Jitter adds a small amount of gaussian (white) or uniform noise to a vector, matrix or N-D array.

This function is ideal for adding noise to a signal for processing, or generating starting conditions for chaotic functions.

The magnitude of the noise is given as either a fraction of the smallest difference between values of the input array, or as a fraction of the range of values of this input array.

See picture for an example.

This file is based upon the R function 'jitter' by written by Werner Stahel and Martin Maechler, ETH Zurich.
See http://stat.ethz.ch/R-manual/R-patched/library/base/html/jitter.html
for details of the original.

I've added support for imaginary inputs and gaussian noise, which weren't in the original.