Signature Management Using Operational Knowledge and Environments (SMOKE)

The Signature Management Using Operational Knowledge and Environments program seeks to develop signature management technologies through two technical areas to generate evasive cyber infrastructure incorporating counter-attribution techniques, quantitatively measure attribution risk in real-time, and maintain evasiveness after infrastructure changes. The Defense Advanced Research Projects Agency will evaluate proposals to accelerate cyber operations and eliminate signatures as an attribution source. Proposals are due by January 31, 2022 for a three year program consisting of two phases with an estimated total funding of $15 million over five years. Proposals will be evaluated based on metrics including time, scalability, attribution precision, recall, and F1 score for the two technical areas.

There is no set-aside designation for this opportunity, with multiple awards anticipated for the two technical areas of automated planning and execution of attribution-aware infrastructure and discovery and generation of infrastructure signatures. Evaluation of the first technical area will consider components based on time, scalability, and attribution, while the second area will be evaluated on precision, recall, and F1 score. The program aims to generate evasive cyber infrastructure and quantify attribution risk in real-time through counter-attribution techniques and maintaining evasiveness after infrastructure changes.