Related Experiment Video
Updated: May 4, 2026
10:08
The Evoked Potential Operant Conditioning System EPOCS: A Research Tool and an Emerging Therapy for Chronic Neuromuscular Disorders
Published on: August 25, 2022
2.8K
EEG/ERP adaptive noise canceller design with controlled search space (CSS) approach in cuckoo and other optimization
M K Ahirwal1, Anil Kumar1, G K Singh2
1PanditDwarka Prasad Mishra Indian Institute of Information Technology, Design & Manufacturing Jabalpur, Jabalpur.
IEEE/ACM Transactions on Computational Biology and Bioinformatics
|January 11, 2014
Summary
This study introduces a controlled search space for swarm intelligence in electroencephalogram (EEG) noise cancellation. The new method enhances adaptive filtering accuracy and power for event-related potential (ERP) extraction.
Area of Science:
- Biomedical Engineering
- Computational Neuroscience
- Signal Processing
Background:
- Adaptive filtering is crucial for noise cancellation in electroencephalogram (EEG) signals.
- Swarm intelligence and evolutionary techniques offer potential for optimizing adaptive filters.
- Event-related potentials (ERPs) are vital for EEG studies but susceptible to noise.
Purpose of the Study:
- To explore swarm intelligence/evolutionary techniques for EEG/ERP noise cancellation.
- To propose a novel controlled search space approach to stabilize swarm intelligence algorithms.
- To enhance the accuracy and power of adaptive noise cancellers for EEG signals.
Main Methods:
- Implemented swarm-based algorithms: Particle Swarm Optimization, Artificial Bee Colony, Cuckoo Optimization Algorithm.
- Developed and tested a controlled search space technique for swarm intelligence.
- Compared results with traditional adaptive filters: LMS, NLMS, and RLMS.
- Utilized simulated and real ERP signals (visual evoked potential, sensorimotor evoked potential).
Main Results:
- The proposed controlled search space significantly improved accuracy and power of swarm intelligence techniques.
- Evolutionary techniques demonstrated an average computational time of 0.821 sec and shape measure of 0.173.
- Traditional algorithms had negligible time consumption but poor ERP shape preservation (0.0141 sec difference, 2.60 shape measure difference).
Conclusions:
- The controlled search space enhances swarm intelligence for robust EEG noise cancellation.
- Swarm intelligence with the proposed method offers superior ERP shape preservation compared to traditional algorithms.
- This approach holds promise for improving the quality of EEG signal analysis in various studies.
