Page 12 - Special Topic Session (STS) - Volume 4

STS556 Mohd Bakri A. et al.

Recovery signal from noise of higher volatility
2
2
Mohd Bakri Adam , Nurul Nisa’ Khairol Azmi , Norhaslinda Ali , Mohd Shafie
1,2
Mustafa 2
1 Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA Negeri Sembilan
2 Institute for Mathematical Research, Universiti Putra Malaysia

Abstract
Compound smoother is a non-linear smoothing technique that has the ability
to reduce heavy noise from signal and at the same time, is resistant to sudden
changes and impulse in a data series. In this study, the compound smoother
of 4253HT has been adjusted in the algorithm, specifically to estimate the
middle point of running median for even span size by applying the following
types of means; geometric, harmonic, quadratic and contraharmonic. The
performance of smoothers in extracting the signal from heavy noise were
assessed via simulation function of sinusoidal of high frequency plus trend
with higher percentage contaminated normal noise added. The result found
that modified 4253HT using geometric mean managed to smooth the data of
high frequency with fluctuation effectively compared to the existing and
others modification.

Keywords
compound smoother; running median; non-linear; 4253HT; signal; noise

1. Introduction
It is has been recognized that linear smoother is optimal to eliminate
Gaussian noise and track trends that are common in practice, Bernholt et al.
(2006). However, noise of high volatility tends to mask the general picture of
a data series. The existence of non well-behaved noise makes the assumptions
of linear model violated. Usually, least square estimation which is well known
for its poor performance in the presence of outliers or long-tailed distribution
data is used. According to Venetsanopoulos and Pitas (1990), linear smoothers
also have a high tendency to blur important features and lack of the ability to
remove impulsive noise. Not only that, linear smoothers are highly vulnerable
to outliers and could not deal well with nonlinearity in a data series. Blurry
edge which leads to the lost of important information is actually due to the
sudden changes in a series, Bernholt et al. (2006). Due to its ability to remove
non-Gaussian noise from a data series, median smoother is usually the favored
smoothing tools. Unfortunately, median smoother tends to over smoothed a
data series since it eliminates Gaussian noise too. One of established types of
median smoothers which have been widely employed in various area settings

1 | I S I W S C 2 0 1 9