NXP's new i.MX RT700 crossover MCU is equipped with eIQ Neutron NPU, enabling AI edge with high performance and low power consumption

Publisher:EE小广播Latest update time:2024-09-24 Source: EEWORLDKeywords:NXP Reading articles on mobile phones Scan QR code
Read articles on your mobile phone anytime, anywhere

New highly integrated i.MX RT700 crossover MCU is designed to deliver significant power savings and features an eIQ Neutron neural processing unit (NPU) to deliver up to 172x AI acceleration at the edge


image.png


Shanghai, China - September 24, 2024 - NXP Semiconductors NV (NXP Semiconductors NV) today announced the launch of the new i.MX RT700 cross-border MCU series, designed to empower edge devices that support intelligent AI, such as wearable devices, consumer medical devices, smart home devices and HMI platforms . The i.MX RT700 series provides an optimized combination of high performance, extensive integration, advanced features and energy efficiency for the new era of edge AI computing.


The i.MX RT700 is equipped with up to five powerful cores in a single device, including the first integrated eIQ® Neutron NPU in a crossover MCU, which can accelerate the processing of AI-related applications by up to 172 times while reducing energy consumption per inference by up to 119 times. The i.MX RT700 crossover MCU also integrates up to 7.5MB of ultra-low-power SRAM, reducing power consumption by 30-70% compared to previous generations.


Significance


There is a growing demand for additional computing power and new features in AI edge computing devices. At the same time, many of these devices are battery-powered, requiring MCUs with lower power consumption.


The i.MX RT700 crossover MCU meets this need with a low-power, multi-core design that integrates powerful graphics capabilities, AI hardware acceleration, advanced security, and perceptual computing subsystems. This enables customers to create a full range of solutions with multi-modal features such as presence sensing, gesture recognition, voice control, etc., all on a unified platform.


“As the pioneer of crossover MCUs, we are not only driving product advancements with the i.MX RT700, we are redefining what’s possible in edge computing,” said Charles Dachs, senior vice president and general manager of Industrial and IoT at NXP. “The i.MX RT700 delivers breakthrough efficiency improvements while significantly reducing power consumption, extending battery life and ensuring reliability for resource-constrained applications. In addition, the integrated eIQ Neutron NPU enables customers to build innovative machine learning applications and improve AI and multitasking capabilities on low-power edge devices.”


More details


The i.MX RT700 crossover MCU family features up to five computing cores. This includes a main Arm® Cortex®-M33 core running at 325 MHz, and an integrated Cadence® Tensilica® HiFi 4 DSP for more demanding DSP and audio processing tasks. It also features NXP's eIQ Neutron NPU and supports the eIQ machine learning software development environment. The series can be extended to an industry-leading 7.5MB of ultra-low power SRAM without waiting for state access. The i.MX RT700 also includes an ultra-low power perception computing subsystem that includes a second Cortex-M33 core and an integrated Cadence Tensilica HiFi 1 DSP. This eliminates the need for an external sensor hub, reducing system design complexity, space usage, and BOM cost.


  • Accelerating AI with eIQ Neutron


To unleash the potential of AI on small embedded controllers like MCUs, the eIQ Neutron NPU effectively accelerates AI inference by offloading the workload of the Cortex-M33. Compared to running these tasks on the Cortex-M33 alone, the eIQ Neutron NPU is able to complete tasks faster, such as 18 times faster anomaly detection or 172 times faster image classification (depending on the model). With 7.5MB of onboard SRAM, the i.MX RT700 can perform more complex multi-mode AI tasks that are challenging for traditional microcontrollers.


  • Low power consumption, long battery life


Power consumption is an important consideration for edge devices. The i.MX RT700 series is designed to optimize power consumption, reducing power consumption by 30-70% compared to previous generations. This is achieved through a series of power architecture innovations, such as advanced adaptive deep sleep technology, optimized clock architecture with low-power caching schemes, advanced wake-up/sleep cycles, etc. As a result, end devices can extend battery life or use smaller batteries, achieving greater design flexibility.


  • Highly integrated MCU to optimize coding


In addition, the i.MX RT700 series also enhances system performance and flexibility through integrated high-efficiency DC-DC converters and basic memory management units, and supports improved analog peripherals. It is also NXP's first crossover MCU to support the emerging eUSB standard, enabling USB 2.0 interfaces to operate at 1V or 1.2V I/O voltages instead of 3.3V.


  • Core security for edge devices


Today, cybersecurity and privacy for smart, connected consumer devices are more important than ever, especially with upcoming cybersecurity regulations such as the US Trustmark and the European Cyber ​​Resilience Act. The i.MX RT700 series provides a path to comply with these regulations as well as other security standards for consumer devices, including ETSI 303 645.


Network resilience and consumer data protection are at the core of the i.MX RT700 series. This new crossover MCU integrates EdgeLock Secure Enclave (Core Profile), which provides advanced security features for smart devices, including secure boot with power saving mode, secure updates, seamless memory encryption, and secure data access. It also has strong device authentication capabilities with built-in physical unclonable function (PuF).


  • Product Availability


The i.MX RT700 crossover MCU family is currently sampling to qualified early access customers.


Keywords:NXP Reference address:NXP's new i.MX RT700 crossover MCU is equipped with eIQ Neutron NPU, enabling AI edge with high performance and low power consumption

Previous article:Intel confirms it will launch a customized version of Xeon 6 processors for Amazon AWS
Next article:IAR fully supports Guoke Huanyu AS32X series RISC-V automotive MCU

Latest Embedded Articles
Change More Related Popular Components

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号