a study-dependent nonlinear transformation, and study dependent (normally distributed) Observation Noise. The algorithm addresses both parameter estimation (the parameters being gene expression means and variances, Observation Noise variances and the nonlinear transformations) and data adjustment, and yields as a result adjusted datasets suitable for aggregation. The method is validated on two case studies. The first one concerns E. Coli expression data, artificially distorted by given nonlinear transformations and additive Observation Noise. The proposed method is able to correct for the distortion, and yields adjusted datasets from which the relevant biological effects can be recovered, as shown by a standard differential analysis. The second case study concerns the aggregation of two real prostate cancer datasets. After adjustment using the proposed algorithm, a differential analysis performed on adjusted datasets yields a larger number of differentially expressed genes (between control and tumor data). The proposed method has been implemented using the statistical software R 1 , and Bioconductor packages 2. The source code (valid for merging two datasets), as well as the datasets used for the validation, and some complementary results, are made available on the web site
-
Discrete-time piecewise linear filtering with small Observation Noise
IEEE Transactions on Automatic Control, 1995Co-Authors: P. Milheiro De Oliveira, Marie-christine RoubaudAbstract:A problem of discrete-time piecewise linear filtering with small Observation Noise is considered. The case in which the Observation function is not one-to-one and has two intervals of linearity is studied, in particular when this function is symmetric. Under some detectability assumption, a procedure is presented to detect the intervals of linearity of the Observation function. During such time intervals, one can then approximate the optimal filter by the corresponding Kalman-Bucy filter.
-
An approximate filter for a partially observed piecewise linear system with small Observation Noise
Systems & Control Letters, 1993Co-Authors: Marie-christine RoubaudAbstract:Abstract We propose a finite dimensional approximate filter for a piecewise linear system with small Observation Noise. The nonlinearity of our system depends only on state components that can be estimated quickly and accurately under a certain ‘detectability hypothesis’. Then we estimate the distance between our approximate filter and the state process to be filtered.
-
General piecewise linear filtering problems with small Observation Noise
Proceedings of the 28th IEEE Conference on Decision and Control, 1Co-Authors: Etienne Pardoux, Marie-christine RoubaudAbstract:An algorithm which is based on several Kalman filters running in parallel is presented. A test procedure for deciding which Kalman filter to follow, which produces a good estimate of the unobserved system process in a nonlinear filtering problem with piecewise linear dynamics and small Observation Noise is developed. The results generalize those of W.H. Fleming, D. Ji, and E. Pardoux (1988). >
-
Discrete time piecewise linear filtering with small Observation Noise
[1991] Proceedings of the 30th IEEE Conference on Decision and Control, 1Co-Authors: P. Milheiro De Oliveira, Marie-christine RoubaudAbstract:The authors describe a procedure that gives an approximate solution to a piecewise linear discrete time filtering problem with small Observation Noise. They present and compare different tests which make it possible to compute the intervals of linearity of the Observation function, under a 'detectability assumption' based on the drift. Over such an interval, one can then approximate the optimal filter by the corresponding Kalman-Bucy filter. >
Tomoaki Ohtsuki - One of the best experts on this subject based on the ideXlab platform.
-
Performance Analysis of Statistical STBC Cooperative Diversity Using Binary Sensors with Observation Noise
IEICE Transactions on Communications, 2006Co-Authors: Tomoaki OhtsukiAbstract:This letter analyzes the performance of statistical cooperative diversity based on space-time block codes (STBC) (Statistical STBC cooperative diversity) considering the effects of quantization and Observation Noise. Binary quantization is used. The bit error rate (BER) and average mutual information of the statistical STBC cooperative diversity with Alamouti's STBC and two active nodes are derived in the presence of general Observation Noise. It is shown that the performance of the statistical STBC cooperative diversity depends on the effects of Observation Noise and the number of cooperating nodes largely. It is also shown how much the communication between sensor nodes or feedback from the fusion center improves the performance of STBC cooperative diversity.
-
FUSION - Estimation Method of False Alarm Probability and Observation Noise Variance in Wireless Sensor Networks
2006 9th International Conference on Information Fusion, 2006Co-Authors: T. Fujita, Tomoaki Ohtsuki, T. KanekoAbstract:Likelihood ratio test (LRT) and best linear unbiased estimator (BLUE) have been researched as estimation methods of Observation event in sensor network. LRT and BLUE can estimate the Observation event with high accuracy at the fusion center (FC) when the FC has a perfect knowledge of Observation statistics of each sensor node. However, all the sensor nodes' Observation statistics are not always available at the FC. In this paper, we propose a method to estimate false alarm probability and Observation Noise variance of each sensor node at the FC, and present the performance of LRT and BLUE using the proposed estimation method. We show that the LRT and BLUE using the proposed method achieve almost the same bit error rate (BER) as the conventional LRT and BL UE with perfect knowledge of them
-
Performance analysis of statistical STBC cooperative diversity using binary sensors with Observation Noise
VTC-2005-Fall. 2005 IEEE 62nd Vehicular Technology Conference 2005., 1Co-Authors: Tomoaki OhtsukiAbstract:We analyze the performance of statistical co- operative diversity based on space-time block codes (STBC) (Statistical STBC cooperative diversity) with considering the effects of quantization, Observation Noise, fading, and channel Noise. We use binary quantization, that is, binary sensors. We derive the bit error rate (BER) and average mutual information of the statistical STBC cooperative di- versity with Alamouti's STBC and two active nodes in the presence of general Observation Noise. We show that the performance of the statistical STBC cooperative diversity depends on the effects of Observation Noise and the number of cooperating nodes largely. We also show how much the communication between sensor nodes or feedback from the fusion center improves the performance of STBC coopera- tive diversity.
Seiichi Nakamori - One of the best experts on this subject based on the ideXlab platform.
-
Filtering Technique Using Auto-Covariance Data of Signal and Observation Noise in Linear Discrete-Time Stochastic Systems
Programmable Device Circuits and Systems, 2016Co-Authors: Seiichi NakamoriAbstract:Usually, in the discrete-time estimation method of the signal, for the purpose of obtaining the state-space model for the signal, the signal process is fitted to the process model, i.e. the autoregressive model, the autoregressive moving average model, etc. Here, the signal is observed with additive white Noise. This paper proposes the filtering technique, which uses the finite number of auto-covariance data of the Observation process or the signal process for the positive lag time, the auto-covariance of the signal process, for the lag time 0, and the variance of the Observation Noise process in linear discrete-time wide-sense stationary stochastic systems. Especially, the auto-covariance data of the signal process is approximated by the Fourier cosine series expansion. The Fourier cosine series expansion leads to the expression in semi-degenerate kernel form of the auto-covariance function. Henceforth, in the current work, the filter using the covariance information calculates the filtering estimate recursively by using only the covariance information of the signal and Observation Noise processes.
-
New RLS Wiener Smoother for Colored Observation Noise in Linear Discrete-time Stochastic Systems
International Journal of Information Technology and Computer Science, 2013Co-Authors: Seiichi NakamoriAbstract:In the estimation problems, rather than the white Observation Noise, there are cases where the Observation Noise is modeled by the colored Noise process. In the Observation equation, the observed value ) (k y is given as a sum of the signal ) ( ) ( k Hx k z and the colored Observation Noise ) (k vc . In this paper, the Observation equation is converted to the new Observation equation for the white Observation Noise. In accordance with the Observation equation for the white Observation Noise, this paper proposes new RLS Wiener estimation algorithms for the fixed-point smoothing and filtering estimates in linear discrete-time wide-sense stationary stochastic systems. The RLS Wiener estimators require the following information: (a) the system matrix for the state vector ) (k x ; (b) the Observation matrix H ; (c) the variance of the state vector ) (k x ; (d) the system matrix for the colored Observation Noise ) (k vc ; (e) the variance of the colored Observation Noise.
-
RLS Wiener Smoother for Colored Observation Noise with Relation to Innovation Theory in Linear Discrete-Time Stochastic Systems
International Journal of Information Technology and Computer Science, 2013Co-Authors: Seiichi NakamoriAbstract:Almost estimators are designed for the white Observation Noise. In the estimation problems, rather than the white Observation Noise, there might be actual cases where the Observation Noise is colored. This paper, from the viewpoint of the innovation theory, based on the recursive least-squares (RLS) Wiener fixed-point smoother and filter for the colored Observation Noise, newly proposes the RLS Wiener fixed-interval smoothing algorithm in linear discrete- time wide-sense stationary stochastic systems. The Observation )
-
Design of RLS Wiener Smoother and Filter for Colored Observation Noise in Linear Discrete-Time Stochastic Systems
Journal of Signal and Information Processing, 2012Co-Authors: Seiichi NakamoriAbstract:Almost estimators are designed for the white Observation Noise. In the estimation problems, rather than the white Observation Noise, there might be actual cases where the Observation Noise is modeled by the colored Noise process. This paper examines to design a new estimation technique of recursive least-squares (RLS) Wiener fixed-point smoother and filter for colored Observation Noise in linear discrete-time wide-sense stationary stochastic systems. The Observation y(k) is given as the sum of the signal z(k)=Hx(k) and the colored Observation Noise vc(k). The RLS Wiener estimators explicitly require the following information: 1) the system matrix for the state vector x(k); 2) the Observation matrix H; 3) the variance of the state vector x(k); 4) the system matrix for the colored Observation Noise vc(k); 5) the variance of the colored Observation Noise; 6) the input Noise variance in the state equation for the colored Observation Noise.
-
Chandrasekhar-type recursive Wiener filter using covariance information in linear discrete-time wide-sense stationary stochastic systems
IFAC Proceedings Volumes, 2004Co-Authors: Seiichi Nakamori, Aurora Hermoso-carazo, J.d. Jiménez-lópez, J. Linares-pérezAbstract:Abstract This paper designs the Chandrasekhar-type recursive Wiener filter for the white Observation Noise in linear discrete-time wide-sense stationary stochastic systems. The system matrix in the state-space model of the signal, the crossvariance function of the state variable of the signal with the observed value, the Observation matrix for the signal, the variance of the white Observation Noise and the observed value are assumed to be known. In particular, this paper extends the Chandrasekhar-type recursive Wiener filter for a scalar Observation equation to the case of a vector Observation equation. The characteristic of the Chandrasekhar-type filter is to calculate the filter gain directly by solving the recursive difference equations.
P. Milheiro De Oliveira - One of the best experts on this subject based on the ideXlab platform.
-
Discrete-time piecewise linear filtering with small Observation Noise
IEEE Transactions on Automatic Control, 1995Co-Authors: P. Milheiro De Oliveira, Marie-christine RoubaudAbstract:A problem of discrete-time piecewise linear filtering with small Observation Noise is considered. The case in which the Observation function is not one-to-one and has two intervals of linearity is studied, in particular when this function is symmetric. Under some detectability assumption, a procedure is presented to detect the intervals of linearity of the Observation function. During such time intervals, one can then approximate the optimal filter by the corresponding Kalman-Bucy filter.
-
Discrete time piecewise linear filtering with small Observation Noise
[1991] Proceedings of the 30th IEEE Conference on Decision and Control, 1Co-Authors: P. Milheiro De Oliveira, Marie-christine RoubaudAbstract:The authors describe a procedure that gives an approximate solution to a piecewise linear discrete time filtering problem with small Observation Noise. They present and compare different tests which make it possible to compute the intervals of linearity of the Observation function, under a 'detectability assumption' based on the drift. Over such an interval, one can then approximate the optimal filter by the corresponding Kalman-Bucy filter. >
Rajeeva L. Karandikar - One of the best experts on this subject based on the ideXlab platform.
-
Path continuity of the nonlinear filter
Statistics & Probability Letters, 2001Co-Authors: Abhay G. Bhatt, Rajeeva L. KarandikarAbstract:We consider the nonlinear filtering model with signal and Observation Noise independent, and show that in case the signal is continuous in probability, the filter admits a version whose paths are continuous. The analysis is based on expressing the nonlinear filter as a Wiener functional via the Kallianpur-Striebel Bayes formula.
-
White Noise theory of robust nonlinear filtering with correlated state and Observation Noises
[1992] Proceedings of the 31st IEEE Conference on Decision and Control, 1Co-Authors: Arunabha Bagchi, Rajeeva L. KarandikarAbstract:In the direct white Noise theory of nonlinear filtering, the state process is still modeled as a Markov process satisfying an Ito stochastic differential equation, while a finitely additive white Noise is used to model the Observation Noise. In the present work, this asymmetry is removed by modeling the state process as the solution of a (stochastic) differential equation with a finitely additive white Noise as the input. This makes it possible to introduce correlation between the state and Observation Noise, and to obtain robust nonlinear filtering equations in the correlated Noise case. >
