Wavelet transformation on a finite interval

Integral transformations on a finite interval with a singular basis wavelet are considered. Using a sequence of such transformations, the problem of nonparametric approximation of a function is solved. Traditionally, it is assumed that the validity condition must be met for a basic wavelet (the aver...

Full description

Saved in:
Bibliographic Details
Main Author: V. M. Romanchak
Format: Article
Language:Russian
Published: National Academy of Sciences of Belarus, the United Institute of Informatics Problems 2021-01-01
Series:Informatika
Subjects:
Online Access:https://inf.grid.by/jour/article/view/1060
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849240295681556480
author V. M. Romanchak
author_facet V. M. Romanchak
author_sort V. M. Romanchak
collection DOAJ
description Integral transformations on a finite interval with a singular basis wavelet are considered. Using a sequence of such transformations, the problem of nonparametric approximation of a function is solved. Traditionally, it is assumed that the validity condition must be met for a basic wavelet (the average value of the wavelet must be zero). The paper develops the previously proposed method of singular wavelets when the tolerance condition is not met. In this case Delta-shaped functions that participate in Parzen – Rosenblatt and Nadaray – Watson estimations can be used as a basic wavelet. The set of wavelet transformations for a function defined on a numeric axis, defined locally, and on a finite interval were previously investigated. However, the study of the convergence of the decomposition on a finite interval was carried out only in one particular case. It was due to technical difficulties when trying to solve this problem directly. In the paper the idea of evaluating the periodic continuation of a function defined initially on a finite interval is implemented. It allowed to formulate sufficient convergence conditions for the expansion of the function in a series. An example of approximation of a function defined on a finite interval using the sum of discrete wavelet transformations is given.
format Article
id doaj-art-48b20e0c8f9240ccb02af53710909892
institution Kabale University
issn 1816-0301
language Russian
publishDate 2021-01-01
publisher National Academy of Sciences of Belarus, the United Institute of Informatics Problems
record_format Article
series Informatika
spelling doaj-art-48b20e0c8f9240ccb02af537109098922025-08-20T04:00:40ZrusNational Academy of Sciences of Belarus, the United Institute of Informatics ProblemsInformatika1816-03012021-01-01174223510.37661/10.37661/1816-0301-2020-17-4-22-35942Wavelet transformation on a finite intervalV. M. Romanchak0Belarusian National Technical UniversityIntegral transformations on a finite interval with a singular basis wavelet are considered. Using a sequence of such transformations, the problem of nonparametric approximation of a function is solved. Traditionally, it is assumed that the validity condition must be met for a basic wavelet (the average value of the wavelet must be zero). The paper develops the previously proposed method of singular wavelets when the tolerance condition is not met. In this case Delta-shaped functions that participate in Parzen – Rosenblatt and Nadaray – Watson estimations can be used as a basic wavelet. The set of wavelet transformations for a function defined on a numeric axis, defined locally, and on a finite interval were previously investigated. However, the study of the convergence of the decomposition on a finite interval was carried out only in one particular case. It was due to technical difficulties when trying to solve this problem directly. In the paper the idea of evaluating the periodic continuation of a function defined initially on a finite interval is implemented. It allowed to formulate sufficient convergence conditions for the expansion of the function in a series. An example of approximation of a function defined on a finite interval using the sum of discrete wavelet transformations is given.https://inf.grid.by/jour/article/view/1060waveletwavelet transformthe parzen – rosenblatt window methodnonparametric estimatornadaraya − watson kernel regression
spellingShingle V. M. Romanchak
Wavelet transformation on a finite interval
Informatika
wavelet
wavelet transform
the parzen – rosenblatt window method
nonparametric estimator
nadaraya − watson kernel regression
title Wavelet transformation on a finite interval
title_full Wavelet transformation on a finite interval
title_fullStr Wavelet transformation on a finite interval
title_full_unstemmed Wavelet transformation on a finite interval
title_short Wavelet transformation on a finite interval
title_sort wavelet transformation on a finite interval
topic wavelet
wavelet transform
the parzen – rosenblatt window method
nonparametric estimator
nadaraya − watson kernel regression
url https://inf.grid.by/jour/article/view/1060
work_keys_str_mv AT vmromanchak wavelettransformationonafiniteinterval