smooth

It is often necessary to smooth high frequency fluctuations out of data streams sequenced by time, position, etc. The 'smooth' utility applies such smoothing using the wavelet algorithm.

This implementation of wavelet smoothing was optimized for use in a data stream. It was adapted from HMMSeg Wavelet.Java by Thomas E. Wilson, University of Michigan. HMMSeg Wavelet.Java was written by Andrew Hemmaplardh, University of Washington.

http://noble.gs.washington.edu/proj/hmmseg/
Unsupervised segmentation of continuous genomic data, Bioinformatics 2007 23:1424-1426

See the above references for a more thorough description of the principles behind wavelet smoothing.