Cyber Threat Insertion and Evaluation Technology for Navy Ship Control Systems

Amida's Hardware Security Platform illuminates microelectronic vulnerabilities through a patented graph transformation of Register Transfer Level (RTL) designs. Unlike traditional approaches that model transfer functions as nodes and connectivity as edges, our methodology inverts this relationship—mapping data elements to nodes and transformations to edges. This innovation answers two critical questions: "Is this line of code vulnerable to exploitation?" and "How much influence does it have across other elements?" The Phase II extension enhances our platform with three objectives: evaluate a real-world Navy RTL design with static vulnerability analysis, perform quantitative dynamic security assessment (quantitative analysis) using operational vectors, and instrument the design for attack detection and remediation.Our static analysis identifies eleven vulnerability types (with an additional one or more planned vulnerabilities) across three categories: Behavior Modifiers (kill-switches, takeovers), Reliability Impacts (accelerated aging, spoofed faults), and Leakers (infiltration, exfiltration). Each vulnerability receives a severity score, creating a rank-ordered list of exploitable nodes. Dynamic assessment will build on static analysis by incorporating operational activity data, providing deeper insights into which vulnerabilities represent actual threats during system operation.The platform includes a sophisticated visualizer that displays the influence graph, vulnerabilities, and linked RTL source code, enabling comprehensive exploration through specialized queries. We will leverage GenAI to create and integrate targeted instrumentation, including detection instruments (value comparators, toggle detectors), learning instruments (signal capture, threshold monitoring), and remediation instruments (redundant voting, error correction). This methodology iteratively evaluates RTL designs to identify and safeguard structures susceptible to hardware attacks. Seven patents from previous phases demonstrate technical success, while current implementation at the Department of Veterans Affairs confirms commercial viability. The platform significantly reduces vulnerability detection time compared to traditional methods, providing quantitative security metrics that objectively measure hardening effectiveness.