PERSON AUTHENTICATION USING BRAINWAVES
- 1. A Seminar on “Person Authentication Using Brain Waves” Presented by: Amena Kauser 3GU14EC005 Under The Guidance Of: Prof. Praveen Sindagi DEPT. OF ELECTRONICS AND COMMUNIATION ENGG. GOVERNMENT ENGINEERING COLLEGE, RAICHUR 584134 Dept of Electronics and Communication Engg
- 2. Introduction Biometric Identification. An authentication (or verification) system involves confirming or denying claimed by a person (one-to-one matchnig). Dept of Electronics and Communication Engg
- 3. Biometrics National Institute of science and technology (NIST) defines biometrics as Automated method of identifying or authenticating an Individual based on his or her physiological or behavioural characteristics. The Brain Waves (BW) pattern of every individual is unique and that the EEG (Electroencephalograms) can be used for biometric identification. Biometrics Brain Waves IRIS Palm Vein Retina FingerprintDept of Electronics and Communication Engg
- 4. Characteristics of BW Main source of EEG is the synchronous activity of thousands of cortical neurons Everyone’s BW signals is a bit different even when they think about the same thing In abnormal adults the EEG shows sudden bursts of electrical activity (spikes) or sudden slowing of BW… These abnormal discharges may be caused by a brain tumour, infection, injury, stroke, or epilepsy Brain dead state: Flat EEG Dept of Electronics and Communication Engg
- 5. Types of BW Dept of Electronics and Communication Engg
- 6. Brainwave-based Authentication The use of brain waves signals for user authentication have security advantages ,i.e by the use of pass-thoughts instead of typing in a pass-word. Authentication process have three steps: 1. Collecting of Brainwave data 2. Data Preprocessing 3. Data Analysis Dept of Electronics and Communication Engg
- 7. Database EEG signals were recorded with a Biosemi system using a cap with 32 integrated electrodes. Illustration of the location of the electrodes on the scalp cap. Dept of Electronics and Communication Engg
- 8. Problem Description An identity authentication system has to deal with two kinds of events Either the person claiming the given identity is the one who he claims to be (client) Or he is not (impostor) In general, the system may take two decisions either accept the client or reject him and decide he is an impostor Dept of Electronics and Communication Engg
- 9. FR FA clientimpostor EER τ EER Person Authentication Illustration of typical errors of biometric system during authentication. An impostor above the threshold is a false acceptance. A client below the threshold is the false rejection. Dept of Electronics and Communication Engg
- 10. Applications Access Control Systems Building Gate Control Digital Multimedia Access Transaction Authentication Voice Mail Dept of Electronics and Communication Engg
- 11. Advantages It is confidential (as it corresponds to mental task) It is very difficult to mimic (as similar mental tasks are person dependent) It is almost impossible to steal (as brain activity is sensitive to the stress and mood of the person, and aggressor can’t force the person to reproduce his or her mental pass phrase) Dept of Electronics and Communication Engg
- 12. REFERENCES 1. F. Babiloni, F. Cincotti, L. Lazzarini, J.d.R. Millan, J. Mourino, M. Varsta, J. Heikkonen, L. Bianchi and M.G. Marciani, “Linear classification of low-resolution EEG patterns produced by imagined hand movements”, IEEE Trans. on Rehabilitation Engineering, vol. 8, pp. 186-188, 2000 2. F. Cardinaux, C. Sanderson and S. Marcel, “Comparison of MLP and GMM Classifiers for Face Verification on XM2VTS,” Proceedings of the 4th International Conference on Audio- and Video-Based Biometric Person Authentication, pp. 911-920, 2003. 3. A.P. Dempster, N.M. Laird and D.B. Rubin. “Maximum-likelihood from incomplete data via the EM algorithm” Journal of Royal Statistical Society, Series B (Methodological), vol. 39, no. 1, pp. 1-38, 1977. Dept of Electronics and Communication Engg