Towards a Science Base for Cybersecurity
Abstract
The goal was to better understand approaches for building attack-resistant cyber-systems. This involved implementing new system software, designing formalisms for specifying system security properties, and developing program analysis techniques for enforcing those properties. Two operating systems were built to better understand how trusted coprocessors could be leveraged for increased assurance that unmodified software and applications are executing: Nexus provides support for a desktop and Cloud Proxy provides support for applications running in cloud. The specification and enforcement of information-use policies that could tag values was also investigated. Here, a theory of RIF (reactive information flow) labels was developed to support re-classification of information as it is transformed by program execution. The theory was then the basis for a new type system, and that type system was retrofit into a programming language.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 08, 2016
- Accession Number
- AD1010314
Entities
People
- Fred B. Schneider
Organizations
- Cornell University