Description of the program: nrlazy

This program performs simple nonlinear noise reduction. Each embedded point is replaced by the average vector calculated in its neighbourhood with a given size. This is different from what is described in Schreiber. There and in the program lazy only the central component of each vector is corrected. It is advisable to give both a try. We found a tendency that lazy performs better on map like data while nrlazy is superiour on flow like data.

Usage:

nrlazy [Options]

Everything not being a valid option will be interpreted as a potential datafile name. Given no datafile at all, means read stdin. Also - means stdin

Possible options are:

Option Description Default

-l# number of points to use whole file

-x# number of lines to be ignored 0

-c# column to be read 1

-m# embedding dimension 5

-d# delay for the embedding 1

-i# number of iterations 1

-r# neighborhood size (interval of the data)/1000

-v# neighborhood size in units of the std. dev. of the data
overwrites the -r option not set

-o# output file name; without file name:'datafile'.laz.n where n is the iteration
(or stdin.laz.n if data were read from stdin)
without -o the last iteration is also written to stdout

-V# verbosity level
  0: only panic messages
  1: add input/output messages
  2: write output of all iterations to files
  4: write the number of neighbors found for each point 3

-h show these options none

Option	Description	Default
-l#	number of points to use	whole file
-x#	number of lines to be ignored	0
-c#	column to be read	1
-m#	embedding dimension	5
-d#	delay for the embedding	1
-i#	number of iterations	1
-r#	neighborhood size	(interval of the data)/1000
-v#	neighborhood size in units of the std. dev. of the data overwrites the -r option	not set
-o#	output file name;	without file name:'datafile'.laz.n where n is the iteration (or stdin.laz.n if data were read from stdin) without -o the last iteration is also written to stdout
-V#	verbosity level 0: only panic messages 1: add input/output messages 2: write output of all iterations to files 4: write the number of neighbors found for each point	3
-h	show these options	none

Description of the Output:

Each of the files produced consists of one column being the filtered time series. If the verbosity level is set accordingly, the second column contains the number of neighbors found for this point. If this number is 1, no correction is done at all for this point.

View the C-sources.