Andriyani, Andriyani (2008) Estimator wavelet melalui pendekatan theresholding dalam regresi nonparametrik (kasus: data sinyal wicara). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 13052011001-33876.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
13052011001-33876.pdf Restricted to Repository staff only Download (7MB) |
Abstract
Analisis wavelet digunakan untuk mengestimasi fungsi prediktor yang terkontaminasi Gaussian noise dalam bentuk deret wavelet pada regresi nonparametrik. Estimator fungsi tersebut diperoleh dengan pendekatan *thresholding* yang dilakukan melalui transformasi wavelet diskrit, pemilihan metode threshold untuk mencari nilai threshold yang digunakan pada aturan thresholding, dan invers transformasi wavelet diskrit untuk membentuk estimator waveletnya. Adapun metode threshold yang digunakan adalah universal threshold dan minimum MSE threshold dengan penentuan level noise melalui Median Absolute Deviation Normal. Sedangkan aplikasi pendekatan tersebut dilakukan pada data sinyal wicara, untuk kemudian hasil estimasinya dibandingkan berdasarkan kriteria perhitungan MSE.
===================================================================================================================================
Wavelet analysis, in the form of a wavelet series, has been used to estimate a predictor function contaminated by Gaussian noise in nonparametric regression. This function estimator can be found by using a thresholding approach through the discrete wavelet transform, a threshold selection method, and the inverse discrete wavelet transform in order to form the wavelet estimator. Universal threshold and minimum MSE threshold, as threshold methods, with median absolute deviation normal are used for the thresholding approach. This approach will be applied to a speech signal, and MSE will be used to compare the results.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | 519.536 And e |
| Uncontrolled Keywords: | Regresi Nonparametrik, Wavelet, Gaussian Noise, Thresholding, MSE, Nonparametric regression, Wavelet, Gaussian nmse, Thresholding, MSE. |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | magang . |
| Date Deposited: | 08 Oct 2026 04:36 |
| Last Modified: | 08 Oct 2026 04:36 |
| URI: | http://repository.its.ac.id/id/eprint/145435 |
Actions (login required)
 |
View Item |