Optimization of Adaptive Noise Canceller with Grey Wolf Optimizer for EEG/ERP Signal Noise Cancellation

This paper presents an optimization algorithm for Adaptive Noise Canceller by taking Electroencephalography/Event-Related Potentials (EEG/ERP) as an input signal. In this paper Adaptive Noise Cancellers are implemented by utilizing gradient based algorithms, swarm-based algorithm and nature inspired algorithm which is Grey Wolf Optimization (GWO). To improve the ANC performance, an optimal weight value is calculated and GWO is used to update optimal weight value. Testing of an adaptive filter has been done through consideration of White Gaussian Noise (WGN) over sample input EEG signals at various SNR levels. The performance of the GWO algorithm is evaluated in terms of mean value, Signal to Noise Ratio (SNR) in dB and correlation. A comparative analysis shows that proposed GWO technique gives better performance when compared with the gradient-based techniques like RLS, LMS and swarm-based technique like Particle Swarm Optimization (PSO). The novelty of this proposed approach can be considered, as because first-time GWO is applied on denoising EEG signals from the contaminated EEG signal, the result shows that extracting the desired EEG component is more effective in the proposed GWO method.