There have been some advancements in embedded technology despite the global pandemic, which has hindered chip production in everything from smartphones to video cards. Despite the shortage, manufacturers have managed to release new SBCs, SoCs, AIoMs, and other embedded systems. Most have been designed to take advantage of machine learning, Edge computing, and other AI applications, which is driving innovation in nearly every industry. In this roundup, we will take a look at several of the latest embedded technologies currently available in 2021.
AI-On-on-Module专为在边缘的工业应用而设计
adlink技术has launched its new LEC-IMX8MP SMARC AI-on-Module designed for industrial AI at the Edge applications. The new module is designed around NXP's i.MX 8M Plus SoC, which packs an Arm Cortex-A53 processor and integrated neural processing unit (NPU), providing up to 2.3 Terra Operations Per Second (TOPS). It also offers VDS/DSI/HDMI graphic output, dual CAN bus/USB 2.0/USB 3.0, dual GbE ports (one with TSN), and an I2S audio interface. The LEC-IMX8MP features BSP support for Debian, Yocto and Android, including MRAA hardware abstraction layer (HAL), and can operate in temperatures ranging from -40°C to +85°C.
新SOM为商用车和移动系统带来了边缘计算
MicroSys Electronics.已推出其新的MPXLX2160A，这是一种专为商用车辆和移动系统设计的坚固型Edge SOM（模块上的系统）。SOM围绕NXP的Qoriq Layerscape LX2160A处理器设计，ARM Cortex-A72和高达128 GB的DDR4 RAM。董事会还通过PCIe Gen3 X8，100GBE / 40GBE / 25GBE / 10GBE和SATA包装24x Serdes Lanes。附加功能包括2x RGMII，2x USB 3.0，2x CAN-FD，UART，I2C和Flex SPI。MPX-LX2160A可以处理高达256 GB的EMMC存储，外接SD卡，温度传感器，RTC，冷却控制器和JTAG调试接口的接口。
新的5G Front Haul Solution Designed for Increased Wireless Throughput
Semtech已宣布采样其新的GN2255双向50Gbps PAM4 Trie Edge CDR现在可用。GN2255采用集成的DML激光驱动器，可启用SFP56 50 Gbps PAM4光学模块，专为5G前运输部署而设计。三边IC具有超低延迟，每个模块小于1.5W，可提供高达10km的性能，达到50 Gbps PAM4 SFP56光学模块。IC的工业工作温度范围为-400℃至1000℃，正在提供具有完整模块级固件的可选参考设计套件（RDK），以快速集成。
AI Vision Board可以帮助工程师加速市场
ETA Computerecently announced its new integrated ECM3532 AI Vision Board designed to help developers accelerate time to market and quickly deploy machine-vision applications. The board is outfitted with an ECM3532 TENSAI SoC with an Arm Cortex-M3 microcontroller and a CoolFlux DSP to boost machine-learning operations. The board is also outfitted with several sensors (ambient light, microphone, accelerometer/gyroscope), a low-power Himax HM0360 camera, and an expansion connector. The ECM3532 has a CR2032 battery socket and uses low-power IoT/Bluetooth Low Energy connectivity, allowing the board to run for months before needing a recharge or battery replacement.
AI辅助车辆计算机升级以增加性能
syslogic.recently took advantage of Nvidia's Jetson Xavier NX module and incorporated it into its RS A3N AI vehicle computer, making it ideal for computer vision, inferencing, or sensor fusion applications, even in adverse conditions. With the Jetson Xavier NX module integrated into the RS A3N, the computer now features a 6-core Nvidia Carmel ARM v8.2 64-bit CPU with 384 CUDA cores and 48 Tensor cores, capable of pumping out 21 TOPS. It also includes 8 Gb of LPDDR4x RAM, a 7-Way VLIW vision processor, and a pair of NVDLA Engines. The computer comes equipped with a pair of Ethernet ports, CAN, USB 3.1 ports, and one HDMI interface. Two USB ports and a DisplayPort interface are also available behind the service cover and an M.2 slot (PCIe) to expand the flash memory from 125 GB to 2 TB. Connectivity options include LTE, GNSS, and Wi-Fi, and Syslogic is expected to expand the interface layout with PoE (Power over Ethernet) and digital I/Os in later revisions.
