Edge AI MCUs Take on Complex Tasks While Drawing Minimal Power

08 November 2023

Intended for addressing artificial intelligence (AI) workloads relating to voice recognition and machine vision, the 32-bit Ensemble microcontroller units (MCUs) from Alif Semiconductor are highly optimised for edge-based implementation.

The company claims that they have more than 2 orders of magnitude better AI processing performance than conventional MCUs in this class can deliver. This is down to the proprietary architecture employed - where each of the Arm Cortex-M55 central processing unit (CPU) cores is complemented by an Arm Ethos-U55 microNPU neural processor. When carrying out object detection inference work, use of one of the microNPUs means that the time involved is 74x shorter than relying on the CPU core alone. In addition, the energy required for conducting such tasks is just 0.27mJ. Prospective use cases will include wearables, IoT, retail, communications infrastructure, etc.

“Early on we saw a huge gap in the market,” states Mark Rootz, Alif’s VP of Marketing. “There are many edge ML applications that require 50-250GOPs to become useful. Typical 32-bit MCUs can’t come close to that performance level, so to find a solution developers had to jump all the way to GPU-based accelerators at about the 1000GOPs level which is extreme overkill in terms of power, size, cost and complexity. Only Alif fills this gap in the middle. This is the sweet spot for battery-powered products on the edge.”


Contact Details and Archive...

Print this page | E-mail this page






This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.