BIOMETRIC AUTHENTICATION SYSTEM USING ELMAN NEURAL NETWORKS

In this paper we investigated the use of brain signature as a possible biometric authentication technique. EEG signals are generated while the subjects perform three various tasks, namely relax, read and spell are acquired from twenty four subjects. We propose an algorithm for recognition of individuals using signal power and backpropagation neural networks. Second methods spell task is more suitable for biometric authentication or verification. The Elman neural network with an average accuracy of 85%- 93%. Twenty four subjects were used in this study. The usage of brain signatures as a possible for biometric verification technique.