Title :
Using publicly known passwords with haptics and biometrics user verification
Author :
Kanneh, Andrea ; Stoute, Valerie ; Smith, Michael
Author_Institution :
Univ. of Trinidad & Tobago, Trinidad and Tobago
Abstract :
The use of username-password combinations is the simplest approach to system access to administer but these systems continue to pose several problems to both the user and to the system administrator. Using Biometrics for system access gives greater security as this offers an approach that is more difficult to copy, share or steal. Haptics technology allows these biometric systems to measure also the relative force a user exerts to carry out a task. Studies, to date, using Haptics and Biometrics for user identity verification have been able to attain accuracy of up to 95%, on average. The work presented here has increased this figure by 3%, with the use of a heuristics decision-making algorithm based on the neural network concept. Key features of the heuristics system include allowing either standard or individualized feature selection and permitting imprecision in decision-making to cater for human variability. The study also illustrates that a unique password can be created even when the participants perform the same or similar actions.
Keywords :
biometrics (access control); decision making; feature extraction; haptic interfaces; neural nets; security of data; biometrics user identity verification system; feature selection; haptics user identity verification; heuristics decision-making algorithm; heuristics system; human variability; neural network; system administrator; username-password combination; Authentication; Conferences; Education; Extraterrestrial measurements; Indexes; Instruments; Neural networks; Biometrics; Data preprocessing; Haptics; Heuristic algorithms; Pattern recognition;
Conference_Titel :
Haptics Symposium (HAPTICS), 2012 IEEE
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4673-0808-3
Electronic_ISBN :
978-1-4673-0807-6
DOI :
10.1109/HAPTIC.2012.6183847