Infineon Technologies AG, a leading semiconductor manufacturer, has recently launched the PSoC Edge microcontroller, an innovative solution designed to revolutionize the landscape of hardware-accelerated machine learning (ML) for various applications within the Internet of Things (IoT) and industrial sectors. The PSoC Edge microcontroller aims to significantly lower the barriers in human-machine interaction and add contextual awareness to end applications, thereby propelling the industry into a new era of responsive computing and control applications.
Enhanced compute performance and localized ML processing
At the core of this groundbreaking technology is the integration of advanced ML capabilities within the high-performance Arm Cortex-M55 processor, along with Arm’s Helium technology, which facilitates enhanced digital signal processing (DSP) and ML capabilities. This combination enables the efficient acceleration of neural network processing, catering to the ever-increasing demands of complex ML applications in resource-constrained IoT devices.
In addition to the Cortex-M55, Infineon’s PSoC Edge microcontrollers feature the integration of the Arm Ethos-U55, a dedicated neural processing unit (NPU) meticulously crafted to expedite ML inference in embedded and IoT devices operating within limited spatial confines. Complementing this, the Cortex-M33 is paired with Infineon’s proprietary NNLite hardware accelerator, known for its ultra-low power consumption, intended to expedite neural network processing in ML applications further. This powerful blend of scalable compute power, memory, and efficient processing capabilities sets the stage for a new era of seamless and robust ML implementation.
Comprehensive software and tool ecosystem
Recognizing the critical role of comprehensive software support in enabling effective utilization of ML-enabled MCUs, Infineon has incorporated the full ML tool suite from Imagimob, a Swedish startup acquired by Infineon earlier this year, into its Modus Toolbox software platform. This integration empowers embedded system developers with a comprehensive suite of development tools, libraries, and embedded runtime assets, facilitating seamless and efficient development and deployment of ML-based applications.
The PSoC Edge microcontrollers boast ample on-chip memory, including non-volatile resistive random-access memory (RRAM), alongside high-speed, secured external memory support, ensuring efficient data storage and retrieval for ML applications. Infineon’s commitment to providing end-to-end support for ML development, from initial data entry to the deployment of intricate models, further reinforces the company’s dedication to fostering a robust and user-friendly ecosystem for developers.
Industry implications and future prospects
Steve Tateosian, senior vice president of microcontrollers at Infineon, hailed the PSoC Edge microcontroller as a game-changer in the realm of compute performance and ML acceleration. With the ability to execute advanced ML tasks locally on the device, the PSoC Edge microcontroller eliminates the dependency on cloud-based solutions for resource-intensive applications, thereby enhancing data privacy and security. Infineon’s strategic amalgamation of cutting-edge hardware and comprehensive software support not only streamlines the development process for engineers but also opens new doors for the seamless integration of AI-driven functionalities across various IoT and industrial applications.
As Infineon continues to push the boundaries of innovation with its PSoC Edge microcontroller, the industry eagerly anticipates the transformative impact of this advanced hardware-accelerated ML solution on the ever-evolving landscape of IoT and industrial technologies.
Infineon’s PSoC Edge microcontroller represents a significant leap forward in integrating hardware-accelerated ML capabilities, promising a host of transformative possibilities for developers and stakeholders across diverse industries. With its robust computing power, optimized memory, and comprehensive software support, the PSoC Edge microcontroller stands poised to redefine the paradigm of responsive computing and control applications in the IoT and industrial sectors.
