Kernel estimation for Time-frequency distribution

Deprem, Z.; Cetin, A.E.

doi:10.1109/SIU.2015.7130250

View / Download
353.8 Kb

Author

Deprem, Z.

Cetin, A.E.

Date

2015

Source Title

2015 23rd Signal Processing and Communications Applications Conference, SIU 2015 - Proceedings

Publisher

Institute of Electrical and Electronics Engineers Inc.

Pages

1973 - 1976

Language

Turkish

Abstract

In this article a method is introduced to obtain a high-resolution and cross term free Cohen's Class Time-frequency (TF) distribution based on kernel estimation. The kernel is estimated via projecting an initial rough TF distribution onto the epigraph set of l1 norm. The kernel is aligned with the signal alignment in ambiguity domain and filters out the cross terms. © 2015 IEEE.

Keywords

Cohen's class
Projection
Time-frequency
Frequency estimation
And filters
Cohen's class
Cross-terms
High resolution
Kernel estimation
Projection
Time frequency
Time-frequency distributions
Signal processing

Permalink

http://hdl.handle.net/11693/28230

Published Version (Please cite this version)

http://dx.doi.org/10.1109/SIU.2015.7130250

Collections

Department of Electrical and Electronics Engineering 3330