Technical Program

Paper Detail

Session:Learning Theory and Models
Time:Thursday, May 20, 15:30 - 17:30
Presentation: Poster
Topic: Machine Learning for Signal Processing: Learning Theory and Modeling
Authors: Mo Chen; Imperial College, London 
 Danilo Mandic; Imperial College, London 
Abstract: Adaptive signal processing research has been directed towardsdesigning adaptive filters with high performance in terms of someperformance measure, however, little is known about how suchfilters influence the nature of the processed signal. Based uponsome recently introduced results on dealing with nonlinearitywithin a signal in hand (DVV method), we provide a criticalassessment of the qualitative performance of common linear andnonlinear filters and their combinations. Therefore, an insightinto the performance of the so called hybrid filters is provided,which is achieved for combinations of standard nonlinear (neural)and linear filters. It is shown that depending on the application,it is important not only to look for best filter performance interms of some quantitative measure of error but also for a filterthat will not change the character of a signal. Simulation resultssupport the analysis.

Home -||- Organizing Committee -||- Technical Committee -||- Technical Program -||- Plenaries
Paper Submission -||- Special Sessions -||- ITT -||- Paper Review -||- Exhibits -||- Tutorials
Information -||- Registration -||- Travel Insurance -||- Housing -||- Workshops

©2015 Conference Management Services, Inc. -||- email: -||- Last updated Wednesday, April 07, 2004