AVIATION

Cignal LLC Awarded Phase 3 Funding by DHS S&T SVIP

Cignal, a technology startup that develops advanced simulation and high-performance computing (HPC) environments to rapidly train Automated Threat Recognition (ATR) and artificial intelligence (AI) systems, announced recently that it was awarded Phase 3 project funding by the Department of Homeland Security (DHS) Science and Technology Directorate’s (S&T) Silicon Valley Innovation Program (SVIP).

In continued collaboration with S&T’s Screening at Speed Program, Cignal will test and validate the accuracy, utility, and realism of volumetric, synthetic X-ray computed tomography (CT) data that are generated within its platform, such as weapons, stream-of-commerce objects, and prohibited items. Cignal also will perform research and development into On-Person Screening (OPS) systems to identify a pathway for integrating millimeter wave (mmWave) transmission and simulation into Cignal’s patent-pending compute and rendering engine. These efforts are essential to Cignal’s goal of solving critical national security problems in aviation security and screening.

“We’ve developed the Cignal Engine to help the industry develop, train, and test advanced AI systems that improve detection while maintaining efficiency and speed at the checkpoint,” saidEric Fiterman, Cignal chief technology officer. “We’re pleased to continue our work with DHS S&T in our Phase 3 project, which allows us to build on our successes with checkpoint CT and apply them to OPS systems. This will provide comprehensive data services for additional checkpoint screening systems and more advanced detection models.”

During its Phase 3 project, Cignal will use the Cignal Engine, which is a flexible suite of HPC software components and environments that enable the dynamic training, evaluation, and deployment of next-generation ATR systems for emerging and complex threats. The Cignal Engine permits users to create new screening paradigms, digital twins, novel objects, and new AI and machine learning (ML) models to protect the public and critical infrastructure. Using this engine and its innovative compression techniques, Cignal created the industry’s first volumetric million-bag dataset with automated labels, which supports advanced X-ray and X-ray CT systems and AI/ML training for large-scale, trillion-parameter ATR models.

Advanced screening, inspection, and security ATR and AI systems need a constant source of balanced training datasets to deliver better detection capability and differentiate legitimate commerce from threats and illicit activity. However, data in these use cases typically resides in the non-visible spectra, such as X-ray, gamma ray, infrared, and mmWave. Traditionally, acquiring training or test data for these spectra is time-consuming and expensive, and it can easily exhaust memory, compute, and storage budgets. Cignal’s innovative approach to data creation allows these non-visible datasets to be generated, shared, and stored faster, with fewer limitations and at a greatly reduced cost, for device manufacturers, researchers, and government personnel responsible for securing ports, transportation infrastructure, mail and cargo systems, and more.

About Cignal LLC

Cignal LLC makes advanced simulation and high-performance computing environments to help next-generation Automated Threat Recognition systems, artificial intelligence models, and human screeners in protecting the public and critical infrastructure against emerging and complex threats. Cignal’s flexible engine for security, screening, and inspection can be deployed on-premises, in the cloud, or at the edge, and it provides users with full control over their data. Cignal LLC is headquartered in Central Pennsylvania. For more information, please visit www.cignal.co or contact innovation@cignal.co.

(Research reported in this press release was supported by the Department of Homeland Security, Science and Technology Directorate under Award Number 70RSAT22T00000014. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Department of Homeland Security.)