RG Smoothing Algorithm Which Makes Data Compression

Abstract

I describe a new method for smoothing a one-dimensional curve in Euclidian space with an arbitrary number of dimensions. The basic idea is borrowed from renormalization group theory which previously was applied to biological macromolecules. There are two crucial differences from other smoothing methods which make the algorithm unique: data compression and recursive implementation. One of the simplest forms of the method that is described in this article has only one free parameter - the number of iterative steps. This means that hardware implementation should be relatively easy because each loop is simple and strictly defined. The method could be beneficially applied to pattern recognition and data compression in future studies.