Title :
Nonstationary statistical tests in time-scale space
Author_Institution :
NASA Ames Res. Center, Moffett Field, CA, USA
Abstract :
Detection and estimation of abrupt changes in a nonstationary environment is an important and challenging problem. A statistical test for selecting the order of a nonstationary AR model is presented based on wavelet vanishing moment and predictive least squares principle. The order of the nonstationary AR is estimated at a different resolution level, which makes the order reliable. Its confidence interval is also determined. The test is derived based on the order probability distribution of the wavelet approximation sequence
Keywords :
autoregressive processes; fault diagnosis; least squares approximations; probability; signal processing; statistical analysis; wavelet transforms; abrupt changes; aerospace; confidence interval; machine monitoring; nonstationary AR model; nonstationary statistical tests; order probability distribution; predictive least squares; time-scale space; wavelet approximation sequence; wavelet vanishing moment; Band pass filters; Energy resolution; Fourier transforms; Low pass filters; Multiresolution analysis; Signal processing; Signal resolution; Testing; Time frequency analysis; Wavelet transforms;
Conference_Titel :
Aerospace Conference Proceedings, 2000 IEEE
Conference_Location :
Big Sky, MT
Print_ISBN :
0-7803-5846-5
DOI :
10.1109/AERO.2000.877913
