EOG artifact removal from EEG using a RBF neural network

In this paper, a new adaptive radial-basis function- networks- (RBFN-) based filter for theadaptive noise cancellation (ANC) problem is proposed. The algorithm of structure identificationand parameters adjustment is developed. The proposed RBFN-based filtering approachimplements Takagi-Sugeno-Kang (TSK) fuzzy systems functionally. The RBFN-based filter hasthree major features: (1) No space pre partitioning is needed; (2) No predetermination, such asthe number of RBF neurons (fuzzy rules), must be given; (3) Fast learning speed is achieved.Simulation results demonstrate that the proposed adaptive RBFN-based filter can cancel thenoise successfully and efficiently with a parsimonious structure.

Keywords:

RBFN , ANC , adaptive filtering