TU Darmstadt / ULB / TUbiblio

Application of the Evidence Procedure to Linear Problems in Signal Processing

Shutin, D. and Koeppl, H. (2004):
Application of the Evidence Procedure to Linear Problems in Signal Processing.
AIP, In: AIP Conference Proceedings, 735, [Online-Edition: http://adsabs.harvard.edu/abs/2004AIPC..735..161S http://sci...],
[Conference or Workshop Item]

Abstract

The presented work addresses application of the evidence procedure to the field of signal processing where ill-posed estimation problems are frequently encountered. We base our analysis on the Relevance Vector Machines (RVM) technique originally proposed by M. Tipping. It effectively locally maximizes the evidence integral for linear kernel-based models. We extend the RVM technique by considering correlated additive Gaussian observation noise and complex-valued signals. We also show that grouping model parameters wvec , such that a single hyperparameter αk controls the kth cluster can be very effective in practice. In particular, it allows to cluster parameters wvec 's according to their potential relevance which in turns leads to highly improved generalization performance of the therewith parametrized models. The developed scheme is then illustratively applied to the problem of nonlinear system identification based on a discrete-time Volterra model. Similar ideas are used to analyze wireless channels from the channel measurement data. Results for synthetic as well as real-world data are presented.

Item Type: Conference or Workshop Item
Erschienen: 2004
Creators: Shutin, D. and Koeppl, H.
Title: Application of the Evidence Procedure to Linear Problems in Signal Processing
Language: German
Abstract:

The presented work addresses application of the evidence procedure to the field of signal processing where ill-posed estimation problems are frequently encountered. We base our analysis on the Relevance Vector Machines (RVM) technique originally proposed by M. Tipping. It effectively locally maximizes the evidence integral for linear kernel-based models. We extend the RVM technique by considering correlated additive Gaussian observation noise and complex-valued signals. We also show that grouping model parameters wvec , such that a single hyperparameter αk controls the kth cluster can be very effective in practice. In particular, it allows to cluster parameters wvec 's according to their potential relevance which in turns leads to highly improved generalization performance of the therewith parametrized models. The developed scheme is then illustratively applied to the problem of nonlinear system identification based on a discrete-time Volterra model. Similar ideas are used to analyze wireless channels from the channel measurement data. Results for synthetic as well as real-world data are presented.

Volume: 735
Publisher: AIP
Divisions: 18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications > Bioinspired Communication Systems
18 Department of Electrical Engineering and Information Technology
18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications
Event Title: AIP Conference Proceedings
Date Deposited: 04 Apr 2014 11:39
Official URL: http://adsabs.harvard.edu/abs/2004AIPC..735..161S http://sci...
Export:

Optionen (nur für Redakteure)

View Item View Item