A translation-invariant based adaptive threshold denoising method formechanical impact signal is proposed. Compared with traditional wavelet denoising methods, itsuppresses pseudo-Gibbs phenomena in the neighborhood o...A translation-invariant based adaptive threshold denoising method formechanical impact signal is proposed. Compared with traditional wavelet denoising methods, itsuppresses pseudo-Gibbs phenomena in the neighborhood of signal discontinuities. To remedy thedrawbacks of conventional threshold functions, a new improved threshold function is introduced. Itpossesses more advantages than others. Moreover, based on utilizing characteristics of signal, aadaptive threshold selection procedure for impact signal is proposed. It is data-driven andlevel-dependent, therefore, it is more rational than other threshold estimation methods. Theproposed method is compared to alternative existing methods, and its superiority is revealed bysimulation and real data examples.展开更多
The author generalizes the classical Lebesgue differential theorem to the case of rather general metrics. Then. the author applies it to the subelliptic metric to check the integral characterization of Holder continuo...The author generalizes the classical Lebesgue differential theorem to the case of rather general metrics. Then. the author applies it to the subelliptic metric to check the integral characterization of Holder continuous functions related to the metric.展开更多
文摘A translation-invariant based adaptive threshold denoising method formechanical impact signal is proposed. Compared with traditional wavelet denoising methods, itsuppresses pseudo-Gibbs phenomena in the neighborhood of signal discontinuities. To remedy thedrawbacks of conventional threshold functions, a new improved threshold function is introduced. Itpossesses more advantages than others. Moreover, based on utilizing characteristics of signal, aadaptive threshold selection procedure for impact signal is proposed. It is data-driven andlevel-dependent, therefore, it is more rational than other threshold estimation methods. Theproposed method is compared to alternative existing methods, and its superiority is revealed bysimulation and real data examples.
基金The project is supported by the Natural Science Foundation of China
文摘The author generalizes the classical Lebesgue differential theorem to the case of rather general metrics. Then. the author applies it to the subelliptic metric to check the integral characterization of Holder continuous functions related to the metric.