Title :
A Bayesian approach for frequency estimation using TV AR model for ocean acoustics Time-Series
Author :
Aunsri, Nattapol ; Hemrungrote, Sirichai
Author_Institution :
Sch. of Inf. Technol., Mae Fah Luang Univ., Chiang Rai, Thailand
Abstract :
In this paper, we investigate a sequential Bayesian filtering method to obtain estimates of instantaneous frequencies using Time-Varying Autoregressive (TVAR) model for ocean acoustics Time-Series. A particle filter along with a nonlinear model for estimating these frequencies is developed. Simulations demonstrate performance of the method compared to standard periodogram.
Keywords :
Bayes methods; acoustic signal processing; autoregressive processes; frequency estimation; nonlinear filters; particle filtering (numerical methods); time series; TVAR model; instantaneous frequency estimation; nonlinear model; ocean acoustic time series; particle filter; sequential Bayesian filtering method; time-varying autoregressive model; Acoustics; Bayes methods; Decision support systems; Filtering; Frequency estimation; Oceans; Time-frequency analysis;
Conference_Titel :
Asia-Pacific Signal and Information Processing Association, 2014 Annual Summit and Conference (APSIPA)
Conference_Location :
Siem Reap
DOI :
10.1109/APSIPA.2014.7041813
