Signal to noise ratio enhancement in BCI with using ICA method in preprocessing step

  • Soheil Mokhlesi Department of Electrical Engineering, Faculty of Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran
  • Hamid Khaloozade
Keywords: Brain-Computer Interface, Preprocessing, Independent Component Analysis, Noise Reduction


The quality of Electroencephalography (EEG) signals is low and these signals mostly are being mixed with physiological and non-physiological noises. The Independent Component Analysis (ICA) is an appropriate and common method for reducing the noises and improving the EEG signals quality. This method reduces the physiological and uncommon noises by allocating each signal to its own source. By doing this, signals become independent and noises are being separated from signals. In this paper we use first dataset of forth competition of BCI Competition website, which this dataset has two class and related to motor imagery of hands and feet. We implement two different experiments on this dataset. The difference of these two experiments is in using the ICA method for reducing noise in preprocessing step of brain-computer interface (BCI). Using ICA increased signal to noise ratio and BCI performance. Common Spatial Pattern (CSP) method is used in feature extraction step of both experiments.


[1] C. S. Kim, J. Sun, D. Liu, Q. Wang, and S. G. Paek, "Removal of ocular artifacts using ICA and adaptive filter for motor imagery-based BCI," IEEE/CAA journal of automatica sinica, 2017.
[2] S. Sanei, S. Sanei, and J. Chambers, "EEG Signal Processing. Centre of Digital Signal Processing Cardiff University, UK (2007)," 2015.
[3] G. Dornhege, J. d. R. Millan, T. Hinterberger, D. J. McFarland, and K.-R. Müller, Toward brain-computer interfacing. MIT press, 2007.
[4] I. Manuel and B. Nunez, "EEG Artifact Detection," Project Report, Department of Cybernetics Czech TechnicaIUniversity, 2010.
[5] E. Hortal et al., "SVM-based Brain–Machine Interface for controlling a robot arm through four mental tasks," Neurocomputing, vol. 151, pp. 116-121, 2015.
[6] R. Kottaimalai, M. P. Rajasekaran, V. Selvam, and B. Kannapiran, "EEG signal classification using principal component analysis with neural network in brain computer interface applications," in 2013 IEEE International Conference on Emerging Trends in Computing, Communication and Nanotechnology (ICECCN), 2013: IEEE, pp. 227-231.
[7] L. F. Nicolas-Alonso and J. Gomez-Gil, "Brain computer interfaces, a review," sensors, vol. 12, no. 2, pp. 1211-1279, 2012.
[8] I. Rejer and P. Górski, "Independent Component Analysis for EEG data preprocessing-algorithms comparison," in IFIP International Conference on Computer Information Systems and Industrial Management, 2013: Springer, pp. 108-119.
[9] B. Blankertz, G. Dornhege, M. Krauledat, K.-R. Müller, and G. Curio, "The non-invasive Berlin brain–computer interface: fast acquisition of effective performance in untrained subjects," NeuroImage, vol. 37, no. 2, pp. 539-550, 2007.
[10] D. Garrett, D. A. Peterson, C. W. Anderson, and M. H. Thaut, "Comparison of linear, nonlinear, and feature selection methods for EEG signal classification," IEEE Transactions on neural systems and rehabilitation engineering, vol. 11, no. 2, pp. 141-144, 2003.
[11] B. Blankertz, S. Lemm, M. Treder, S. Haufe, and K.-R. Müller, "Single-trial analysis and classification of ERP components—a tutorial," NeuroImage, vol. 56, no. 2, pp. 814-825, 2011.
[12] B. Blankertz, G. Curio, and K.-R. Müller, "Classifying single trial EEG: Towards brain computer interfacing," in Advances in neural information processing systems, 2002, pp. 157-164.