Quantifying Information Leakage for Fully Probabilistic Systems
Yunchuan Guo, Yin Lihua, Yuan Zhou, Fang Binxing · 2010
Quantifying the improper leakage of confidential information is a great challenge. In this paper, we propose a method to quantify the information leakage for a fully probabilistic system. Our approach relies on αmutual information (αMI). In our analysis, system is modeled as a fully probabilistic automata; information leakage is identified by means of the weak probabilistic trace equivalence, and then measured via αMI. The accuracy of our approach is demonstrated by experiments.