Smart Cities Archives - Page 188 of 577 - SmartCitiesElectronics.com

Basler uses Nvidia Jetson Orin Nano SoMs for entry level AI

Thu, Sep 22, 2022

Add-on camera kits announced by Basler at the Nvidia GTC conference this week, support the Nvidia Jetson Orin Nano system on modules (SoMs). These SoMs have set a new baseline for entry-level edge AI and robotics, said Basler. The five and 13Mpixel camera kits are ready to use vision extensions that include an adapter board, camera, lens and cables.

The Basler Camera Enablement software package enables smooth integration of Basler’s Camera Software pylon with the Nvidia JetPack software development kit. A Camera Enablement package includes a Basler camera kernel driver, device-tree overlays for the supported boards, GenTL Producer and the pylon Viewer application. The camera kits are claimed to offer the most cost-effective prototyping approach and are suitable for mass production.

The camera kits are scheduled to be available from the beginning of 2023. Basler explained that customers can develop edge AI, robotics, AIoT (AI of things) and embedded solutions on the Nvidia Jetson AGX Orin developer kit and then deploy and scale applications on the AGX Orin production modules. When the Orin Nano and Orin NX modules become available they will also be compatible.

“Our portfolio ranges from different cameras and products to complement vision applications to the Pylon software framework for AI-based image processing solutions,” said Malene Fricke, head of module business at Basler.

“The upcoming Nvidia Orin Nano SoMs are an ideal processing platform to develop advanced edge AI vision systems for . . . customers focusing on automation applications in smart factories, the medical and life sciences market, robotics and logistics use cases, as well as intelligent traffic systems,” she added.

Basler manufactures imaging components for computer vision applications. In addition to classic area scan and line scan cameras, lenses, frame grabbers, light modules, 3D products, and software, the company offers embedded vision products, consulting services, customer specific software development as well as customised products.

The company’s products are used in a variety of markets and applications, including factory automation, medical, logistics, retail, and robotics.

Founded in 1988, the Basler Group employs around 1,000 people at its headquarters in Ahrensburg, Germany and other locations in Europe, Asia, and North America.

http://www.baslerweb.com

> Read More

Automated test and validation accelerates measurements in high speed devices

Thu, Sep 22, 2022

Engineers can test PCIe 6.0 (Base) receivers using the PCIe 6.0 receiver test option developed by Tektronix in collaboration with Anritsu.

The PCIe 6.0 (64GTransfers per second) automated test runs on the Tektronix DPO70000SX series real time oscilloscope and the Anritsu MP1900A BERT.

The automated test and validation option is claimed to render faster test times, and has an intuitive user interface.

The cohesive PCIe 6.0 transmitter and receiver test presents the ability to test increasing amounts of data as it is transferred at increasing speeds. Data-intensive sectors, like AI, cloud, and data centres use PCIe as a de-facto standard, explained Tektronix. While the recently released Gen 6.0 standard provides fast speeds to handle such workloads, testing brings challenges to those charged with integrating PCIe 6.0 on devices.

PCIe receiver validation is plagued by the sensitivity of calibrating the stressed eye signal across a high loss channel. Tektronix’s PCIe receiver test targets the required bit error rate (BER) while intuitive step-by-step tools provide link training routines for the Anritsu MP1900A BERT to ensure the receiver is tested accurately.

The Tektronix PCIe 6.0 Base Rx solution includes what is claimed to be the fastest 64GTransfers per second (PAM4) Rx calibration in the industry. Novel algorithms calibrate the transmitter’s additional pre-cursor added to the 6.0 Base specification.

The Tektronix PAMJET DSP tool provides critical 64GTransfers per second jitter and noise measurements with instrument noise compensation. Tektronix said that the automated test leverages the market’s first 6.0 Base Tx solution and enhances the capabilities to address receiver challenges at the latest data rate introduced by the PCI Special Interest Group (PCI-SIG).

The TekRxTest automation software provides a single control panel with the power to manage a Tektronix oscilloscope and Anritsu BERT during receiver calibration. This intuitive software wizard guides users through short- and long-channel calibration steps with ease, resulting in an accurate and repeatable calibration at 64 GTransfers per second.

Integrating the Anritsu MP1900A and Tektronix DPO70000SX real time oscilloscopes deliver a complete physical layer PCIe test system that can be expanded to PCIe 6.0 as the market continues to develop technologies for high-speed networks, said Kazuhiro Fujinuma, director of solution marketing department service infrastructure division for Anritsu.

Synopsys also collaborated with Tektronix on early validation of the PCIe 6.0 DesignWare IP, providing early-stage prototypes for Tektronix’s partnership with Anritsu.

“The PCIe 6.0 interface is instrumental to enabling real-time data connectivity with low latency and high throughput in HPC [high performance computing], storage, and AI applications,” said John Koeter, senior vice president of marketing and strategy for IP at Synopsys. “Conducting early, accurate and efficient validation of the Synopsys silicon-proven PCIe 6.0 IP . . . demonstrates that the IP adheres to the latest PCI-SIG specifications, enabling designers to deliver interoperable products that include Synopsys PCIe 6.0 controller, PHY, verification, and integrity and data encryption (IDE) IP faster and with less risk.”

Tektronix’s PCIe 6.0 receiver test is available now worldwide for use with DPO70000SX real time oscilloscopes.

http://www.Tek.com

> Read More

Nvidia advances AI with IGX edge AI computing

Thu, Sep 22, 2022

At this week’s GTC event, Nvidia has announced the IGX edge AI computing platform for autonomous systems in healthcare, manufacturing and logistics.

The edge AI computing platform is secure for safety in these demanding market sectors. It is programmable and configurable to meet different application use-cases.

For manufacturing and logistics, IGX provides an additional layer of safety in highly regulated physical-world factories and warehouses, explained Nvidia. For medical edge AI use cases, IGX delivers secure, low-latency AI inference to address the clinical demand for instantaneous insights from instruments and sensors for procedures, such as robotic-assisted surgery and patient monitoring.

“As humans increasingly work with robots, industries are setting new functional safety standards for AI and computing,” said Jensen Huang, founder and CEO of Nvidia. He said that IGX will help companies build the next generation of software-defined industrial and medical devices that can safely operate in the same environment as humans,” he said.

IGX uses Nvidia IGX Orin, the compact and energy-efficient AI supercomputer for autonomous industrial machines and medical devices.

IGX Orin developer kits will be available early next year for prototyping and testing. Each kit comes with an integrated GPU and CPU for AI compute and an Nvidia ConnectX-7 SmartNIC for low latency, secure networking.

There is also a software stack with critical security and safety capabilities that can be programmed and configured for different use cases. Features allow enterprises to add proactive safety into environments where humans and robots work side by side, such as warehouse floors and operating rooms, Nvidia said.

The IGX platform can run Nvidia AI Enterprise software, which optimises the development and deployment of AI workflows and ensures organisations have access to necessary AI frameworks and tools. Nvidia is working with operating system partners like Canonical, Red Hat and SUSE to bring full-stack, long-term support.

Nvidia Fleet Command is for the management of IGX in industrial and medical environments. It allows organisations to deploy secure, over-the-air software and system updates from a central cloud console.

Adlink, Advantech, Dedicated Computing, Kontron, Leadtek, MBX, Onyx, Portwell, Prodrive Technologies and YUAN will be among the first embedded-computing manufacturers to create products based on the IGX design.

Siemens is collaborating with Nvidia to implement IGX at the edge for autonomous factories. The two companies are collaborating with projects across industrial computing, including with digital twins and for the industrial metaverse.

For healthcare, the IGX platform supports Nvidia Clara Holoscan, a real-time, AI computing platform used by more than 70 medical device companies, start ups and medical centres to advance AI in clinical settings and surgical systems.

http://www.nvidia.com

> Read More

Baseband IP is for 5G RAN ASICs

Thu, Sep 22, 2022

To address the MIMO compute challenges around 5G Open RAN (O-RAN), Ceva has introduced PentaG-RAN which is claimed is the industry’s first baseband IP for ASICs.

The IP for ASICs targets cellular infrastructure in both base station and radio configurations, including distributed units (DU), and remote radio units (RRU), from small cell to massive multiple-input, multiple output (mMIMO). This heterogeneous baseband compute platform has been designed to “significantly reduce the entry barriers” for companies wishing to break into opportunities available in O-RAN equipment.

PentaG-RAN addresses the inability of the current platforms to scale effectively to mMIMO dimensions or support new O-RAN use cases where power and cost are significant factors. The platform for a complete L1 PHY (physical layer 1) with optimal hardware/software partitioning incorporates powerful vector DSPs, PHY control DSPs, flexible 5G hardware accelerators and other specialised components required for modem processing chains. According to Ceva, it delivers up to 10 times savings in power and area compared to available FPGA and commercial off the shelf (COTS) CPU-based alternatives. To further reduce design risk and expediate ASIC design, Ceva offers its co-creation services to PentaG-RAN customers to develop the PHY subsystem which extends up to designing the complete chip.

The PentaG-RAN platform addresses both base station and radio compute configurations. For the base station, it supports macro DU/vDU and small cells with scalability for L1 inline DU/vDU acceleration. This configuration handles the acceleration of the main processing chains (data and control), for both symbol-to-bit domains (including FEC) and frequency processing (including FFT and equalisation). Advanced algorithms including channel estimation and MMSE calculation are mapped to the Ceva-XC DSP for processing and power efficiency. It includes a powerful resource pool for accelerating COTS platforms and supports high and low-PHY 7.2x split partitioning based on Open RAN specifications.

For the radio portion, it supports Open RAN low PHY, massive MIMO and a beamformer for a scalable compute platform for massive MIMO beamforming processing on the RRU side, including beamformer and bBeamforming weight calculation. This configuration offers compute and PPA efficiency, enabling cost reduction and integration options with transceivers, compared to FPGAs and other COTS alternatives. It supports a range of use cases from small cell and macro, to massive MIMO 32TR to 64TR, for both sub-6 and mmWave and supports O-RAN 7.2x splits.

PentaG-RAN is supported by Ceva’s Virtual Platform Simulator (VPS), a unified System-C modelling environment that allows pre-silicon software development, dimensioning, architecture proof-of-concept, and modelling of all platform components. The VPS also includes reference software implementation for main processing chains, as well as beamforming use cases.

PentaG-RAN will be available for general licensing in Q4 2022.

http://www.ceva-dsp.com

> Read More

Basler uses Nvidia Jetson Orin Nano SoMs for entry level AI

Automated test and validation accelerates measurements in high speed devices

Nvidia advances AI with IGX edge AI computing

Baseband IP is for 5G RAN ASICs

About Smart Cities