The Commissariat à l’Énergie Atomique et aux Énergies Alternatives (CEA), is a major public research institute active in three main fields: energy, information and health technologies, and defense and national security. CEA represents 16000 employees with a portfolio of more than 5800 patents.
CEA List, a CEA institute, is a key software systems and technology research centre. it focuses on three areas with disruptive societal and economic implications: embedded systems, interactive systems, signal detection and processing. CEA List research activities span from conceptual design of systems to pre- industrial prototypes, with a strong emphasis on technology transfer.
The LSL Laboratory of CEA List has been conceiving and developing software verification tools for many years, and counts several successful industrial transfers, most notably its open source Frama-C platform used at Airbus, EDF and NASA. More recently, LSL has acquired a strong expertise in applying formal methods to software security (e.g., verification of PolarSSL, collaborations with Thales and Bureau Veritas), especially for binary-level security analysis through the BINSEC platform. LSL is also strongly involved in several validation- and security-related committees, both national (Allistene GT Cybersecurity) and European (ECSO GTs Certification & Roadmap).
Many crucial security analyses must be performed at the binary level, i.e. on executable files, either because source code is not available (commercial off-the-shelf, mobile or legacy code) or in order to discover subtle bugs introduced at compile time. Typical examples include vulnerability detection and malware analysis. Those tasks are currently poorly tooled: while syntactic approaches (based on the way the program is written) are very easy to fool, dynamic approaches (based on the observation of one or few execution traces) miss rare behaviors
CEA LIST will thus bring its strong expertise in security-oriented binary code analysis to the project. In particular, CEA LIST will focus on adapting and expanding its open-source BINSEC platform to meet the needs of the project. It will thus design and implement new binary-level analyses to provide a continuum from precise dedicated analyses to large-scale analyses within the specific contexts of security-oriented monitoring and forensic activities. The spectrum of analyses CEA LIST aims to develop will encompass lightweight syntactic approaches to more time-consuming semantic methods in order to provide the necessary finetuning and scaling for the different aspects of C4IIoT. Furthermore, we will add suitable parametrization mechanisms to enhance the adaptability of the underlying techniques. The tools developed by CEA LIST will act as a link in C4IIoT’s toolchain and will thus work in conjunction with detection, remediation and attribution activities. In particular, it will exploit and export relevant data given or needed by these activities. As such, CEA will be leading the relevant task concerning the mitigation engine (T3.3).
Furthermore, CEA will strongly support the deployment of the implemented technologies from integration and testing to the implications of users on the use-cases of the project (leader of T1.4). Finally, it will be responsible for project quality planning (T7.1) and also be taking part in the scientific dissemination – industrial and academic – of the project.