IoT Archives - Page 399 of 456 - SmartCitiesElectronics.com

Axiomtek’s PoE embedded vision system delivers AI for IoT

Mon, Jul 01, 2019

For applications in machine vision, edge computing, traffic vision, deep learning, and artificial intelligence (AI) of things, the eBOX671-521-FL is a graphics processing unit (GPU) computing embedded system with four-channel PoE and a MXM 3.1 Type A slot.

The eBOX671-521-FL is powered by the 8th gen Intel Core i7, i5, i3 (Coffee Lake-S) processor with either up to six-core, Intel Pentium processor, or Intel Celeron processor and comes with Intel Q370 or optional Intel C246 chipset.

The fanless embedded GPU-based system is equipped with dual DDR4 error code correction (ECC) /non-ECC SO-DIMM slots for up to 64Gbyte of system memory. The embedded vision system has an optional MXM type A slot for NVIDIA GTX1030 and GTX1050 graphics modules to enhance the performance of the visual display.

The Intel Coffee Lake-based embedded box PC is designed with four Gigabit 802.3at (Power-over-Ethernet or PoE) -compliant Ethernet ports with a power budget of 60W, six USB 3.1 ports and two Gigabit LANs to connect different devices or sensors.

“Axiomtek’s eBOX671-521-FL is dedicated to AI inference, GPU computing, and machine learning and image processing,” explains Janney Lee, a product manager of Product PM Division at Axiomtek. In addition to the expansion options of an internal MXM 3.1 type A connector, two full-size PCI Express Mini Card slots and two SIM slots, there are two swappable SATA HDD drive bays with a height of up to 9.5 mm and one mSATA for RAID 0 and 1 feature, for storage. Lee continues: “We have designed a flexible I/O window slot for incorporating application-oriented I/O functions to meet the requirements of various cases. This embedded vision computer can be mounted using wall or DIN-rail mounting for versatile use in various environments.”

I/O interfaces include two RS-232/422/485, one DVI-I port, one HDMI port, one VGA port, and two DisplayPort outputs through an optional MXM graphics kit. There is also a Phoenix-type VDC power input connector, AT/ATX quick switch, ATX power switch, reset switch, and four SMA-type antenna openings. The four-channel Gigabit PoE embedded vision system supports a 24V DC power input with power protection.

The eBOX671-521-FL supports Windows 10 IoT and Linux operating systems and supports Axiomtek’s AXView 3.0 software for smart device monitoring and remote management in the industrial IoT world.

http://www.axiomtek.com

> Read More

Multi-chip package is set to advance 5G CPE modem roll out

Thu, Jun 27, 2019

Memory supplier, Winbond Electronics, supports accelerated roll-out of 5G customer premises equipment (CPE) modems as it releases a 2Gbit+2Gbit NAND+LPDDR4x multi-chip package.

The W71NW20KK1KW combines non-volatile flash/high-speed DRAM for cost and memory capacity in static 5G modem applications, says Winbond.

The 1.8V device is provided in a compact 8.0 x 9.5 x 0.8mm multi-chip package (MCP).

The W71NW20KK1KW combines robust single level cell (SLC) NAND flash and high-speed, low-power LPDDR4x memory. It provides sufficient memory capacity for 5G cellular modems that are intended for use as CPE in homes and offices.

Static 5G CPE modems can operate with memory capacities of 2Gbit NAND/2Gbit DRAM. This combination in a single package enables 5G modem manufacturers to meet the system requirements of CPE units at the lowest possible materials and production cost, says Winbond, to help accelerate consumer adoption of 5G as an alternative to fixed-line copper or optical xDSL links in the last mile of high-speed broadband networks.

According to Wilson Huang, product marketing manager at Winbond, Winbond is the only MCP chip manufacturer that produces both NAND and LPDDR4x in its own wafer fabrication plants to assure customers of supply quantities and schedules, quality and service.’

The W71NW20KK1KW is a 149-ball ball grid array (BGA) MCP consisting of a 2Gbit SLC NAND flash die and a 2Gbit LPDDR4x DRAM die. The robust SLC NAND flash offers endurance and high data integrity, adds Winbond. The SLC NAND only requires 4-bit error code correction (ECC) to achieve high data integrity, but the device’s 2kbyte+128byte page size provides enough space for the use of 8-bit ECC.

The W71NW20KK1KW has an 8-bit bus, and is organized in blocks of 64 pages. The NAND die’s performance specifications include a maximum page Read time of 25 micro seconds and a typical page program time of 250 micro seconds.

The LPDDR4x DRAM die, which operates at a high frequency of 1866MHz, provides an LVSTL_11 interface and features eight internal banks for concurrent operation. It offers a data rate of up to 4267M transfers per second, supporting the fast data-transfer rates to be offered by 5G cellular networks, according to Winbond.

The W71NW20KK1KW is available now in production volumes.

http://www.winbond.com

> Read More

Raspberry Pi 4 computer triples speed, says Farnell

Tue, Jun 25, 2019

Distributor Farnell has introduced the Raspberry Pi 4 Model B computer, describing it as the most powerful Raspberry Pi to date.

It is three times faster and offers greater memory capacity, with improved interface and connectivity, says Farnell.

The Raspberry Pi 4 Model B Computer offers significant enhancements in processor speed, multimedia performance, memory and connectivity for general desktop computer users, hobbyist and makers and professional developers working with compute-intensive embedded applications such as computer vision and artificial intelligence (AI).

“The Raspberry Pi 4 Model B Computer … delivers a real jump in capabilities and addresses the needs of the growing market for Raspberry Pi,” said Hari Kalyanarama, global head of single board computers at Farnell.

The Raspberry Pi 4 Model B Computer is the first model to use a 28nm SoC, delivering a significant increase in performance and energy efficiency, reports Farnell. It has a quad-core ARM Cortex-A72 64-bit processor clocked at 1.5GHz, which accounts for the three-fold increase in operating speed.

For video and sound there are two micro HDMI ports support dual-display output at resolutions up to 4K. For fast data transfer, there is a SuperSpeed USB 3.0 to deliver transfer rates to mass-storage devices (up to 5Gbits per second). There is true Gigabit Ethernet connectivity to deliver network data rates of up to 1Gbits per second.

Dual-band wireless networking at 2.4 and 5.0GHz delivers data rates in excess of 100Mbits per second. Modular compliance certification allows the board to be designed into end products with what Farnell descrbies as “significantly reduced compliance testing” to accelerate time to market.

Memory options are 1.0, 2.0 and 4Gbyte LPDDR4 and for multimedia applications the Raspberry Pi 4 Computer has H.265 decode (4kp60), H.264 decode (1080p60) and H.264 encode (1080p30), OpenGL ES 3.0 graphics and an image sensor pipeline.

There is a 40-pin general purpose I/O header with full backward compatibility and additional multiplexed UART, I2C and SPI peripherals. There is also a Micro SD card slot for loading operating system and data storage and Power over Ethernet (PoE) support, using a separate HAT accessory.

According to Raspberry Pi Trading Chief Executive Eben Upton, AI and machine learning (ML) data can be processed right on the board, reducing the need to send data to the cloud and addressing real-world concerns about privacy. The Raspberry Pi 4 Model B can be used as a desktop computer, with a powerful processor, multimedia acceleration and dual-display support.

The Raspberry Pi 4 Model B Computer is the first model to use a 28nm SoC, delivering a significant increase in performance and energy efficiency.

The Raspberry Pi 4 Model B Computer will remain in production until at least January 2026. Also available alongside the Raspberry Pi 4 Computer is a new case and a range of accessories including power supplies and micro HDMI cables.

The Raspberry Pi 4 Computer is available now from Farnell across EMEA and CPC in the UK, and Newark in North America.

http://www.element14.com

> Read More

Xilinx ships Versal ACAP devices

Wed, Jun 19, 2019

Xilinx has announced that it has shipped Versal AI Core series and Versal Prime series devices to multiple tier one customers through the company’s early access program.

Versal is the industry’s first Adaptive Compute Acceleration Platform (ACAP), described as heterogeneous compute devices with capabilities that far exceed those of conventional CPUs, GPUs, and FPGAs.

The integrated, multi-core, heterogeneous compute platform can be changed at both the hardware and software levels to dynamically adapt to the needs of applications and workloads in data centre, automotive, 5G wireless, wired and defence markets.

The Versal ACAP is natively software programmable and its architecture features a flexible, multi-Tbit per second network on chip (NoC). By seamlessly integrating all engines and key interfaces, the NoC ensures that the ACAP is available at boot and easily programmed, whatever a developer’s specialism, be they software developers, data scientists or hardware developers, says Xilinx. Tools, software, libraries, IP, middleware, and frameworks are available to dynamically customize accelerated computing using industry-standard design flows.

The Versal ACAP portfolio is based on TSMC’s 7nm process technology ad is claimed to be the first platform to combine software programmability with dynamically configurable domain-specific hardware acceleration. It includes a mix of next-generation scalar engines for embedded compute, adaptable engines for FPGA silicon programmability, and intelligent engines for AI inference and advanced signal processing for advantageous raw performance and performance per Watt capabilities compared to CPU and GPU implementations, adds Xilinx.

The Versal AI Core series delivers the portfolio’s highest compute and lowest latency, for AI inference throughput and performance through the AI Engine. This core is optimised for cloud, networking and autonomous technology and is claimed to offer the highest range of AI and workload acceleration available in the industry.

The Versal Prime series is optimised for connectivity and in-line acceleration to operate across multiple markets and a diverse set of workloads, continues the company.

Both the Versal AI Core series and Versal Prime series include multiple devices, each with dual-core Arm Cortex-A72 application processors, dual-core Arm Cortex-R5F real-time processors, over 2m logic cells of adaptable hardware, and over 3,000 DSP engines optimised for high-precision floating point and low latency. The Versal AI Core devices offer up to 400 AI Engines optimised for AI inference and advanced signal processing workloads.

The Versal portfolio includes four additional series of devices, architected to deliver scalability and AI inference capabilities for diverse applications, from cloud and networking to wireless communications, edge computing, and endpoints.

The Versal AI Core and Versal Prime series will be generally available in the second half of 2019.

http://www.xilinx.com

> Read More

Axiomtek’s PoE embedded vision system delivers AI for IoT

Multi-chip package is set to advance 5G CPE modem roll out

Raspberry Pi 4 computer triples speed, says Farnell

Xilinx ships Versal ACAP devices

About Smart Cities