Abstract: As a statistical model of wavelet coefficients, hidden Markov tree (HMT) can consider the statistical dependencies and non-Gaussian statistics of wavelet coefficients. Due to shift-variance of discrete wavelet transform (DWT), if DWT- based HMT model is used for machine condition diagnosis, it is likely to get incorrect results. To obtain shift-invariance, DT CWT-based (dual-tree complex wavelet transform) HMT model is developed. Experiments show that DT CWT-based HMT model can get much higher recognition rates in comparison with DWT-based HMT model.

Key words: Condition diagnosis, Dual-tree complex wavelet transform, Hidden Markov tree, Shift-invariance