Numem Selected by NASA for 'DNN Radiation Hardened Co-Processor Companion Chip to NASA's Upcoming High-Performance Spaceflight Computing Processor'
A New Spin on Embedded Memory
SUNNYVALE, Calif., October 8, 2020 (Newswire.com) - Numem, the leading provider of NuRAM & SmartMem Smart Compute Memory IP Cores based on MRAM, has been selected for Phase I of the “DNN Radiation Hardened Co-processor as companion chip to NASA’s upcoming High-Performance Spaceflight Computing Processor” to enable AI processing.
This AI core developed with Numem’s Partners comprises of reconfigurable DNN Engine with multiple compute units which can support a wide range of DNN models and frame rates and a Numem NuRAM MRAM-based Memory enabling a 2-3x smaller memory area and 20x to 50x lower standby power than SRAM for low-power, highly efficient processing.
The on-board AI processing will enable spacecraft to efficiently process large volumes of raw sensor-data into scientific knowledge or actionable data to overcome limitations in downlink communication. It will also enable spacecraft to formulate decisions for critical operations. It will solidify HSPC avionics ecosystem with robust AI capabilities for space autonomy, small satellite constellations/autonomous science, human explorations and operations Habitat and deep space missions.
Numem ’s NuRAM and Smart Compute Memory (SmartMem) IP Cores are 2-3x smaller area and >20x lower leakage current than SRAM with the added advantage of being non-volatile and able to retain data in case of power failure or for low power operation.
Numem NuRAM and SmartMem provide an excellent memory solution for IoT Wearables, Microcontroller, AI /Neuromorphic, Storage, Autonomous Vehicle, Digital Security Cameras and in general SoC Designs requiring ultra-low power or high-density memories, in addition to Space and Defense applications requiring radiation hardened AI microelectronics. They can also be used as lower power/higher endurance/higher performance eFlash, or OTP for permanent data such as PUF/Root of Trust for Security Applications.
Source: Numem Inc.
Categories: Embedded Computing