Signal processing component of a BCI typically consists
of two parts: feature extraction and decoding (translation)
algorithms. Feature extraction part processes recorded neural
signals, in order to extract signal features, assumed to reflect
specific aspects of a user’s current neural signal. Decoding
algorithms take those abstracted feature vectors and transform
them into application-specific commands. Depending on the
application, many different decoding algorithms are being used
in BCIs. As pointed out in [42], effective decoding algorithms
are able to adapt to: (1) individual user’s signal features,
(2) spontaneous variations in recorded signal quality, and (3)
adaptive capacities of the brain (neural plasticity).
Signal processing component of a BCI typically consistsof two parts: feature extraction and decoding (translation)algorithms. Feature extraction part processes recorded neuralsignals, in order to extract signal features, assumed to reflectspecific aspects of a user’s current neural signal. Decodingalgorithms take those abstracted feature vectors and transformthem into application-specific commands. Depending on theapplication, many different decoding algorithms are being usedin BCIs. As pointed out in [42], effective decoding algorithmsare able to adapt to: (1) individual user’s signal features,(2) spontaneous variations in recorded signal quality, and (3)adaptive capacities of the brain (neural plasticity).
การแปล กรุณารอสักครู่..
