Analyzing and Improving Stochastic Network Security: A Multicriteria Prescriptive Risk Analysis Model
Abstract
This research optimized two measures of network security by hardening components and improving their reliability. A common measure of effectiveness (MOE) for networks is statistical reliability, which ignores the effects of hostile actions. A new MOE which includes hostile actions was developed. Both measures require component reliability functions, derived using fault trees. Fuzzy Logic and Monte Carlo simulation were used to quantify uncertainty. Results from the model are compared to traditional Risk Assessment results.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1997
- Accession Number
- ADA326934
Entities
People
- David L. Lyle
Organizations
- Air Force Institute of Technology