The comparison of threshold techniques in one-cycle sliding window length for transient event / Saidatul Habsah Asman, Nofri Yenita Dahlan and Ahmad Farid Abidin

Asman, Saidatul Habsah and Dahlan, Nofri Yenita and Abidin, Ahmad Farid (2018) The comparison of threshold techniques in one-cycle sliding window length for transient event / Saidatul Habsah Asman, Nofri Yenita Dahlan and Ahmad Farid Abidin. Journal of Electrical and Electronic Systems Research (JEESR), 12: 18. pp. 123-128. ISSN 1985-5389

Abstract

Wavelet transform (WT) is an effective method to denoise the signal based on several attainment de-noising through
wavelet threshold methods. In this study, four types threshold
technique namely rigrsure, minimaxi, heursure, and sqtwolog
as proposed have been tested to de-noise the signal with
transient event. This study has been focusing on decomposition
coefficient at all levels signals for analysis. The mean square
error (MSE) and correlation coefficient (CC) are evaluated to
indicate the performance of proposed threshold. The analysis
signal is simulated using signal processing tool. From the
analysis, rigrsure is the best threshold that can be used for denoising signals with transient event because it is performed the
highest CC and lowest MSE for both one-cycle sliding window
and total 21 cycles reconstructed coefficient.

Metadata

Item Type: Article
Creators:
Creators
Email / ID Num.
Asman, Saidatul Habsah
UNSPECIFIED
Dahlan, Nofri Yenita
UNSPECIFIED
Abidin, Ahmad Farid
UNSPECIFIED
Subjects: Q Science > QA Mathematics > Wavelets (Mathematics)
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Electrical Engineering
Journal or Publication Title: Journal of Electrical and Electronic Systems Research (JEESR)
UiTM Journal Collections: UiTM Journal > Journal of Electrical and Electronic Systems Research (JEESR)
ISSN: 1985-5389
Volume: 12
Page Range: pp. 123-128
Keywords: Wavelet transform, threshold, de-noise, transient
Date: June 2018
URI: https://ir.uitm.edu.my/id/eprint/63055
Edit Item
Edit Item

Download

[thumbnail of 63055.pdf] Text
63055.pdf

Download (492kB)

ID Number

63055

Indexing

Statistic

Statistic details