| Product Specification

NXP Semiconductors i.MX 93 Evaluation Boards for Power-Efficient Applications Processors

An advanced evaluation board for cost-efficient machine-learning application for Linux-based systems.

Specifications

Product TypeEvaluation Board
BrandNXP Semiconductors
Processors1-2x Arm Cortex A-55 1.7GHz, Arm Ethos U-65 microNPU, Arm Cortex-M33 250Mhz. EdgeLock secure enclave
ApplicationsAutomotive, Smart cities, Industrial and Smart homes
Camera Interface8-bit parallel YUV,RGB and 1 x 1080p60
ConnectivityEthernet, USB, UART, USART< and Flex IO Interfaces
Supported OSLinux, Greenhills, FreeRTOS, QNX, VxWorks

Overview

It evaluates the i.MX93 processors to deliver power efficiency performance for high-end machine learning operations on Linux-based systems. Thanks to an Arm Cortex A-55 core, heat protection and accelerated machine learning performance is ensured. Multicore processing, extensive connectivity features, compatibility with multiple operating systems, and external peripheral support for camera and audio devices make it a versatile evaluation board for various research and industrial applications.

Accelerated Machine Learning with Advanced Performance Handling Features

The i.MX 93 Evaluation board features multicore processing capabilities. It has an integrated EdgeLock secure enclave that ensures advanced security, making it an ideal option for sensitive industrial applications. The board also features an Arm Ethos - U65 microNPU to accelerate machine learning applications at a higher efficiency. Moreover, the Arm Cortex-A55 core provides seamless scalability allowing developers to expand the testing and evaluation range at a large scale. 

Thanks to a multicore architecture, the I.MX93 evaluation board integrates multiple computing cores in a single chip. As a result, it can run multiple tasks simultaneously without compromising on speed and efficiency. That’s why it’s a suitable option for machine learning applications. 

With faster data processing, the evaluation board enables developers to handle extensive and complex ML-based and graphical applications. 2D GPU composition and blending support with resizing and color space conversion features exist. 

Moreover, the board can work at an optimal temperature range of -40°to 125°degrees Celsius. Hence, it doesn’t wear out in complex and graphic-heavy applications. Its flexible compatibility features enable it to work with Linux, Greenhills, QNX, VxWorks, and FreeRTOS operating systems. 

Seamless Connectivity with External Peripherals for Extensive Applications

The i.MX 93 is a powerful evaluation board due to its power handling and temperature-efficient capabilities. Hence, it offers various connectivity options to onboard development boards and external IO devices. The board features 2 x 32-pin Flex IO interfaces that allow the user to connect various display and input devices simultaneously. For instance, there are two camera interfaces. First, there is a 1x 1080p60 featuring MIPI-CSI (2-lane) with a physical layer having a capacity of 1.5Gbps per lane to allow high-speed data transmission with cameras or external image sensors. The second camera interface features an 8-bit parallel YUV/RGB camera connectivity. 

The display module comprises an 18-bit parallel RGB interface, a 1080p60 MIPI-DSI (4 lanes) physical layer with a 1.5Gbps per lane capacity. Also, there is 720p60 LVDS (4-lane) display connectivity for high-definition video transmission. 

Further connectivity modules feature two USB 2.0 type C connectors, two Gb Ethernet AVB connectors, two CAN-FD connectors for high-efficiency communication in Automotive and industrial equipment, I2C, UART, and USART communication protocols. It also provides connectors for the Serial IO interface with external development boards and sensor packages. 

There is support for external memory devices, including SPI NOR and SPI NAND memories. The board can also process audio signals via 8-channel PDM microphone input, I2S TDM input, and an MQS output with a sigma-delta modulator.

With a wide range of connectivity features, onboard modules, and advanced power handling capabilities, the I.MX93 evaluation can be a cost-efficient and reliable option for automotive, Smart Homes, and Industrial applications.

References

Mouser Electronics [Internet]

Tags

embedded machine learningembedded systemslow power designmachine learning

Continue Reading

From Single Models to Full MLOps, Edge Impulse Is The New Instruction Engine For A Smarter Industrial Sector.

Embedded Machine Learning Is Giving Industrial Machines The Brains They Always Wanted.

The CyberSpec framework is designed to detect anomalous behavior linked to cyber-attacks against crowd-sensing spectrum sensors, even when said sensors are running on lightweight resource-constrained hardware like a Raspberry Pi.

CyberSpec turns machine learning onto the problem of spectrum sensor attacks, without overloading the host

ABI Research forecasts that TinyML unit volumes will explode from 15 million units in 2020 to 2.5 billion units in 2030. Find out about this machine-learning technique now.

TinyML: What is It and Why Does It Matter?