In this paper,shortcoming of traditional wavelet denoising in real-time signal processing is discussed,requirements of online denoising are considered,and a moving window is introduced into traditional wavelet transfo...In this paper,shortcoming of traditional wavelet denoising in real-time signal processing is discussed,requirements of online denoising are considered,and a moving window is introduced into traditional wavelet transform.Using the moving window,an online wavelet denoising approach is proposed.Some problems of online denoising,such as border distortion and pseudo-Gibbs phenomena,are discussed.To solve these problems,window extension and window cycle spinning are also proposed.Different approaches are tested by the signal widely used in denoising domain.Both the visual results and the quantitative measures are presented to highlight the availability of the new approach.展开更多
A novel two-step method is employed, for the first time, to fabricatc nonvolatile memory devices that have metal nanoerystals. First, size-averaged Au nanocrystals are synthesized chemically; second, they are assemble...A novel two-step method is employed, for the first time, to fabricatc nonvolatile memory devices that have metal nanoerystals. First, size-averaged Au nanocrystals are synthesized chemically; second, they are assembled into memory devices by a spin-coating technique at room temperature. This attractive approach makes it possible to tailor the diameter and control the density of nanocrystals individually. In addition, processes at room temperature prevent Au diffusion, which is a main concem for the application of metal nanocrystal-based memory. The experimental results, both the morphology characterization and the electrical measurements, reveal that there is an optimum density of nanocrystal monolayer to balance between long data retention and a large hysteresis memory window. At the same time, density-controllable devices could also feed the preferential emphasis on either memory window or retention time. All these facts confirm the advantages and novelty of our two-step method.展开更多
基金Supported by National Science Fund for Distinguished Young Scholars(60625302)National Key Fundamental Research Project of China(2002CB3122000)National High Technology Research and Development Program of China(863 Program)(20060104Z1081)
文摘In this paper,shortcoming of traditional wavelet denoising in real-time signal processing is discussed,requirements of online denoising are considered,and a moving window is introduced into traditional wavelet transform.Using the moving window,an online wavelet denoising approach is proposed.Some problems of online denoising,such as border distortion and pseudo-Gibbs phenomena,are discussed.To solve these problems,window extension and window cycle spinning are also proposed.Different approaches are tested by the signal widely used in denoising domain.Both the visual results and the quantitative measures are presented to highlight the availability of the new approach.
文摘A novel two-step method is employed, for the first time, to fabricatc nonvolatile memory devices that have metal nanoerystals. First, size-averaged Au nanocrystals are synthesized chemically; second, they are assembled into memory devices by a spin-coating technique at room temperature. This attractive approach makes it possible to tailor the diameter and control the density of nanocrystals individually. In addition, processes at room temperature prevent Au diffusion, which is a main concem for the application of metal nanocrystal-based memory. The experimental results, both the morphology characterization and the electrical measurements, reveal that there is an optimum density of nanocrystal monolayer to balance between long data retention and a large hysteresis memory window. At the same time, density-controllable devices could also feed the preferential emphasis on either memory window or retention time. All these facts confirm the advantages and novelty of our two-step method.
