SampEn {FWHVA} | R Documentation |
A SampEn algorithm used in this study to estimate the SE is available from Physione website (http://www.physionet.org/physiotools/sampen/c/). The algorithm of SampEn has three input parameters, (1) m: the embedded dimension, (2) r: the similarity criterion, (3) n: the length of a time series. In this study, two SE features ( : m=2, r=0.15, and : m=2, r=0.2) of each epoch of EEG signals are extracted.
This algorithm excutes the HVG with $O(n)$ and also include the weighted information.
The limitation, the input length of FWHVA is less than 4098, if the input length is large than it, the stack may be overflow.
SampEn(data, mvector, r_noise, std_flag)
data |
input data, one dimension |
mvector |
m: the embedded dimension |
r_noise |
r: the similarity criterion |
std_flag |
if std_flag=='v' stdand deveratin will be used |
Package: | FWHVA |
Type: | Package |
Version: | 1.0 |
Date: | 2013-07-28 |
License: | GPL (version 2 or later) |
sample Entropy[1..6] |
Guohun Zhu
Maintainer: Guohun Zhu <zhuguohun@163.com> or <Guohun.Zhu@usq.edu.au>
Zhu, Guohun, Li, Yan, & Wen, Peng(Paul). (2013). A weighted horizontal visibility graph algorithm and its application in Epileptic EEG signal classification. submitted
and
Physione website (http://www.physionet.org/physiotools/sampen/c/)
FWHVG
~~ <FWHVA>
~~
a=c(10,8,6,4,2,1,3,5,7,9,11) SampEn(a,3,0.2,'n') # no std