Exsisting security mechanisms protect computers and networks from unauthorized use through access controls, such as passwords. However, if these access controls are compromised or can be bypassed, an abuser may gain unauthorized access an thus can cause great damage and disruption to systems operation. Modes of protection can be devised against this, however, an intruder is likely to exhibit behavior patterns different than that of a legitimate user. While many systems collect audit data, most do not have any capability for automated analysis of that data. Moreover, those that do collect audit data generally collect large volumes of data that are not necessarily security relevant. Thus, for security analysis, a security officer (SO) must wade through stacks of printed output of audit data. Thus, the capability for automated security analysis of audit trails is needed. The Next-generation Intrusion-Detection Expert System (NIDES) is the result of research that started in the Computer Science Laboratory at SRI International in the early 1980’s and led to a series of increaslingly sophisticated prototypes that resulted in the current NIDES Beta release. The current version, described in this final report and in greater detail in [1,2,3], is designed to operate in real time to detect instruions as they occur. NIDES is a comprehensive system that uses innovative statistical algorithms for anomaly detection, as well as an expert system that encodes known intrusion scenarios.

This report describes a formal security policy model for a secure relational database system.  This model is intended to meet the formal model requirement specified in the DoD Trusted Computer System Evaluation criteria.  The model is formulated in two layers, one corresponding to a reference monitor that enforces mandatory security, and the second defining multilevel relations and formalizing policies for labeling new and derived data, data consistency, discretionary security, and transaction consistency.  The development of a formal security policy model is the second task of the SeaView project to design a multilevel secure database system meeting the Criteria for Class A1.

Generally Accepted System Security Principles incorporate the consensus at a particular times as to the practices, conventions, rules, mechanisms, and procedures that 1) information security professionals should employ, or that 2) information processing products should provide, to achieve, preserve, and restore the properties of integrity, availability, and confidentiality of information and information systems. GSSP is a technical security term encompassing the practices, conventions, rules, mechanisms, and procedures that are needed to define accepted security practice at a particular time. It includes broad guidlines and detailed practices and procedures.