Signal-adaptive evolutionary spectral analysis using instantaneous frequency estimation

Akan A.

FREQUENZ, cilt.59, ss.201-205, 2005 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 59
  • Basım Tarihi: 2005
  • Doi Numarası: 10.1515/freq.2005.59.7-8.201
  • Dergi Adı: FREQUENZ
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.201-205
  • İstanbul Üniversitesi Adresli: Evet

Özet

In this paper we present a signal-adaptive evolutionary spectrum estimation method based on the discrete evolutionary transform (DET). The DET provides a representation for non-stationary signals and a time-frequency kernel that permits us to obtain the time-dependent spectrum of the signal. We show that the instantaneous phase and the corresponding instantaneous frequency (IF) can also be computed from the evolutionary kernel. Implementation of the IF estimation is done by means of a time-frequency masking. The proposed estimation is valid for mono-component as well as multi-component signals. Using the IF estimate of signal components in the representation, we obtain a signal-adaptive discrete evolutionary transform that results in spectral estimates with improved time-frequency resolutions.

In this paper we present a signal-adaptive evolutionary spectrum estimation method based on the discrete evolutionary transform (DET). The DET provides a representation for non-stationary signals and a time-frequency kernel that permits us to obtain the time-dependent spectrum of the signal. We show that the instantaneous phase and the corresponding instantaneous frequency (IF) can also be computed from the evolutionary kernel. Implementation of the IF estimation is done by means of a time-frequency masking. The proposed estimation is valid for mono-component as well as multi-component signals. Using the IF estimate of signal components in the representation, we obtain a signal adaptive discrete evolutionary transform that results in spectral estimates with improved time-frequency resolutions.