EEG/ERP Adaptive Noise Canceller Design with Controlled Search Space (CSS) approach in Cuckoo and other Optimization Algorithms
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TCBB.2013.119
M. K. Ahirwal , PanditDwarka Prasad Mishra Indian Institute of Information Technology, Design & Manufacturing Jabalpur, Jabalpur
A. Kumar , PanditDwarka Prasad Mishra Indian Institute of Information Technology, Design & Manufacturing Jabalpur, Jabalpur
G. K. Singh , Indian Institute of Technology Roorkee, India, Jabalpur and University of Malaya, Malaysia
This paper explores the migration of adaptive filtering with swarm intelligence /evolutionary techniques employed in the field of electroencephalogram /Event related potential (EEG /ERP) noise cancellation or extraction. A new approach is proposed in form of Controlled Search Space (CSS) to stabilize randomness of swarm intelligence techniques specially for EEG signal filtering. Three most recently developed swarm population based algorithms such as Particles Swarm Optimization (PSO), Artificial Bee Colony (ABC), and Cuckoo Optimization algorithm (COA) with their several variants are implemented to design optimized adaptive noise canceller. The proposed controlled search space technique is tested on each of the swarm intelligence techniques, and it is found to be more accurate and powerful. Adaptive noise canceller with traditional algorithms such as LMS, NLMS and RLS algorithms are also implemented to compare the simulation results. Three types of ERP data are used such as simulated visual evoked potential (S-VEP), real visual evoked potential (R-VEP) and real sensorimotor evoked potential (R-SEP) due to their physiological importance in various EEG studies.
Optimization, Medical simulation
M. K. Ahirwal, A. Kumar and G. K. Singh, "EEG/ERP Adaptive Noise Canceller Design with Controlled Search Space (CSS) approach in Cuckoo and other Optimization Algorithms," in IEEE/ACM Transactions on Computational Biology and Bioinformatics.