TítuloDistributed Multivariate Regression with Unknown Noise Covariance in the presence of Outliers: A Minimum Description Length Approach
Tipo de publicaciónConference Paper
Year of Publication2016
AutoresLópez-Valcarce, R, Romero, D, Sala, J, Pagès-Zamora, A
Conference NameIEEE Workshop on Statistical Signal Processing (SSP)
Conference LocationPalma de Mallorca, Spain
Palabras clavecompass, wsn
Resumen We consider the problem of estimating the coefficients in a multivariable linear model by means of a wireless sensor networkwhich may be affected by anomalous measurements. The noise covariance matrices at the different sensors are assumedunknown. Treating outlying samples, and their support, as additional nuisance parameters, the Maximum Likelihoodestimate is investigated, with the number of outliers being estimated according to the Minimum Description Lengthprinciple. A distributed implementation based on iterative consensus techniques is then proposed, and it is shown effectivefor managing outliers in the data.