Year
1988
Abstract
This paper describes a prototype safe- guards analysis tool that automatically identi- fies system weaknesses against an insider ad- versary and suggests possible upgrades to improve the probability that the adversary will be detected. The tool is based on this premise: as the adversary acts, he or she creates a set of facts that can be detected by safeguards com- ponents. Whenever an adversary's planned set of actions create a set of facts which the secu- rity personnel would consider irregular or unusual, we can improve the security system by implementing safeguards that detect those facts. Therefore, an intelligent computer pro- gram can suggest upgrades to the facility if we construct a knowledge base that contains in- formation about: (1) the facts created by each possible adversary action, (2) the facts that each possible safeguard can detect, and (3) groups of facts which will be considered irregular whenever they occur together. We describe the structure of the knowledge base and show how the above information can be represented in it. We also describe the proce- dures that a computer program can use to identify missing or weak safeguards and to suggest upgrades.