Israeli fabless semiconductor company Polyn Technology has demonstrated a proof of technology for its Neuromorphic Analogue Signal Processor, or NASP. Based on Polyn’s brain-mimicking architecture and containing several neural networks, the chip is the first Tiny AI “true analogue” design to be used with sensors. Tiny AI is localised AI, which promises to speed up processes such as mobile-based medical-image analysis or self-driving cars reaction times, as close to real time as possible.
“Our chip represents the most advanced technology bridging analogue computations and the digital core. It is designed with neuroscience in mind, replicating pre-processing the primary cortical area of the human brain does at the periphery before learning at the centre,” said Aleksandr Timofeev, CEO and founder of Polyn Technology.
The chip is implemented in 55nm CMOS technology, and the proof of technology also confirms the chip’s scaleability as well as the efficiency of its design tools. Polyn’s Neural-Net-To-Chip automation tools support fast and cost-effective development of tailored Tiny AI solutions, which perform AI computations on-device.
“Our first chip is created from trained neural networks by NASP compiler and synthesis tools that generated netlist and the silicon engineering files from the software math model simulation. We will continue to refine our technology for creation of new generation chips,” said Yaakov Milstain, COO of POLYN.
The chip will be available to customers early next year, followed by the first wearable product that will fuse photoplethysmography (PPG) sensors – essential in monitoring cardiac functions, and inertial motion unit (IMU) sensors, which blend sensors such as gyroscopes, magnetometers and accelerometersto calculate the exact force, angular rate and direction of a body.