Avoiding the inverse fractal problem for compressive sampling of 1/f data sets

We present a novel fractal Iterated Function System (IFS based) data interpolation algorithm enabling compressive sampling of 1/f data sets. We avoid the classical inverse IFS parameter estimation problem by using a novel analytical function driven variant of the random Iterated Function System (IFS) algorithm. We attempt to optimize the parameters of the analytical driver equation to optimize the data reconstruction by minimizing errors using various state-of-the-art genetic algorithms. We demonstrate our encouraging results and detail our methods and findings.