SoC Provides Neural Network Acceleration

Brainchip has claimed itself as the first company to bring a production spiking neural network architecture to market. Called the Akida Neuromorphic System-on-Chip (NSoC), the device is small, low cost and low power, making it well-suited for edge applications such as advanced driver assistance systems (ADAS), autonomous vehicles, drones, vision-guided robotics, surveillance and machine vision systems. Its scalability allows users to network many Akida devices together to perform complex neural network training and inferencing for many markets including agricultural technology (AgTech), cybersecurity and financial technology (FinTech).
According to Lou DiNardo, BrainChip CEO, Akida, which is Greek for ‘spike,’ represents the first in a new breed of hardware solutions for AI. Artificial intelligence at the edge is going to be as significant and prolific as the microcontroller.

The Akida NSoC uses a pure CMOS logic process, ensuring high yields and low cost. Spiking neural networks (SNNs) are inherently lower power than traditional convolutional neural networks (CNNs), as they replace the math-intensive convolutions and back-propagation training methods with biologically inspired neuron functions and feed-forward training methodologies. BrainChip’s research has determined the optimal neuron model and training methods, bringing unprecedented efficiency and accuracy. Each Akida NSoC has effectively 1.2 million neurons and 10 billion synapses, representing 100 times better efficiency than neuromorphic test chips from Intel and IBM. Comparisons to leading CNN accelerator devices show similar performance gains of an order of magnitude better images/second/watt running industry standard benchmarks such as CIFAR-10 with comparable accuracy.

The Akida NSoC is designed for use as a stand-alone embedded accelerator or as a co-processor. It includes sensor interfaces for traditional pixel-based imaging, dynamic vision sensors (DVS), Lidar, audio, and analog signals. It also has high-speed data interfaces such as PCI-Express, USB, and Ethernet. Embedded in the NSoC are data-to-spike converters designed to optimally convert popular data formats into spikes to train and be processed by the Akida Neuron Fabric.

Spiking neural networks are inherently feed-forward dataflows, for both training and inference. Ingrained within the Akida neuron model are innovative training methodologies for supervised and unsupervised training. In the supervised mode, the initial layers of the network train themselves autonomously, while in the final fully-connected layers, labels can be applied, enabling these networks to function as classification networks. The Akida NSoC is designed to allow off-chip training in the Akida Development Environment, or on-chip training. An on-chip CPU is used to control the configuration of the Akida Neuron Fabric as well as off-chip communication of metadata.

The Akida Development Environment is available now for early access customers to begin the creation, training, and testing of spiking neural networks targeting the Akida NSoC. The Akida NSoC is expected to begin sampling in Q3 2019.

Brainchip | www.brainchip.com

NXP i.MX RT1060 Crossover Processors Released

First announced in February at Embedded World 2018, NXP Semiconductors has released its i.MX RT1060 Crossover processor, with the company claiming a mere ten months from concept to market launch.

The i.MX RT1060 is the latest addition to what NXP calls a crossover processor series and expands the i.MX RT series to three scalable families. The i.MX RT1060 doubles the On-Chip SRAM to 1 MB while keeping pin-to-pin compatibility with i.MX RT1050. This new series introduces additional features ideal for real-time applications such as High-Speed GPIO, CAN-FD, and synchronous parallel NAND/NOR/PSRAM controller. The i.MX RT1060 runs on the Arm Cortex-M7 core at 600 MHz.

This device is fully supported by NXP’s MCUXpresso Software and Tools, a comprehensive and cohesive set of free software development tools for Kinetis, LPC and i.MX RT microcontrollers. MCUXpresso SDK also includes project files for Keil MDK and IAR EWARM.

The i.MX RT crossover are designed to bridge the gap between high-performance and integration while minimizing costs to meet today’s need for high performance embedded processing at the edge node. According to NXP the series were designed to combine high performance MCU processing with the functionality of applications processors, at reduced costs, thereby enabling advanced computation and machine learning capabilities in millions of connected edge devices. The i.MX RT1060 is available now, and is priced at $3.48 (10,000s).

NXP Semiconductors | www.nxp.com

NXP and Alibaba Cloud Team up for IoT Deal

NXP Semiconductors has announced a strategic partnership with Alibaba Cloud, the cloud computing and business unit of Alibaba Group. The two companies are working together to enable development of secure smart devices for edge computing applications and have plans to further develop solutions for the IoT.

NXP_logo_RGB_webAs part of the partnership, AliOS Things, the Alibaba IoT operating system has been integrated onto NXP applications processors, microcontroller chips, and Layerscape multicore processors. Both NXP’s i.MX and Layerscape processors are currently the only embedded systems on the market using the Alibaba Cloud TEE OS platform. The new solution benefits various markets including automotive, smart retail and smart home. And it is currently being applied in applications such as automotive entertainment and infotainment systems, QR code payment scanning applications and smart home speakers.

With the partnership between NXP and Alibaba Cloud Link in the field of IoT security, NXP has become a council member of the ICA IoT Connectivity Alliance. In the future, the two companies plan to jointly develop solutions to support application development in different fields including smart manufacturing and smart city.

The ‘Annual Report of China IoT Development 2015-2016’ predicts that the amount of equipment connected to IoT globally will reach 20-50 billion by 2020, with 80 percent of that equipment in China. NXP’s robust product portfolio covers offering from the edge node to gateway and comprehensive cloud IoT solutions. NXP’s products are widely used in smart homes, smart cities, smart transportation and secure connectivity.

NXP Semiconductors | www.nxp.com

IoT Tool Suite Supports Bluetooth 5

Rigado has announced its Edge Connectivity Suite with full support for Bluetooth 5. Designed for large-scale commercial IoT deployments, Rigado’s Edge Connectivity solution is comprised of Bluetooth 5 end-device modules and the Vesta IoT Gateway, which includes cloud-based tools for secure deployment and updating.

The Edge Connectivity Suite actively addresses a growing need for low-power wireless within commercial IoT applications like asset tracking, smart lighting and connected retail and hospitality. The company’s Bluetooth 5-enabled solutions support the flexibility, interoperability and security demands of large-scale commercial IoT deployments. Moreover, the suite addresses the market need for Edge Computing at scale, paving a secure and cost-effective road for data from device-to-cloud.

Specifically designed for companies who need to develop, deploy and manage a large number of connected devices and gateways, the Rigado Edge Connectivity Suite provides seamless integration between IoT devices and the Cloud. It includes:

  • BMD-340 angleCertified end device modules – Rigado modules (see photo) save connected product teams six months and $200K+ in design, test and certification. Fully Bluetooth 5 enabled, Rigado modules also feature mesh networking capabilities, ideal for applications like smart lighting, asset tracking, and connected retail.
  • Edge computing gateways – Rigado Vesta gateways manage connectivity to end devices and ensure data reaches public and private cloud services. They also support custom edge applications to process data and offer local device control. Flexible wireless options and customizability mean that companies can optimize their gateway for cost-effective enterprise deployment.
  • Cloud-based tools for secure deployment and updating– Companies require a scalable solution to securely manage updates to devices in the field. With that in mind, every Rigado gateway ships with Rigado’s provisioning and release management system that integrates with existing development tools for secure updating at scale.

Rigado | www.rigado.com

New Consortium to Focus on Industry 4.0

Advantech has announced that it has partnered with Mitsubishi Electric, Omron, NEC, IBM Japan and Oracle Japan to establish the “Edgecross Consortium” to overcome boundaries between companies and industries in order to realize collaboration between factory automation (FA) and IT. The objective is to create new value centered on edge computing. The joint press conference to announce the establishment of the Edgecross Consortium was held on November 6, 2017, in Tokyo, Japan, and the establishment of the consortium is planned for November 29, 2017.

content-image-1510189308969

The Edgecross Consortium will focus on the easy realization of FA and IT collaboration through its open software platform (“Edgecross Software Platform”), which is based on edge computing. By residing on the edge layer, the Edgecross Software Platform facilitates connectivity between factory shop floors and value chains, while enabling the rapid acquisition, analysis, and utilization of data, which are essential to smart manufacturing. In addition, user applications can be easily created and shared on the edge to realize various IoT utilization purposes.

The consortium aims to co-develop the platform specifications for and achieve certification of Edgecross products, in addition to co-marketing promotion and global co-selling for consortium members. Starting from Japan and then globally, you will be able to catch a glimpse of the Edgecross booth display at the SCF 2017 Fair and Smart Factory EXPO 2018. Advantech is organizing the IoT Co-Creation Summit from November 1 to 3, 2018, in Suzhou, China, with more than 2500 attendees from all over the world. The summit will seek to foster the three phases of growth brought about by IoT and to demonstrate the co-creation achievement of IoT WISE-PaaS, EISs, and SRPs.

Advantech | www.advantech.com