Adaptive filtering of EEG/ERP through noise cancellers using an improved PSO algorithm

In this paper, event related potential (ERP) generated due to hand movement is detected through the adaptive noise canceller (ANC) from the electroencephalogram (EEG) signals. ANCs are implemented with least mean square (LMS), normalized least mean square (NLMS), recursive least square (RLS) and evolutionary algorithms like particle swarm optimization (PSO), bacteria foraging optimization (BFO) techniques, genetic algorithm (GA) and artificial bee colony (ABC) optimization technique. Performance of this algorithm is evaluated in terms of signal to noise ratio (SNR) in dB, correlation between resultant and template ERP, and mean value. Testing of their noise attenuation capability is done on EEG contaminated with white noise at different SNR levels. A comparative study of the performance of conventional gradient based methods like LMS, NLMS and RLS, and swarm intelligence based PSO, BFO, GA and ABC techniques is made which reveals that PSO technique gives better performance in average cases of noisy environment with minimum computational complexity.