WAVELET-BASED IN-BAND DENOISING TECHNIQUE FOR CHAOTIC SEQUENCES
William L. B. Constantine, Per G. Reinhall · International Journal of Bifurcation and Chaos · 2001
A novel wavelet-based denoising technique (MODRA) is introduced and shown to be an effective means of denoising in-band contaminated chaotic sequences. MODRA is compared to existing WaveShrink methods and shown to be a more effective denoising technique based on a mean-squared-error statistic for simulated contaminations. For each case, an estimate of the correlation dimension D2 is calculated to quantify the results. As an integral part of MODRA, a best basis cost functional for the maximum overlap discrete wavelet packet transform using Shannon entropy is also introduced.