EEG signal classification with different signal representations

If several mental states can be reliably distinguished by recognizing patterns in EEG, then a paralyzed person could communicate to a device like a wheelchair by composing sequences of these mental states. In this article, the authors report on a study comparing four representations of EEG signals and their classification by a two-layer neural network with sigmoid activation functions. The neural network is implemented on a CNAPS server (128 processor, SIMD architecture) by Adaptive Solutions Inc., gaining a 100-fold decrease in training time over a Sun Sparc10 for a large number of hidden units.