STMicroelectronics is set to release STM32N6, its first microcontroller integrated with a full neural processing unit (NPU).
Unlike other STM32 MCU, the STM32N6 includes a proprietary NPU and an ARM Cortex core. It also offers the same AI performance as a quad-core processor equipped with an AI accelerator but at one-twelfth the power consumption and one-tenth the cost.
As a leading developer of ARM microcontroller cores, ST has used the ARM M7 core in the dual-core M4/M7 STM32H7 series and the ART AI accelerator in the STM32F7 series.
There is no disclosing the ARM core it will use in the newly launched MCU. However, based on the perspective of performance and function data, it is more likely to use the ARM M55, even the latest M85. The ARM Cortex M85 is the highest performance M core, issuing up to three instructions per cycle with internal accelerators for AI that help to boost the performance.
ARM Cortex M85
NPU is a kind of processor specially applied to network application data packets. It adopts the architecture of "data-driven parallel computing" and can be used to process massive multimedia data such as video and images.
For example, in mobile phone applications, NPU is a professional chip constructed by simulating biological nerves. It is born for deep learning and enables mobile phones to obtain stronger computing and processing capabilities in the field of artificial intelligence. Its miniaturization and low power consumption are also ideal for UAV applications, which can improve take-off and endurance.
ST’s First NPU integrated STM32 MCU
The choice of architecture is key to adding security support and integrating with cloud services, the M55 and M85 support the ARMv8.1-M architecture. Yesterday, STMicroelectronics announced deals with Microsoft and Amazon to connect microcontrollers to the cloud.
ST is integrating its 160MHz ARM Cortex-M33 core-based STM32U5 microcontroller (MCU) with Microsoft Azure real-time operating system (RTOS) and IoT middleware and certified ARM Trusted Firmware-M (TF-M) security implementation Embedded System Services.
The integration uses the hardened security features of the STM32U5, complemented by the hardened key storage of the STSAFE-A110 secure element.
The integration with Amazon Web Services (AWS) also uses the STM32U5, this time using Amazon's FreeRTOS real-time operating system and ARM Trusted Firmware (TF-M) for embedded systems.
The reference implementation is based on ST's B-U585I-IOT02A IoT Node Discovery Kit and supports USB, WiFi, and Bluetooth Low Energy connectivity and multiple sensors. The STSAFE-A110 secure element support comes preloaded with IoT object credentials to help secure and simplify the connection between connected objects and the AWS cloud.
FreeRTOS includes a kernel optimized for resource-constrained embedded systems and software libraries for connecting various types of IoT endpoints to the AWS cloud or other edge devices. AWS's Long Term Support (LTS) for two years on FreeRTOS releases provides developers with a stable platform to deploy and maintain their IoT devices.
Hardware cryptographic accelerators, secure firmware installation and updates, and increased resistance to physical attacks provide PSA Certified Level-3 and SESIP 3 certification.
ST will release the STM32Cube-based Azure and AWS Integration Reference Implementation Integration in Q3 2022, leveraging tight integration with the wider STM32 ecosystem to further simplify IoT device design.
Supporting AI in industrial and embedded designs
“The new STM32N6 Neural MCU is dramatically lowering the AI technology implementation price point. This breakthrough supports our roadmap of new generation intelligent sensors allowing rapidly growing adoption in Smart Cities,” said Vincent SABOT, Executive Managing Director of developer Lacroix.
Driven by supporting AI in industrial and embedded designs, ST acquired Frech software developer Cartesian, which is specialize in the development of AI tools and enables ARM-based microcontrollers to have machine learning and reasoning capabilities.
In addition, ST aims to hit the $20 billion revenue ambition by 2027 at the latest. The total sales of STMicroelectronics in 2021 is $12.76 billion. In 2022, sales are expected to be between $14.8 billion and $15.3 billion.
In the severe economic environment in recent years, the global core shortage has severely hit the automobile and other manufacturing industries. Most chip manufacturing plants are currently operating at full capacity. STMicroelectronics is no exception, its customers include iPhone maker Apple and electric vehicle leader Tesla, so it has had to invest heavily to expand production capacity. This also means that ST may achieve new sales and gross margin targets higher.