i.MX 8M Plus Features Dedicated Neural-Net Processing Unit

Jan. 7, 2020

MCUs and SoCs can handle machine-learning chores, but they can do the job faster and more efficiently with hardware like that found on NXP’s i.MX 8M Plus.

William G. Wong

Related To: NXP Semiconductors

NXP Semiconductors added to its EdgeVerse portfolio with the i.MX 8M Plus application processor (see figure). It's the company's first i.MX EdgeVerse chip to include dedicate machine-learning (ML) hardware. The i.MX 8 family has been very popular and the addition of ML acceleration greatly expands the possible applications it can run with a dedicated Neural Processing Unit (NPU). It allows for advanced ML inference for the Internet of Things (IoT).

The i.MX 8M Plus includes an NPU that delivers 2.3 TOPS along with a 2 GHz quad-core Arm Cortex-A53 CPU.

The i.MX 8M Plus is built using 14-nm LPC FinFET process technology. The NPU delivers 2.3 TOPS. It's complemented by a 2-GHz quad-core Arm Cortex-A53 CPU. Real-time support is courtesy of an 800-MHz Cortex-M7 and a a high-performance 800-MHz Tensilica audio HiFi 4 DSP. The latter can handle voice and natural language processing chores and supports Immersiv3D audio with Dolby Atmos and DTS:X.

The chip also manages streaming video—a dual-camera image signal pocessor (ISP) and 3D GPU handle graphics rendering. The ISP can take on two high-definition cameras or a single 12-Mpixel camera. This includes support for high dynamic range (HDR) and fisheye lens correction. The system also includes H.265 or H.264 HD video encoder and decoders.

The NPU uses ML models to process streaming mutltimedia data. This allows the i.MX 8M Plus to manage chores like object recognition for public safety, industrial machine vision, robotics, hand gesture, and emotion detection with natural language processing—all without going to the cloud.

The i.MX 8M Plus can tackle ML chores like multi-object identification or speech recognition of over 40,000 English words. It's able to process 500 images/s using the Mobilenet ML interference model. The NPU can handle multiple ML models at the same time.

The chip also includes a Gigabit Ethernet port that supports time-sensitive networking (TSN). The internal and external memory systems include error-correction-code (ECC) support.

“The edge is the perfect destination to deploy machine-learning applications, especially as technology advancements are enabling accurate localized decision-making,” said Martyn Humphries, vice president and general manager of i.MX application processors for consumer and industrial markets at NXP. “NXP engages with leading companies in the consumer and industrial IoT marketplace, and we look forward with great excitement to the innovative products they will introduce based on the i.MX 8M Plus. The i.MX 8M Plus is a significant milestone and a trendsetter for processing solutions that are capable of transforming the ‘smart’ edge to an ‘intelligent’ edge.”

The family will be available in versions within the industrial temperature range of –40 to 105ºC. There will also be versions with a 15-year longevity commitment.

About the Author

William G. Wong | Senior Content Director - Electronic Design and Microwaves & RF

I am Editor of Electronic Design focusing on embedded, software, and systems. As Senior Content Director, I also manage Microwaves & RF and I work with a great team of editors to provide engineers, programmers, developers and technical managers with interesting and useful articles and videos on a regular basis. Check out our free newsletters to see the latest content.

You can send press releases for new products for possible coverage on the website. I am also interested in receiving contributed articles for publishing on our website. Use our template and send to me along with a signed release form.

Check out my blog, AltEmbedded on Electronic Design, as well as his latest articles on this site that are listed below.

You can visit my social media via these links:

I earned a Bachelor of Electrical Engineering at the Georgia Institute of Technology and a Masters in Computer Science from Rutgers University. I still do a bit of programming using everything from C and C++ to Rust and Ada/SPARK. I do a bit of PHP programming for Drupal websites. I have posted a few Drupal modules.

I still get a hand on software and electronic hardware. Some of this can be found on our Kit Close-Up video series. You can also see me on many of our TechXchange Talk videos. I am interested in a range of projects from robotics to artificial intelligence.