DocumentCode :
573179
Title :
A cross-terms geometry based method for components instantaneous frequency estimation using the Cross Wigner-Ville distribution
Author :
Malnar, Damir ; Sucic, Victor ; Boashash, Boualem
Author_Institution :
Fac. of Eng., Univ. of Rijeka, Rijeka, Croatia
fYear :
2012
fDate :
2-5 July 2012
Firstpage :
1217
Lastpage :
1222
Abstract :
A novel method for the signal components instantaneous frequency (IF) estimation based on the CrossWigner-Ville distribution (XWVD) is presented. The cross-terms in the XWVD are deliberately formed between the analyzed signal and a reference signal. The proposed method yields a scaled and time shifted image that closely resembles the instantaneous frequency laws of the components present in the signal. As the interferences location follow geometrical rules, and by using a reference signal well localized in time and frequency, the time-frequency coordinates of the analyzed signal components IF can be calculated by an automatic procedure described below. The performance of the method is tested on both synthetic and real-life signals, showing improvements over another recently proposed components extraction method.
Keywords :
Wigner distribution; feature extraction; frequency estimation; geometry; time-frequency analysis; IF estimation; XWVD; analyzed signal; components extraction method; cross Wigner-Ville distribution; cross-terms geometry based method; reference signal; scaled image; signal component instantaneous frequency estimation; time shifted image; time-frequency coordinates; Atomic clocks; Estimation; Frequency estimation; Frequency modulation; Indexes; Signal to noise ratio; Time frequency analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Science, Signal Processing and their Applications (ISSPA), 2012 11th International Conference on
Conference_Location :
Montreal, QC
Print_ISBN :
978-1-4673-0381-1
Electronic_ISBN :
978-1-4673-0380-4
Type :
conf
DOI :
10.1109/ISSPA.2012.6310478
Filename :
6310478
Link To Document :
بازگشت