Noise enhancement in robust estimation of location

dc.contributor.authorPan, Y.
dc.contributor.authorDuan, F.
dc.contributor.authorChapeau-Blondeau, F.
dc.contributor.authorAbbott, D.
dc.date.issued2018
dc.description.abstractIn this paper, we investigate the noise benefits to maximum likelihood type estimators (M-estimator) for the robust estimation of a location parameter. Two distinct noise benefits are shown to be accessible under these conditions. With symmetric heavy-tailed noise distributions, the asymptotic efficiency of the estimation can be enhanced by injecting extra noise into the M-estimators. With an asymmetric contaminated noise model having a convex cumulative distribution function, we demonstrate that addition of noise can reduce the maximum bias of the median estimator. These findings extend the analysis of stochastic resonance effects for noise-enhanced signal and information processing.
dc.description.statementofresponsibilityYan Pan, Fabing Duan, François Chapeau-Blondeau, Derek Abbott
dc.identifier.citationIEEE Transactions on Signal Processing, 2018; 66(8):1953-1966
dc.identifier.doi10.1109/TSP.2018.2802463
dc.identifier.issn1053-587X
dc.identifier.issn1941-0476
dc.identifier.orcidAbbott, D. [0000-0002-0945-2674]
dc.identifier.urihttp://hdl.handle.net/2440/114288
dc.language.isoen
dc.publisherIEEE
dc.rights© 2018, IEEE.
dc.source.urihttps://doi.org/10.1109/tsp.2018.2802463
dc.subjectMaximum likelihood estimation; robustness; reactive power; noise level; distribution functions; stochastic resonance
dc.titleNoise enhancement in robust estimation of location
dc.typeJournal article
pubs.publication-statusPublished

Files