Keystroke dynamic analysis using relative entropy & timing sequence Euclidian distance

Biometric authentication systems are becoming more and more popular because of increased security concerns. Keystroke Dynamics is one of the important behavior based biometric trait. It has moderate uniqueness level and low user cooperation is required. In this paper keystroke dynamics analysis using relative entropy and Euclidian distance between keystroke timing sequence is discussed. In this approach keystroke timing sequence are calculated and normalized then this information is used for generating normalized probability distribution of dynamic passwords, two distance measures namely relative entropy & Euclidian distance are used for classification. This approach is simple and feasible for multimodal biometric systems.