Removal of power line interference in EEG signals with spike noise based on robust adaptive filter
Jianqiang Lin, Xu Sun, Jiafei Wu, S. C. Chan, Weichao Xu · 2016
This paper proposes a general robust active noise control (RANC) framework for removing power line interference (PLI) from the Electroencephalogram (EEG) signals when both reference and primary signals are contaminated by spike noise. It is obtained by exploiting the robust property of M-estimation function against impulses. According to this new framework, two commonly used methods, namely, least mean squares (LMS) and normalized least mean squares (NLMS), are extended in parallel to least mean M-estimate (LMM) and normalized least mean M-estimate (NLMM), respectively. Experimental results based on the benchmark MIT-BIH Polysomnographic Database show the sufficient capability of our methods for cancelling PLI in EEG signals with noisy observations and excellent performance for rejecting spikes existing in PLI and EEG signals.