TED (version 1.0)

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]))