NXP Semiconductors has released a new portfolio of MCX microcontrollers that enable developers to maximize software reuse in the portfolio to accelerate development. The portfolio also instantiates for the first time its new, specialized neural processing unit for accelerating inference at the edge, delivering 30x faster machine learning throughput than CPU cores alone. Designed to drive innovation in smart homes, smart factories, smart cities, and emerging industrial and IoT edge applications. The product portfolio includes four families of devices built on a common platform and supported by the widely adopted MCUXpresso development tools and software suite.
The MCU has come a long way in the past few decades, driven by the proliferation of edge devices. While they are at the heart of many edge applications today, the next generation of smart applications will require a new class of MCUS that will enable developers to navigate complex power, performance and security requirements, and connectivity options while balancing total system cost and energy efficiency. The portfolio, built on a common foundation of core technologies and supported by a unified software suite for maximum software reuse, will provide the flexibility needed to meet this challenge. The breadth of the portfolio enables developers to choose the device that best fits their application needs, thus enabling them to invest in the differentiated aspects of their design.
Ron Martino, executive vice president and general manager of edge processing at NXP Semiconductors, said: "As we approach the milestone of 75 billion connected devices, we are entering a new era of edge computing that requires us to fundamentally rethink how best to build A flexible MCU portfolio that is scalable, optimized, and can be the foundation for energy-efficient industrial and IoT edge applications today and for decades to come." "By building on our strong heritage in MCUs, this new portfolio will provide the required performance and integration to address the next wave of innovative real-time workloads.”
The portfolio is based on the high-performance Arm Cortex-M kernel. It includes a comprehensive set of peripherals to improve design flexibility. And the combination of the four series of development is to simplify the migration, and according to the need to expand or shrink, maximize the reuse of software, to maximize the lower development costs the equipment has up to 4 MB of flash memory on chip, low-power cache and advanced memory management controller, and up to 1 MB of internal SRAM, in order to improve the real-time performance of the edge of the application.
The company's eIQ ML software development environment will support machine learning and runtime reasoning. EIQ provides easy-to-use tools for developers to train ML models for NPU or CPU cores and deploy them on MCU. Families built to follow its secure design approach will provide secure startup with immutable trust roots, hardware-accelerated encryption, and built-in EdgeLock security subsystems on specific families.
