SOBAS

The evolution of systems embedded into commercial aircraft and spacecraft is raising security problems related to: - increased openness to applications and equipment provided by the clients (airline companies or satellite mission customers) and to digital networks (air traffic control, airline computer networks, Internet access offered to passengers, ground stations managed by satellite mission customers); - increased sharing of computing resources among different applications, possibly with different safety requirements; - increased use of Commercial-Off-The-Shelf (COTS) hardware and software components. To solve these security problems, it is necessary to apply or adapt security methods and techniques that have proven their efficiency in other contexts: - Formal methods for specification, development and verification; - Security mechanisms and tools (firewalls, VPNs, access control, IDS, etc.); - Vulnerability assessment and countermeasure provision. This latter point is the main focus of the project SOBAS. Indeed, few research has been devoted to this security aspect, as opposite with formal methods. Yet, there is currently no formal model able to cover both high-level abstractions, where security properties can be defined, and low-level implementation details, where most vulnerabilities lie: OS kernel implementation of address space protection, interrupt management, context switching, etc.; and hardware implementation of protection mechanisms and other ancillary functions that could be exploited by attackers. These low-level vulnerabilities are the main concern of SOBAS. Traditionally, low-level vulnerability discovery and analysis are performed when the system has been fully implemented, since it is in the implementation that most vulnerabilities are present. The main breakthroughs aimed at by this project are: - integrating low-level vulnerability assessment and countermeasure provision into the development process, so as to identify classes of vulnerabilities before the system is fully implemented; to do that, security assurance activities (similar to quality assurance activities) will be developed to guide the development process; - developing generic countermeasures for these vulnerability classes, generic in the sense that they can be adapted to different implementations, and efficient even for still unknown vulnerabilities; - evaluating the efficiency of these countermeasures and their applicability to aerospace systems; computing system architectures embedded into aircraft being very different from those embedded into spacecraft, two case studies will be provided by Airbus and by Astrium, and the solutions will be implemented into two platforms dedicated to these case studies. Specific tools will be developed to evaluate the efficiency of the countermeasures in these two contexts, aeronautics and space. The partners participating in this project are: - LAAS-CNRS, which brings its competence on security and dependability, in particular in the aerospace domain; - XLIM, which brings its competence on security of embedded systems, and on virtualisation-based embedded systems; - Airbus, which provides a case study based on its future avionics systems, and participates in the security assurance activities and in the evaluation of applicability to the aeronautics domain; - Astrium, which provides a case study based on future multi-mission satellites, and participates in the security assurance activities and in the evaluation of applicability to the space domain; - EADS-IW, which brings its experience in embedded systems vulnerability analysis, as well as in security assurance and security evaluation; - ANSSI, which is the French National Agency for Information System Security.