Descendent adaptive noise cancellers to improve SNR of contaminated EEG with gradient-based and evolutionary approach

In this paper, an Adaptive Noise Canceller (ANC) technique for different artefacts cancellations from the Electroencephalogram (EEG) signals is presented. The proposed technique is based on gradient based adaptive algorithms such as Least Mean Square (LMS), Normalised Least Mean Square (N-LMS) and Recursive Least Square (RLS) algorithms and an evolutionary algorithm like particle swarm optimisation (PSO) technique. Descendent structure is made through three adaptive noise cancellers for the removal of line noise, ECG and EOG artefacts. When compared, the adaptive noise canceller technique based on PSO performs better than all gradient based approaches. Several examples are included to illustrate the effectiveness of the proposed method in terms of the quality, for better and correct interpretation of EEG.