U.S. Navy Leverages Biometric Authentication Technologies with GBL Systems and AWS

GBL Systems helped the U.S. Navy use AWS services to create a biometric identification system that satisfies stringent security requirements, complex regulations, and mounting privacy concerns in an operationally efficient system for naval applications. GBL developed FutureOrb-ID, which takes advantage of the AI/ML capabilities of Amazon Rekognition and other AWS serverless and highly secure and compliant technologies.

The U.S. Navy relies on a Common Access Card (CAC) with PIN to authenticate an individual’s access to restricted buildings, ships, and other controlled spaces and systems. Although CAC systems work well in office environments, they are vulnerable to card loss, damage, or theft and may not be optimal in certain naval operational environments.

The Navy is collaborating with GBL Systems Corporation, an AWS Partner and leading Department of Defense (DoD) technology company, to explore the use of sophisticated biometric technology that could improve the authentication process. “We’re interested in new capabilities that provide the robust security that we need as well as support the actual operational mission that our sailors are trying to accomplish,” says Alan Jaeger, Naval PHE, manager of research and technology applications. “We chose GBL for this development because of their expertise in areas relevant to our mission set and the way they can help accelerate bringing these capabilities to our fleet.”

Using facial recognition and cloud-based technologies that can deliver services all the way to the tactical edge, GBL has developed an innovative solution called FutureOrb-ID. This patented technology creates a two-sided marketplace where organizations can access biometric data for authentication and individuals can opt in or out of having their biometric data used.

GBL chose to build this new application on Amazon Web Services (AWS) to leverage its global footprint that includes one-third of the world’s cloud infrastructure. “With that comes economies of scale,” explains Michael Soltys, GBL’s director of cloud strategy and services. “We can deploy sophisticated solutions that are affordable in the long term without making capital investments at the beginning.”

AWS was able to deliver the high performance, low latency, and industry-leading storage reliability across multiple Availability Zones that were crucial for this application. In addition, GBL benefited from AWS architectural design support that helped its team build a highly efficient solution with special attention given to security and regulatory compliance at every step.

Security is a primary concern when transmitting and storing such sensitive data. Jim Buscemi, chief executive officer at GBL, says, “By reading all the white papers on security and the AWS approach to it, we were assured that our data and compute would be safe in AWS.”

In less than one year, FutureOrb-ID went from concept to prototype, which was shown at the 2022 FATHOMWERX Summit. At the demonstration, GBL used AWS DeepLens with artificial intelligence (AI) on-site and an AWS-powered backend to rapidly authenticate personnel who had opted in for using their facial biometrics. Within seconds, the system accurately recognized individuals as either approved for entrance (indicated by a green light), denied entrance (red light), or unknown to the system (yellow light).

GBL also demonstrated its FutureOrb-ID Ghosting (FOG) technology, which applies advanced AI and machine learning (ML) algorithms to images in order to mask biometric data. This enables naval personnel and others to share pictures online without the risk of biometrics being scraped for nefarious uses.

FutureOrb-ID relies on Amazon Rekognition, which automates image recognition in applications without the customer having to build ML models and infrastructure from scratch. This powerful service can analyze millions of images in seconds, augmenting human review tasks with AI, and is able to scale as needed.

The solution runs on AWS Lambda, for a serverless environment that eliminates the need for upfront capital expenditures and enables GBL engineers to focus on innovation. GBL also chose AWS Lambda for its scalability. “Whether we are testing it on 10 users, 10,000 users, or 10 million users, AWS Lambda will automatically scale without us having to write a scaling module,” says Soltys.

This innovative technology also takes advantage of Amazon DynamoDB, a key-value and document database that delivers single-digit millisecond performance at any scale. “Amazon DynamoDB is very good because it has global tables, so we can quickly scale the application to a global community,” says Soltys. “It’s also a way to store data that's unstructured, and unstructured data is growing at a much faster rate than structured data.”

In addition, FutureOrb-ID uses Amazon Simple Storage Service (Amazon S3) for object storage with robust controls and easy data management at any scale. Soltys adds, “It comes with tremendous security measures in place. All data at rest and in transit is AES-256 encrypted, so we feel that our data is safe in Amazon S3.”

The next iteration of FutureOrb-ID will be shown at the 2023 FATHOMWERX Summit as the U.S. Navy continues to explore its use in land-based and shipboard applications for more robust authentication.

Moving forward, GBL aims to deploy FutureOrb-ID to businesses through a software as a service (SaaS) capability. This would enable businesses to offload the burden of processing identity and biometric information to FutureOrb-ID, which is compliant with international, federal, and state regulations.

“We’re doing pretty advanced research and stretching the art of what’s possible using cloud infrastructure,” says Buscemi. “In a very short time, we have been able to build this end-to-end, highly secure system that can scale quickly across both commercial and Department of Defense applications.”