Electronic information used under EHR(Electronic Health Recode) environment can be searched, modified or deleted in all the systems connected with EHR system inside a hospital, thanks to network systems. Inside a hospital, after logging in EHR system, RBAC system, which allots proper rights to users, is being used. In RBAC(Role-based access control), users' role can be misused or abused through packet modulation by network intrusion and modulation on log-in information of users. Current RBAC system is preventing misuse or abuse of users by managing authority allotment included in role, but users' information that is abused through network modulation can be detected through efficient network monitoring and network traffic analysis. However, in the environment where EHR system is connected even to the Internet, much analysis time is needed to find abused and misused roles only by monitoring network traffic. In this thesis, we, using Principal Component Analysis, can draw principal components of related data transmitted from RBAC system, make linear transformation, analyze correlation and detect use of improper role efficiently among a lot of network traffic.
展开▼
