The repr_dwt computes DWT (Discrete Wavelet Transform) representation (coefficients) from a time series.
Usage
repr_dwt(x, level = 4, filter = "d4")
Arguments
x
the numeric vector (time series)
level
the level of DWT transformation (default is 4)
filter
the filter name (default is "d6").
Can be: "haar", "d4", "d6", ..., "d20", "la8", "la10", ..., "la20", "bl14", "bl18", "bl20",
"c6", "c12", ..., "c30". See more info at wt.filter.
Value
the numeric vector of DWT coefficients
Details
This function extracts DWT coefficients.
You can use various wavelet filters, see all of them here wt.filter.
The number of extracted coefficients depends on the level selected.
The final representation has length equal to floor(n / 2^level), where n is a length of original time series.
References
Laurinec P, Lucka M (2016)
Comparison of representations of time series for clustering smart meter data.
In: Lecture Notes in Engineering and Computer Science: Proceedings of The World Congress on Engineering and Computer Science 2016, pp 458-463