TED (version 1.1.1)

cbfs: Generate an artificial event with white noise

Description

This function generates a box, cliff-ramp, ramp-cliff or a sine function with different levels of white noise as the background noise. Length of the generated event is 128. Generation of events are similar to that of Cylinder-Bell-Funnel dataset in the reference below (Keogh and Lin 2005).

Usage

cbfs(type = c("box", "rc", "cr", "sine"), A = 10, sigma = 1)

Arguments

type

type of the event to be generated. There are four options: `box', `rc',`cr',`sine' representing a box, cliff-ramp, ramp-cliff or a sine function.

A

amplitude of the event; default is 10.

sigma

a scalar specifying the level of white noise. Default is 1, which means the standard deviation of noise is 1.

Value

an artificial event with white noise.

References

Eamonn Keogh and Jessica Lin (2005). Clustering of time-series subsequences is meaningless: implications for previous and future research. Knowl. Inf. Syst., 8(2), 154-177. http://dblp.uni- trier.de/db/journals/kais/kais8.html#KeoghL05.

Yanfei Kang, Kate Smith-Miles, Danijel Belusic (2013). How to extract meaningful shapes from noisy time-series subsequences? 2013 IEEE Symposium on Computational Intelligence and Data Mining, Singapore, 65-72. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6597219&isnumber=6597208.

Yanfei Kang, Danijel Belusic, Kate Smith-Miles (2014). Detecting and Classifying Events in Noisy Time Series. J. Atmos. Sci., 71, 1090-1104. http://dx.doi.org/10.1175/JAS-D-13-0182.1.

Examples

# generate a box function with white noise
set.seed(123)
x1 = cbfs(type = 'box', sigma = 1)
# generate a box function with higher level noise
set.seed(123)
x2 = cbfs(type = 'box', sigma = 3)
# plot them
par(mfrow=c(1,2))
plot(x1,type='l',xlab='t',ylab=expression(x[1]))
plot(x2,type='l',xlab='t',ylab=expression(x[2]))