Experimental Evaluation of EEG Preprocessing Pipelines for Low-Channel Real-Time Brain–Computer Interfaces
DOI:
https://doi.org/10.32515/2664-262X.2026.13(44).90-96Keywords:
computer system, EEG signal processing, data processing algorithms, digital filtering, signal quality control, human-machine interface (HMI)Abstract
The purpose of this study is to substantiate and experimentally evaluate effective EEG signal preprocessing strategies for low-cost non-invasive brain–computer interfaces intended for real-time operation. Such systems are characterized by a limited number of channels, restricted computational resources, and high sensitivity to noise and artifacts, which makes the choice of preprocessing methods critical for signal quality and result reproducibility. The work aims to identify practically applicable preprocessing sequences that ensure a balance between noise suppression, preservation of informative components, and computational efficiency under real-time constraints.
The study is based on an experimental analysis of an 8-channel EEG recording acquired at a sampling rate of 250 Hz. The influence of key preprocessing stages, including DC component removal, notch filtering at 50 Hz, band-pass filtering in the 8–30 Hz range, re-referencing, normalization, and artifact control, is systematically investigated. Both time-domain and frequency-domain analyses are employed to assess the effects of individual processing steps. Quantitative metrics such as spectral power in selected frequency bands, normalized contribution of power-line interference, and root mean square amplitude are used to evaluate signal quality. Several generalized preprocessing pipelines are formulated for real-time operation, offline analysis, and artifact detection, with attention to phase characteristics, processing latency, and reproducibility.
The results demonstrate that appropriate combinations of DC removal and notch filtering effectively suppress power-line interference without distorting relevant EEG rhythms, while band-pass filtering concentrates signal energy within the sensorimotor frequency range. The proposed preprocessing sequences provide a transparent and reproducible intermediate validation stage prior to classification tasks in low-cost BCI systems. The findings confirm the suitability of simple, interpretable preprocessing methods for real-time applications and form a practical basis for further evaluation of their impact on motor imagery classification performance.
References
Список літератури
1. Caiado F., Ukolov A. The history, current state and future possibilities of the non-invasive brain computer interfaces. Medicine in Novel Technology and Devices. 2025. Vol. 25. P. 100353. DOI: 10.1016/j.medntd.2025.100353.
2. Nonstationary nature of the brain activity as revealed by EEG/MEG: Methodological, practical and conceptual challenges / A. Y. Kaplan et al. Signal Processing. 2005. Vol. 85, no. 11. P. 2190–2212. DOI: 10.1016/j.sigpro.2005.07.010.
3. Linear Modeling of Neurophysiological Responses to Speech and Other Continuous Stimuli: Methodological Considerations for Applied Research / M. J. Crosse et al. Frontiers in Neuroscience. 2021. Vol. 15. DOI: 10.3389/fnins.2021.705621.
4. Venkatachalam K. L., Herbrandson J. E., Asirvatham S. J. Signals and Signal Processing for the Electrophysiologist. Circulation: Arrhythmia and Electrophysiology. 2011. Vol. 4, no. 6. P. 965–973. DOI: 10.1161/circep.111.964304.
5. Rejer I., Cieszyński Ł. Independent component analysis for a low-channel SSVEP-BCI. Pattern Analysis and Applications. 2018. Vol. 22, no. 1. P. 47–62. DOI: 10.1007/s10044-018-0758-4.
6. Rejer I., Górski P. MAICA: an ICA-based method for source separation in a low-channel EEG recording. Journal of Neural Engineering. 2019. Vol. 16, no. 5. P. 056025. DOI: 10.1088/1741-2552/ab36db.
7. Comparison of the accuracy of machine learning algorithms for brain-computer interaction based on high-performance computing technologies / V. Stefanyshyn et al. Scientific journal of the Ternopil national technical university. 2024. Vol. 115, no. 3. P. 82–90. DOI: 10.33108/visnyk_tntu2024.03.082
8. Матвєєва Н., Іваниця Д. Попередня обробка та аналіз ЕКГ сигналів. Herald of Khmelnytskyi National University. Technical sciences. 2025. Т. 355, № 4. С. 388–394. DOI: 10.31891/2307-5732-2025-355-55
9. Mdluli B., Khumalo P., Maswanganyi R. C. Signal Preprocessing, Decomposition and Feature Extraction Methods in EEG-Based BCIs. Applied Sciences. 2025. Vol. 15, no. 22. P. 12075. DOI: 10.3390/app152212075.
10. Orekhova E. V., Wallin B. G., Hedström A. Modification of the Average Reference Montage. Journal of Clinical Neurophysiology. 2002. Vol. 19, no. 3. P. 209–218. DOI: 10.1097/00004691-200206000-00004.
11. Aydın S., Melek M., Gökrem L. Intersession Robust Hybrid Brain–Computer Interface: Safe and User-Friendly Approach with LED Activation Mechanism. Micromachines. 2025. Vol. 16, no. 11. P. 1264. DOI: 10.3390/mi16111264.
12. A Review on Signal Processing Approaches to Reduce Calibration Time in EEG-Based Brain–Computer Interface / X. Huang et al. Frontiers in Neuroscience. 2021. Vol. 15. DOI: 10.3389/fnins.2021.733546.
References
1. Caiado, F., & Ukolov, A. (2025). The history, current state and future possibilities of the non-invasive brain computer interfaces. Medicine in Novel Technology and Devices, 25, 100353. https://doi.org/10.1016/j.medntd.2025.100353
2. Kaplan, A. Y., Fingelkurts, A. A., Fingelkurts, A. A., Borisov, S. V., & Darkhovsky, B. S. (2005). Nonstationary nature of the brain activity as revealed by EEG/MEG: Methodological, practical and conceptual challenges. Signal Processing, 85(11), 2190–2212. https://doi.org/10.1016/j.sigpro.2005.07.010
3. Crosse, M. J., Zuk, N. J., Di Liberto, G. M., Nidiffer, A. R., Molholm, S., & Lalor, E. C. (2021). Linear modeling of neurophysiological responses to speech and other continuous stimuli: Methodological considerations for applied research. Frontiers in Neuroscience, 15. https://doi.org/10.3389/fnins.2021.705621
4. Venkatachalam, K. L., Herbrandson, J. E., & Asirvatham, S. J. (2011). Signals and signal processing for the electrophysiologist. Circulation: Arrhythmia and Electrophysiology, 4(6), 965–973. https://doi.org/10.1161/circep.111.964304
5. Rejer, I., & Cieszyński, Ł. (2018). Independent component analysis for a low-channel SSVEP-BCI. Pattern Analysis and Applications, 22(1), 47–62. https://doi.org/10.1007/s10044-018-0758-4
6. Rejer, I., & Górski, P. (2019). MAICA: An ICA-based method for source separation in a low-channel EEG recording. Journal of Neural Engineering, 16(5), 056025. https://doi.org/10.1088/1741-2552/ab36db
7. Mdluli, B., Khumalo, P., & Maswanganyi, R. C. (2025). Signal preprocessing, decomposition and feature extraction methods in EEG-based BCIs. Applied Sciences, 15(22), 12075. https://doi.org/10.3390/app152212075
8. Matvieieva, N., & Ivanytsia, D. (2025). Poperednia obrobka ta analiz EKH syhnaliv. Herald of Khmelnytskyi National University. Technical Sciences, 355(4), 388–394 [in Ukrainian]. https://doi.org/10.31891/2307-5732-2025-355-55
9. Stefanyshyn, V., Stefanyshyn, I., Pastukh, O., & Kulikov, S. (2024). Comparison of the accuracy of machine learning algorithms for brain-computer interaction based on high-performance computing technologies. Scientific Journal of the Ternopil National Technical University, 115(3), 82–90. https://doi.org/10.33108/visnyk_tntu2024.03.082
10. Orekhova, E. V., Wallin, B. G., & Hedström, A. (2002). Modification of the average reference montage. Journal of Clinical Neurophysiology, 19(3), 209–218. https://doi.org/10.1097/00004691-200206000-00004
11. Aydın, S., Melek, M., & Gökrem, L. (2025). Intersession robust hybrid brain–computer interface: Safe and user-friendly approach with LED activation mechanism. Micromachines, 16(11), 1264. https://doi.org/10.3390/mi16111264
12. Huang, X., Xu, Y., Hua, J., Yi, W., Yin, H., Hu, R., & Wang, S. (2021). A review on signal processing approaches to reduce calibration time in EEG-based brain–computer interface. Frontiers in Neuroscience, 15. https://doi.org/10.3389/fnins.2021.733546
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Serhii Solovey
This work is licensed under a Creative Commons Attribution 4.0 International License.
