North Star Labs - CyPHER Network Sensor Phase II

North Star Labs proposes to adapt and transition the Cyber Phenomenology Exploration & Reasoning (CyPhER) sensor,  originally developed under the Better Networks project at the Department of the Air Force AI Accelerator at the Massachusetts Institute of Technology Lincoln Labs (MIT LL), for deployment across the United States Air Force (USAF) network security infrastructure. CyPhER is a novel, lightweight, and signature-independent cybersecurity sensor designed to detect adversarial activity across large-scale, long-range networks.  It can be deployed in conventional networks or emerging space architectures, autonomous edge-based mesh networks, and cyber physical systems. By leveraging novel, anonymized data structures and advanced AI techniques, CyPhER provides full network coverage to identify emerging threats early in their planning and staging phases, significantly ahead of conventional detection systems. Laboratory tests by MIT and North Star Labs benchmark the solution as at least 200X faster than current market solutions, and provide full network security coverage with no downsampling. The solution is currently Technology Readiness Level (TRL) 5-6. With active support from the 16th Air Force, the 688th Cyberspace Wing, the 33rd Cyberspace Operations Squadron, and USAF Cyberworx, North Star Labs will act upon the findings of the Phase I SBIR feasibility study within a P2 STTR with MIT Lincoln Labs. With our MIT LL partner, we will test and validate identified warfighter needs, and iteratively mature the capability using representative data and environments, codify performance optimizations to transform research code into a commercial grade product, conforming to relevant Security requirements. Core areas of focus include optimizations for network data processing, continuous monitoring, threat detection, and mitigation.