A new method based on EMD and LZ complexity algorithms for discrimination of mental tasks

In this paper, we search for a suitable technique for discrimination of mental tasks and we succeeded to suggest the EMD-LZ method by using combination of two process: Empirical mode decomposition(EMD) and improved Lempel-Ziv(LZ) complexity measure. This technique was applied in EEG signals of 7 subjects performing 5 mental tasks. Each mode obtained from the EMD and each EEG channel are fed into improved LZ algorithm. Therefore a feature vector of 30 components is obtained for each trial. The Wilks' lambda parameter was applied for the selection of the most important variables and reducing the dimensionality of the feature vector. The classification of mental tasks was performed using Linear Discriminate Analysis (LDA). With this method, the average classification over all subjects in database was 92.46%. It was concluded that the EMD-LZ kernel allows getting better performances in the classification of mental tasks than the results obtained with other traditional methods, like spectral analysis.