ECE MS Project Defense by Hsin Yu-Yu

ABSTRACT:
The authentication of patients is critical for health care applications. Especially, in the wireless body area networks (WBANs), we need to make sure that the wirelessly collected vital signs are from the person we identified. Some attackers could impersonate our patients and make the medical system risky and disclose private medical information. Then, the challenging question is how we recognize the right patients wearing wireless body sensors and distinguish them from other or susceptible patients.

In this project, rather than using the password-based methods to authenticate a patient, a biometric-based authentication approach is implemented to distinguish patients through collected ECG (Electrocardiography) signals. The QRS complex extracted from noisy ECG are unique features that are used to identify a person. In this project, I investigated two typical classification approaches for the authentication: SVM (support vector machine) and KNN (K nearest neighbor) under different noise (i.e., white, brown and pink) environments. A comprehensive ECG feature extraction procedure is implemented. The experimental results demonstrate that the implemented approach can achieve high accuracy for the patient authentication. This bio-metric based authentication approach could be used for future healthcare-security-related applications over WBANs.

NOTE: All ECE Graduate Students are ENCOURAGED to attend.
All interested parties are invited to attend.
Open to the public.

Advisor: Dr. Honggang Wang
Committee Members: Dr. Paul J. Fortier and Dr. Liudong Xing