A Real-Time Intrusion Detection and Protection System at System Call Level under the Assistance of a Grid

摘要

In this paper, we propose a security system, named the Intrusion Detection and Protection System (IDPS for short) at system call level, which creates personal profiles for users to keep track of their usage habits as the forensic features, and determines whether a legally login users is the owner of the account or not by comparing his/her current computer usage behaviors with the user's computer usage habits collected in the account holder's personal profile. The IDPS uses a local computational grid to detect malicious behaviors in a real-time manner. Our experimental results show that the IDPS's user identification accuracy is 93%, the accuracy on detecting its internal malicious attempts is up to 99% and the response time is less than 0.45 sec., implying that it can prevent a protected system from internal attacks effectively and efficiently.