This combination of software and tools allows developers to reuse a large portfolio of available software, thus accelerating development. This range also incorporates, for the first time, NXP's new dedicated neural processing unit (NPU) to deliver improved performance in edge computing applications, processing machine learning algorithm data 30 times faster than a single CPU core.
Driven by the proliferation of smart devices, the microcontroller (MCU) has undergone a dramatic evolution over the past few decades. While microcontrollers are at the heart of smart applications, the next generation of applications will demand a new class of microcontrollers that allows developers to navigate a complex environment defined by power, performance, and security requirements, as well as connectivity options, while balancing total system cost and energy efficiency. The MCX range, built around a common core technology and supported by a unified software portfolio for maximum utilization, will provide the flexibility required to meet this challenge. MCX's wide range allows developers to select the devices that best suit their application's needs and gives them the freedom to invest in differentiating aspects of their design.
“We are approaching a new starting point, with 75 billion connected devices, entering a new era of Edge Computing that demands we rethink how best to design a microcontroller architecture that is flexible, scalable, optimized, and serves as the foundation for low-power Industrial and IoT Edge Computing applications, both now and in the coming decades,” said Ron Martino, Executive Vice President and General Manager of Edge Processing at NXP Semiconductors. “Building on our strong microcontroller legacy, this new portfolio will deliver the levels of performance and integration needed to handle real-time workloads in the new wave of innovation.”.
The four series that make up the MCX range are designed for ease of use, as well as simplified migration and scalability, to maximize software performance and minimize development costs. The range is based on high-performance Arm Cortex-M cores and integrates a wide range of peripherals for greater design flexibility. MCX devices feature up to 4MB of flash memory, low-power cache, and advanced memory management controllers, as well as up to 1MB of SRAM to further enhance real-time performance for applications requiring edge processing.
Machine learning and runtime inference will be supported by NXP's eIQ® ML software development environment. Developers can use eIQ's user-friendly tools to train machine learning models specifically designed for the NPU or CPU and then deploy them on the microcontroller. The MCX families are built on NXP's "security-by-design" approach, offering secure and immutable boot, hardware-accelerated cryptography, and, in some families, the inclusion of an EdgeLock® secure subsystem.
About the MCX range:
• The MCX N Advanced series is designed for secure and intelligent applications, with families that integrate the EdgeLock security module and a dedicated NPU with a high-efficiency computing architecture for real-time inference.
• The MCX A Essential series is optimized to provide functionality under critical circumstances in a wide variety of applications, such as motor control, that demand limited cost and advanced analog capabilities such as high-precision data converters, as well as a fast time to market.
• The MCX W Wireless series offers low-power, narrowband connectivity, including Bluetooth Low Energy. It is designed to simplify the addition of wireless connectivity to IoT devices, and its low-power radio helps extend battery life in small systems.
• The MCX L Ultra-Low Power series is designed for applications where power consumption is paramount. These devices, with some of the lowest static and dynamic power consumption on the market, will help significantly extend battery life compared to traditional microcontrollers.
