Featured in ASSCC 2025

A 28 nm AI microcontroller with tightly coupled zero standby power weight memory featuring standard logic compatible 4 Mb 4-bits/cell embedded flash technology

‍

Microcontrollers targeted for battery powered smart edge devices are often required to run inferencing tasks at a place where sensor data are generated for real-time response using a locally stored AI model once it is trained from the cloud. Power gating technique is often deployed to save an idle mode power consumption of the microcontrollers for low power applications, and the AI model can be stored and updated in an embedded Non-Volatile Memory (eNVM) during a lifetime of the device without consuming standby power during the idle mode. Typically, such a multiple-time programmable eNVM technology requires additional fabrication steps beyond a standard logic process and is configured to store only single bit information per a unit memory cell limiting efficiency of AI computation [1]. In this work, we report an AI microcontroller with zero-standby power weight memory featuring standard logic compatible 4-bits/cell Embedded FLASH (EFLASH) technology, tightly coupled to a Near-Memory Computing Unit (NMCU) for cost effective and low power edge AI computing applications.

Source:

https://ieeexplore.ieee.org/document/11349555