Performance Comparison of Known ICA Algorithms to a Wavelet-ICA Merger

These signals are however contaminated with artifacts which must be removed to have pure EEGsignals. These artifacts can be removed by Independent Component Analysis (ICA). In this paperwe studied the performance of three ICA algorithms (FastICA, JADE, and Radical) as well as ournewly developed ICA technique. Comparing these ICA algorithms, it is observed that our newtechniques perform as well as these algorithms at denoising EEG signals.