NewsAlif Semiconductor Enhances Edge Computing with AI-Enabled Microcontrollers

Alif Semiconductor Enhances Edge Computing with AI-Enabled Microcontrollers

Category articles

Introduction of AI-Enabled Microcontrollers by Alif Semiconductor

Alif Semiconductor has announced the availability of its ‘Ensemble’ series, a new lineup of 32bit AI-enabled microcontrollers. These integrated circuits are engineered to address the computational demands of AI algorithms in battery-powered edge devices, featuring a combination of various Arm cores and enhanced security to safeguard intellectual property.

Meeting Edge Computing Demands

The Ensemble series aims to fill a critical performance gap in edge machine learning (ML) applications that require operational speeds ranging from 50 to 250 Giga operations per second (Gop/s). Traditional 32bit microcontrollers fall short in this aspect, whereas GPU-based accelerators provide excessive performance at higher power, size, cost, and complexity. Alif’s Ensemble series microcontrollers offer a balanced solution optimized for power-sensitive edge applications.

Ensemble Series Lineup

The Ensemble family of integrated circuits is divided into four distinct tiers, each with a unique configuration of Arm Cortex-M55 MCU, Cortex-A32 MPU, and Ethos-U55 neural processing units, catering to various application requirements.

  • E1 series: Single-core configuration with a focus on wearable technologies.
  • E3 series: Dual-core setup aimed at more complex edge applications.
  • E5 series: Triple-core, Linux-capable processors designed for advanced edge computing.
  • E7 series: Quad-core, Linux-capable processors for the most demanding edge applications.

Performance Benchmarks

Alif Semiconductor asserts that the M55 + U55 core combination significantly outperforms previous-generation Cortex-M microcontrollers in terms of speed and energy efficiency. This combination allows for rapid image classification inferences, highlighting its capability for intensive edge AI tasks.

Peripheral Integration and Security Features

With an emphasis on wearables, the E1 series is equipped with a range of peripherals including dual-lane MIPI-DSI interfaces and multiple audio inputs. Alif has placed a premium on security, implementing robust features to safeguard against intellectual property theft, malware attacks, and unauthorized data access.

Secure Enclave and Lifecycle Management

The Ensemble series incorporates a secure enclave that includes a root-of-trust, hardware cryptographic acceleration, and configurable firewalls. These features ensure security throughout the device’s lifecycle, from manufacture to deployment and eventual disposal.

Development Support and Application Kit

Alif provides an application kit designed to assist developers in assessing the AI algorithm performance of the Ensemble microcontrollers. This toolkit includes a range of sensors and camera inputs, allowing for the collection and use of live data to fine-tune AI applications.

Michal Pukala
Electronics and Telecommunications engineer with Electro-energetics Master degree graduation. Lightning designer experienced engineer. Currently working in IT industry.