Distributed Coded Computing for Content Management at the Tactical Edge

As more and more handheld devices and embedded systems are being deployed to tactical environments, there is a dire need for the ability to efficiently process and store large amounts of data in disconnected, intermittent and low-bandwidth (DIL) environments.  While the compute power of these devices has seen major improvements over the past years, they still lack the ability to rapidly execute modern artificial intelligence (AI) algorithms and applications, particularly given the large amounts of data and data types available.  To address this gap RAM Laboratories is proposing the Distributed Artificial Intelligence with Content Oriented Networking (DAICON) solution, combining state-of-the-art distributed coded computing algorithms with a future Information Centric Networking (ICN) architecture.  These distributed coded computing algorithms enable computationally onerous AI operations to be distributed across available compute resources so that tasks can be completed faster and more efficiently.  In addition, DAICON builds off our successful Opportunistic Named Data Networking (NDN) baseline prototype developed as part of the DARPA Secure Handhelds on Assured Resilient networks at the tactical Edge (SHARE) program that was successfully demonstrated to support the ability for devices to securely communicate in peer-to-peer DIL environments.