Scalable Compute Platform for Sensor Processing
Ian Land, Michael Parker · 2024
High-performance DoD applications like sensor arrays must continue to push the state-of-the-art (SOTA) size, weight, and power (SWaP) while advancing system performance processing of the individual sensors. A scalable Vector Processor (VP) based compute platform can be created to deliver ASIC-like performance/power while incorporating FPGA-like flexibility and latency. This paper presents a study of key processors and assesses the strengths and challenges where a vector-engine based platform can deliver capabilities beyond the traditional methods that leverage FPGAs, while adding in AI acceleration and consideration of multiple sensors, sensor fusion, and strategic applications.