Access the latest quantum technology

Quantum technology in Bristol and bath - find out more about how you can access the commercialisation of quantum technology for sensing and security

Monday, September 25, 2017

Power news this week - details from the iPhone 8 teardown

By Nick Flaherty

. Teardown shows iPhone 8 power play

. Infineon sells Newport fab to power foundry startup

. Gogoro raises $300m for its smart scooter technology

. Spinning a solid-state carbon nanofibre supercapacitor

. Harnessing electricity from blood flow

. Stretchable waterproof organic solar cells for wearables and IoT designs


. Superjunction trench MOSFETs save space in rugged designs

. 12V DC-DC converter for embedded applications eliminates heatsink

. Monitor chip boosts software power measurement accuracy to 99 percent in Windows 10

TECHNICAL PAPERS . CUI: Improved Longevity and Performance in DC Fans

. Intersil: Battery management system tutorial

Rugged 900MHz wireless for embedded and OEM applications in critical industries

By Nick Flaherty

FreeWave Technologies has launched a new line of 90MHz wireless modules for long range connectivity in secure M2M and IoT applications.

The ZumLink Z9-PC ( radios are designed to function in rugged environments for a variety of applications across Industrial IoT (IIoT) markets, including precision agriculture, oil and gas, unmanned systems, smart cities and utilities. Future versions of the module will feature full programmability with third-party applications for IIoT projects.
"The Z9-PC is our most advanced embeddable radio to date," said Phil Linker, senior product manager at FreeWave Technologies. "Not only does it support the functionality of FreeWave's ZumLink 900 MHz Z9-PE radio, but will include the ZumIQ App Server with the ability to host third-party applications at the edge of IIoT networks. Companies in critical industries can utilize data and provide command and control like never before."

Over the past year, FreeWave has introduced the ZumLink 900 MHz Series of radios to include the Z9-P, Z9-PE, Z9-C, and Z9-T. The Z9-PC is the culmination of the 900 MHz radio series, as it provides versatility and flexibility, and, most importantly, provides programmability for industries in need of future-proof IIoT technology. The radios use  FreeWave's ZumBoost Network Acceleration Pack with techniques such as compression, packet aggregation, forward error correction, and a patent-pending Adaptive Spectrum Learning technology to ensure maximum throughput to meet the demands of today's wireless applications.

Additional features include five RF link rates from 115.2 kbps up to 4 Mbps, 128bit and 256bit AES counter mode encryption, multiple link rates, channel sizes and modulations, as well as frequency hopping and single channel operation.

Related stories:

Boosting post-quantum IoT security

By Nick Flaherty at

Engineers, developers, and embedded security experts face a major challenge. Designs using 8-, 16-, and 32-bit processors lack the computing and memory resources needed to incorporate contemporary security methods like Elliptic Curve Cryptography (ECC) that are not vulnerable to the coming generation of quantum computers.

“A common misconception is that constrained IoT devices cannot run public-key solutions, and this is true about legacy methods—either they don't fit or, if they can be made to fit, they don’t perform,” said Derek Atkins, Chief Technology Officer at SecureRF.  He points to two of the company’s security solutions—the Ironwood Key Agreement Protocol (Ironwood KAP) and Walnut Digital Signature Algorithm (WalnutDSA) that are suited in terms of size, speed, and power for the small processors powering the IoT. Based on Group Theoretic Cryptography methods, SecureRF’s tools are at least 60 times faster than ECC and consume up to 140 times less energy.
Ironwood KAP and WalnutDSA provide critical protection against all known quantum attacks. As recently announced by the United States National Security Agency (NSA), quantum resistance is a critical requirement for users looking to deliver secure sensors, devices, and actuators that have multi-year lifespans. These systems can protect IoT devices even when quantum computers become available and render currently used methods such as ECC and RSA obsolete.

Related stories:

Sunday, September 24, 2017

Imagination broken up and sold to two venture capital firms

By Nick Flaherty

UK chip design supplier Imagination Technologies has been broken up and bought by two US venture capital firms.

Imagination includes the MIPS processor cores, PowerVR graphics cores and ENSIGMA wireless cores. The company had been in trouble since losing a deal with Apple for graphics technology. That dispute is still on-going.

The MIPS division, including all the patents, has been bought by Tallwood MIPS Inc, a company indirectly owned by Tallwood VC of Palo Alto, for $65m in cash. The deal is expected to close next month.
This removes one of the main blocks to Canyon Bridge Capital Partners buying the rest. The venture capital that last week was blocked by US president Trump from buying Lattice Semiconductor over concerns about its backing by Chinese funds.

“This transaction, which is separate to the offer for Imagination by Canyon Bridge, will ensure MIPS remains an independent licensing business," said Andrew Heath, Chief Executive of Imagination. “MIPS has gained notable momentum over the last year or so having secured a number of design wins, and made good progress towards profitability. This significant progress, allied to Tallwood’s investments in the semiconductor industry, will provide increased customer choice and new job opportunities based on the continued growth of the MIPS business.”

Tallwood has investments in startups what were bought by Qualcomm (such as SiRF), Silicon Labs and Broadcom as well as direct investments in Cavendish Kinetics and Accent.

The Canyon bridge deal values the bulk of Imagination at $550m, almost ten times that of MIPS. Imagination acquired MIPS in 2013 for $60m, although the majority of the patents at that time were sold to ARM.

"Imagination has made excellent progress both operationally and financially over the last 18 months until Apple’s unsubstantiated assertions and the subsequent dispute forced us to change course," said Heath. "The acquisition will ensure that Imagination – with its strong growth prospects – remains an independent IP licensing business, based in the UK, but operating around the world. Imagination employs a large number of hugely talented individuals who have developed our market leading technology. They and the business as a whole will benefit from Canyon Bridge’s investment in Imagination as it moves to the next stage of its development.”

Related stories:

Friday, September 22, 2017

Facebook and Microsoft complete 160Tbit/s fibre link

By Nick Flaherty at

Facebook and Microsoft have completed a new trans-Atlantic undersea cable that will carry up to 160Tbit/s of data and are getting it ready for operation.

The new "MAREA" cable will be the highest-capacity subsea cable to ever cross the Atlantic with eight fibre pairs. The 6,600 km submarine cable system will be operated and managed by Telxius, Telefónica's new telecommunications infrastructure company, and connect the US to southern Europe, from the data hub of Northern Virginia to Bilbao, Spain and then to network hubs in Europe, Africa, the Middle East and Asia. This route is south of other transatlantic cable systems, helping ensure more resilient and reliable connections for customers in the US and Europe, and Telxius will sell on capacity on the cable. 

A key change is that MAREA is designed to be interoperable with a variety of networking equipment, reducing the cost. 

“Marea comes at a critical time,” said Brad Smith, president of Microsoft. “Submarine cables in the Atlantic already carry 55 percent more data than trans-Pacific routes and 40 percent more data than between the U.S. and Latin America. There is no question that the demand for data flows across the Atlantic will continue to increase and Marea will provide a critical connection for the US, Spain, and beyond.”

"As the world is increasingly moving toward a future based on cloud computing, Microsoft continues to invest in our cloud infrastructure to meet current and future growing global demand for our more than 200 cloud services, including Bing, Office 365, Skype, Xbox Live and the Microsoft Azure platform," said Christian Belady, General Manager, Datacentre Strategy, Planning & Development at Microsoft. "By building the cable along this new southern route, we will also increase the resiliency of our global network, helping ensure even greater reliability for our customers."

"Facebook wants to make it possible for people to have deep connections and shared experiences with the people who matter to them most — anywhere in the world, and at any time," said Najam Ahmad, Vice President of Network Engineering at Facebook. "We're always evaluating new technologies and systems in order to provide the best connectivity possible. By creating a vendor-agnostic design with Microsoft and Telxius, we can choose the hardware and software that best serves the system and ultimately increase the pace of innovation. We want to do more of these projects in this manner — allowing us to move fast with more collaboration. We think this is how most subsea cable systems will be built in the future."
"Working with Facebook and Microsoft in this project will reinforce Telxius' position as a worldwide leading infra company not only expanding on-net capabilities to Spain and the USA but also allowing us to capture market opportunities in Western Europe and new emerging routes in Northern Africa and other regions," said Rafael Arranz, Chief Operating Officer of Telxius Cable.

The cable was landed in Spain in June, and the physical link has been completed. HOwever the equipment still needs to be installed and commissioned, so the link will not be operational until early 2018.

IoT trust frameworks go head to head with launch of blockchain alliance

By Nick Flaherty at

A new open source software foundation has been launched to create a secure, scalable, interoperable, and trusted IoT ecosystem using the distributed ledger technology of blockchain.

However, the Trusted IoT Alliance (TIA) will be up against existing collaborative frameworks that have already released their specifications.

The TIA was formed from the Blockchain & IoT Protocol working group that arose out of conversations amongst technologists during 2016 working on different aspects of the IoT and blockchain problem, all recognising the need for a "smart contract" interface between IoT devices and blockchain systems across the industry.

At the inaugural meeting of the group in Berkeley, CA in 2016, they agreed to create a foundation for industry collaboration in a few basic areas. It will promote technologies that have been successful for members, such as Hyperledger Fabric, Ethereum, Quorum and BigChain DB.
It is not necessarily pushing blockchain as a term. “The word ‘blockchain’ is not in the title because we are unsure of the persistence of the term ‘blockchain’ in this fast-evolving technology landscape,” says the Alliance. “In 2013 the buzz term was ‘Bitcoin’, in 2014-16 ‘blockchain’ became more common, and now in 2017 the term ‘DLT’ or ‘Distributed Ledger Technology’ is gathering momentum. DLT is being used as an umbrella term to describe both permissioned and public ledgers, and could possibly displace the use of the term blockchain, hence our caution.”

A key point is that the TIA is backed by Bosch, Cisco and Gemalto on the operational board but it will be competing with other Internet security groups that already have IoT frameworks available, although not yet with blockchain technologies.

Back in March 2016 The Online Trust Alliance (OTA) formally ratified and released its Internet of Things (IoT) Trust Framework. The Framework was the first global, multi-stakeholder effort to address IoT risks comprehensively including 30 principles providing guidance for device manufacturers and developers to enhance the security, privacy and sustainability of their devices and data they collect.

This was based on work by the OTA’s IoT Trustworthy Working Group (ITWG) hat was established in January 2015 address security, privacy and life-cycle sustainability in IoT products and services, which is pretty much the same remit as the TIA.

The initial scope of the ITWG focused on connected / smart home products and consumer facing health and fitness wearable technologies and the IoT Trust Framework reflects feedback from nearly 100 US and international companies and organisations ranging from major retailers and device manufacturers, to security and privacy subject matter experts, to consumer testing and advocacy organisations, to academia, government and law enforcement.

"The framework demonstrates the ability of the public and private sectors to coalesce and develop meaningful and actionable guidelines, enhancing security and privacy practices globally," said Craig Spiezle, Executive Director and President of the Online Trust Alliance. "Adopting the Framework, is a step towards realising the promise of IoT and making security and privacy competitive product features."

Thursday, September 21, 2017

Intel shows early details of 10nm Falcon Mesa FPGAs

By Nick Flaherty at

Intel has released the first details of the FPGAs it plans to make on its 10nm FINFET process. Although the latest processors on the technology have been delayed, FPGAs are a good process prover as they have lots of repeatable elements and lots of redundancy that can survive low process yields.

The Falcon Mesa FPGAs will target data centerr wireless 5G, network function virtualization (NFV), automotive, industrial and military/aerospace applications with 112 Gbps serial transceiver links and the latest PCI Express Gen4 x16 support with data rates up to 16 GT/s per lane for next-generation data centres.

The new family will build upon several innovative technologies from the current Intel Stratix 10 14 nm FPGAs designed by Altera. The Hyperflex architecture, which uses registers, called hyper-registers, throughout the FPGA, is being optimised for 10nm as you would expect, and this will be included in the Quartus Prime high-level design tools.

The Embedded Multi-Die Interconnect Bridge (EMIB) packaging technology will allow the higher performance transceivers to be made on a different process technology and integrated into the package alongside the FPGA fabric. 

A next-generation high bandwidth memory (HBM) DRAM memory architecture will also deliver 10x the performance of discrete memory designs in a smaller form factor with lower power consumption.

More details at a later date, says Intel, and you can see where the technology came from in this story from 2010: Altera heads to 28nm for Stratix V FPGA Family

Related stories:

LPWAN technologies open up IoT opportunities

By Nick Flaherty

Mobile operators have a significant opportunity to offer low power wide area systems for the Internet of Things, but competition will be intense, says Analysys Mason in a new report.

The report gives an insight into the technology being used, with AT&T using LTE-M, Orange using LoRa and LTE-M and Vodafone using NB-IoT. However, smaller operators have an opportunity in using different technologies.

The report gives a good view of selected operator LPWAN technology choices:

The alternative technology will be better suited to particular use cases. It is uncertain how large the differences between NB-IoT and LTE-M will be in terms of price and performance, but if NB-IoT does have clear price advantages or longer battery life than LTE-M, it will be more attractive for some applications such as metering. This could benefit T-Mobile USA, says the report.

An alternative technology could open up the enterprise market where most challenger operators have a limited presence.

However, the report warns that operators also need to consider the consequences of being late to market if they adopt a wait-and-see approach. The appeal of waiting is clear: the existing enterprise base is limited and NB-IoT or LTE-M are relatively quick to deploy. However, operators with ambitions to play a role in IoT risk foregoing a valuable learning experience in the early LPWA market and the opportunity to capture some of the early demand. Bouygues Telecom, KPN and Swisscom all gained valuable expertise by entering the market early with LoRa, which should help them when developing 3GPP IoT networks say the analysts.

Related stories:

First functional silicon of server DIMM buffer chipset for DDR5 memory

By Nick Flaherty at

Rambus is sampling the first functional silicon of a double data rate (DDR) server DIMM (dual inline memory module) buffer chipset prototype for the next-generation DDR5 memory technology. 

This represents a key milestone for Rambus as the industry’s first silicon-proven memory buffer chip prototype capable of achieving the speeds required for the upcoming DDR5 standard.

“Data-intensive applications like Big Data analytics and machine learning will be key drivers for the adoption of DDR5, with enterprise close behind,” said Luc Seraphin, senior vice president and general manager of the Rambus Memory and Interfaces Division. “We are proud to provide an early path to adoption with the first working buffer chip prototype running at the anticipated performance of next-generation DDR5. This demonstrates our continued dedication to be first to market and remain on the leading edge of industry standards.”

According to JEDEC, next-generation DDR5 memory will offer improved performance and power efficiency, providing double the bandwidth and density over DDR4. With that, server DIMM chipsets, like registered clock drivers and data buffers, will be critical to enabling higher memory capacities while maintaining peak performance. This Server DIMM chip prototype uses the signal integrity and low-power, mixed-signal design expertise of Rambus to enable development of next-generation data centres.

The new features in DDR5 will target the specific problems being caused by the exponential growth in data generated by IoT, cloud services and real-time data analytics, and address the need to continuously process, move, and analyse that data faster. These enhancements are expected to include higher density, a new command structure and new power saving features. DDR5 will also likely introduce signal equalisation and error correction, and these will work their way down to industrial designs.

The Rambus Memory and Interface Division looks at the power, performance, and capacity challenges of the communications and data centre computing markets with standards-compatible and custom memory and serial link solutions include chips, architectures, memory and SerDes interfaces, IP validation tools, and system and IC design services. 

Related stories:

Wednesday, September 20, 2017

Zero power sensor is always on

By Nick Flaherty

Researchers at Northeastern University have developed an infrared sensor that can be always on but also essentially zero power, a combination that could drive applications in the Internert of Things (IoT).

The work for the US DARPA agency is led by Electrical and Computer Engineering Associate Professor Matteo Rinaldi and uses a plasmonically-enhanced micromechanical photoswitch.

The sensor uses a grid of nanoscale patches whose specific dimensions limit them to absorb only particular IR wavelengths. “The charge-based excitations, called plasmons (that can be thought of somewhat like ripples on the surface of water), are highly localized below the nanoscale patches and effectively trap specific wavelengths of light into the ultra-thin structure, inducing a relatively large and swift spike in its temperature,” said Rinaldi. These temperature spikes, in turn, lead to an upstream sequence of events that culminates in circuit-completing deformations of other parts of the sensor.

“What is really interesting about the Northeastern IR sensor technology is that, unlike conventional sensors, it consumes zero stand-by power when the IR wavelengths to be detected are not present,” said Troy Olsson, manager of the N-ZERO Program in DARPA’s Microsystems Technology Office. “When those IR wavelengths are present and impinge on the Northeastern team’s IR sensor, the energy from the IR source heats the sensing elements which, in turn, causes physical movement of key sensor components. These motions result in the mechanical closing of otherwise open circuit elements, thereby leading to signals that the target IR signature has been detected.”

“The technology features multiple sensing elements—each tuned to absorb a specific IR wavelength,” said Olsson. “Together, these combine into complex logic circuits capable of analysing IR, which opens the way for these sensors to not only detect IR energy in the environment but to specify if that energy derives from a fire, vehicle, person or some other IR source.”

A primary goal of the N-ZERO program is to develop fundamental technologies that open the way to new and more capable sensor systems. The NU team points to the IoT with hundreds of billions of devices, ranging from cars, to appliances, to remotely deployed sensors. “The capability of consuming power only when useful information is present will result in nearly unlimited duration of operation for unattended sensors deployed to detect infrequent but time-critical events, with a groundbreaking impact on the proliferation of the Internet of Things,” said Rinaldi.

A deep dive into Oracle's latest SPARC processor

By Nick Flaherty at

I have a soft spot for the SPARC processor architecture and all its struggles over the years, so I was interested to see the launch of the eight generation chip by Oracle.

The 32-core M8 chip is naturally optimised for Oracle's database technology as well as Java, adding security and performance boosts that support up to 256 threads in as little as 2U of space in the data centre.

The SPARC M8 processor-based systems, including the Oracle SuperCluster M8 engineered systems and SPARC T8 and M8 servers, are designed to integrate with existing infrastructures and include fully integrated virtualisation and management for private cloud. 

It includes Silicon Secured Memory, which provides always-on hardware-based memory protection for intrusion protection and end-to-end encryption and Data Analytics Accelerators (DAX) with open API’s.

The M8 is a four-wide issue, out-of-order core with up to 192 instructions in-flight, and supports up to eight hardware threads. It provides dynamic threading to optimize for the highest possible per-thread performance and software can activate up to eight hardware threads (strands) on each core via critical threads optimisation. The processor hardware then dynamically and seamlessly allocates core resources among the active strands. 
The per-thread performance is improved with the enhanced on-chip L2 and L3 cache design and increased processor frequency. The 64 MB L3 cache is fully shared and hot cache lines are migrated to the closest partition to minimise latency. 

The cores and the last level cache are organised as two partitions each containing 16 cores and 32 MB of L3 cache and each core has its own 32 KB L1 instruction cache, 16 KB L1 data cache, and 128 KB L2 data cache. Four cores then share a 256 KB L2 instruction cache for a total of 64 MB L3 cache per processor that is fully shared, 16-way set-associative, and inclusive of all inner caches. Any L3 partition may serve a request from any of the 32 cores of the SPARC M8 processor. Hot cache lines are migrated to the closest L3 cache partition to optimize for performance.

The processor can dynamically trade per-thread performance for throughput by running up to 256 threads, or it can run fewer higher-performance threads by devoting more resources to each thread. This flexibility allows the system to balance overall throughput versus per-thread performance.

“Oracle has long been a pioneer in engineering software and hardware together to secure high-performance infrastructure for any workload of any size,” said Edward Screven, chief corporate architect at Oracle. “SPARC was already the fastest, most secure processor in the world for running Oracle Database and Java. SPARC M8 extends that lead even further.”

The M8 is twice as fast at encryption and hasing as x86 devices (read Xeon) which are comparable to the M7, says Oracle. 
The database gives twice the OLTP (online transaction processing) performance per core over x86 and 7x database analytics than x86, with twice the Java performance. The latest DAX v2 produces 8x more efficient Java streams processing, improving overall application performance.

While there is some doubt over the future of SPARC, Oracle has also publicly committed to supporting the Solaris operating system until at least 2034.

WND aims for pan-UK Sigfox network by 2019

By Nick Flaherty at

One of the main low power wide area network operators in the UK is planning to have 95% coverage of the country by 2019 to connect up devices in the Internet of Things. This uses the French proprietary Sigfox network.

“The feedback we’re getting from many of our channel partners is that some customers remain uncertain about the reliability, security and costs associated with using the Sigfox network," said Neal Forse, chief executive of WND UK. "We’re spending a lot of time reassuring customers that Sigfox is proven technology and the basis for a commercially viable IoT network. Unfortunately, the failure of previous network operators to deliver on their promises has left us with a confused market."
“If we look at what WND is achieving in international markets such as Europe and Latin America as it continues to deploy Sigfox networks throughout Brazil, Mexico, Columbia and Argentina, then customers should be reassured that we will achieve the same outcomes here in the UK," he said.

He points to a difference business model based on usage. "We firmly believe that it is the responsibility of the network operator to fund the development of the network infrastructure. We base licence fees on message volume, resulting in very low ‘per message’ charges for high-volume users. Furthermore, unlike other IoT networks, you don’t require a gateway to use Sigfox.”

WND UK started its rollout in March and currently provides coverage to over 37% the UK’s population and is on track to cover 50% before the end of 2017, with over 450 base station installations across the UK. However, IoT applications are often in areas of low population so using a measure of population can be misleading for coverage. The advantage of the lower frequency LPWAN technology is greater range to cover such areas. 

Ideally equipment makers can add a Sigfox module to a system and connect directly to the network, which then uses the Sigfox server (or Azure cloud) to make the connection to the wider Internet as required.

"Sigfox base stations can be deployed at customer sites anywhere in the UK and be fully operational in a matter of weeks," said WND partner Tim Streather at Spica Technologies.

Using a non-IP, proprietary protocol is a benefit for security, says Forse. “Although Sigfox-ready devices are classed as IoT objects, they are not directly connected to the internet and do not communicate using internet protocol," he said. "Sigfox-enabled devices have a built-in behaviour; when this requires data to be transmitted or received, a device will communicate via a radio message. Each message is picked up by several access stations and is delivered to the Sigfox cloud network over a secure VPN, which then relays it to a predefined destination, typically an IoT application. Because Sigfox devices don’t have IP addresses, they are not addressable for rogue hackers to gain access.”

Such a security design ensures that Sigfox-ready devices are prevented from sending data to arbitrary devices via the internet and are shielded from interception by strict firewall measures.

Related stories on the Embedded blog:

Monday, September 18, 2017

Lime launches £100 software defined radio board

By Nick Flaherty at

Lime Microsystems has launched a smaller, less expensive version of its popular LimeSDR software defined radio prototyping boards costing just £100.

The LimeSDR Mini is available for pre-order immediately through Crowd Supply at $139 (£100, €115). ( The price of LimeSDR Mini has been set to make it affordable for virtually anyone to develop applications using a cutting edge hardware.

The original 2016 LimeSDR campaign raised over $1.25 million and had the backing of over 5,000 developers, including BT/EE.

To simplify the process further, Lime has partnered with Ubuntu, launching an app store for LimeSDR systems. Using these, developers can download apps and get a mobile, IoT or other wireless network running in minutes.

Lime hopes that by simplifying wireless system development and cutting the cost it can attract a new generation of talented minds to wireless system design, and with it speed innovation.

BT/EE is using the hardware to increase coverage in remote regions; and has also run hackathons to encourage innovation. Vodafone has announced it will use the hardware as part of its OPEN RAN vision.

The boards support virtually any type of wireless between 10 MHz and 3.5 GHz - such as UMS, LTE, SDM, LoRa, Bluetooth, zigbee, RFID, Zigbee and digital broadcasting. LimeSDR solutions are programmable and based on open source technologies with an extensive community of developers, offering the potential to democratise wireless communications and enable global wireless connectivity.

To register for details of the new campaign, or receive technical specifications, please visit the Crowd Supply LimeSDR Mini page. (

Related stories:

Power news this week

POWER NEWS from eeNews Europe 

. Blockchain system tackles component counterfeiting

. Zero power sensor technology spins out into startup

. Flexible organic ultracapacitor targets pacemakers and medical implants

. Configurable VLIW core boosts energy efficiency for long battery life

. Dual metal ion battery approaches performance of supercapacitors

. Dual input prioritizer reserves power for backups

. New package options for 600V power modules simplify motor drives

. Offline power controller for adaptor designs with no heatsink


. C&K: Optimized Switch Solutions for Smart Metering Applications

. Intersil: Battery management system tutorial

By Nick Flaherty

Software AG re-launches Cumulocity IoT network with self service option

By Nick Flaherty at

Following its acquisition of the Nokia Siemens spinout Cumulocity back in March, Software AG has re-launched the cloud-based IoT platform.

Cumulocity IoT will address today’s market demand for easy, fast, and scalable IoT solutions to support enterprises of all sizes, regardless of the level of sophistication of their IoT requirements, says the company. Recent market research shows that 90 percent of executives surveyed feel the Industrial Internet of Things (IIoT) is critical to their companies’ success, but only 16 percent have a comprehensive IIoT roadmap, so the new Cumulocity IoT portfolio is offering self-service IoT solutions to enable organizations develop their own IoT applications and solutions. 
The Cumulocity IoT portfolio operates at the edge, in the cloud or on premise, giving flexibility in designing, building, deploying and upgrading enterprise IoT platform, solutions and services. It includes a range of pre-packaged solutions such as Condition Monitoring, Predictive Maintenance and Track & Trace, as well as quick and easy device and sensor management.

The original Cumulocoty technology has been integrated with Software AG’s Digital Business Platform to target companies using IoT to enhance their own products and services and IoT service providers seeking to build and commercialize an IoT platform.

“Cumulocity IoT is unique in that it provides enterprises with the ability to very quickly start limited scale, cloud-based IoT projects and, as their sophistication grows, be confident to build IoT platform and IoT services that will grow effortlessly with them," said Wolfram Jost, Chief Technology Officer at Software AG. "Enterprises can develop and implement IoT services at the optimal balance between risk and ambition, at their pace, at their choice. Cumulocity IoT is already in use by leading technology companies like Deutsche Telekom, Gardner Denver, Octo Telematics, NTT Communication, Reliance Group, and many more. This gives us a head start in our ambition to become a leading Technology provider in the IoT domain.”

“We recently announced a joint venture in the IoT area which fully underlines our commitment to this growing market. ADAMOS is a major IoT collaborative alliance with the leading manufacturing enterprises Dürr, Zeiss, DMG Mori and ASM PT. The purpose of this IoT alliance is to establish global industry standards for the manufacturing sector using Software AG’s Cumulocity IoT as the technology foundation. ADAMOS partners and their customers will be able to leverage a comprehensive IoT platform and services at a fraction of the cost of developing their own solutions in a significantly shorter period of time.”

Related stories:

Toshiba pushes 5GHz RF CMOS receiver under 1V

By Nick Flaherty at

Toshiba has developed a low voltage 5GHz receiver for IEEE802.11ax wireless LAN, targetting the Internet of Things.

802.11ax has initially been aimed at the enterprise, using the MIMO antenna structures to provide four times more performance. But the continued growth in the Internet of Things requires high-speed communications in environments increasingly crowded with numerous wireless LAN devices. 
This requires CMOS devices operating with supply voltages lower than 1.0V. Unlike digital circuits, the performance of the RF CMOS receivers, degrades drastically as voltage falls, and new circuit technology that can operate below 1.0V is required to overcome this problem.

Toshiba's research labs have developed three new technologies for low voltage RF receivers.
The first is a variable linearity RF amplifier (RFAMP). The signal amplitude is affected by power supply voltage so the linearity of RFAMPs decreases as voltage falls. Conventional RFAMP uses variable resistance to improve the linearity, but internal DC voltage and amplifier performance are also changed by adjusting the resistor value. In order to avoid these problems, the new RFAMP simultaneously employs two input paths: a high lnearity path and an internal voltage adjustment path. The RFAMP can adjust linearity without any change in the internal DC voltage.

The second is a low noise frequency converter. Conventional frequency converters use a DC current source to improve the performance of conversion switches in low voltage designs, but this additional current source degrades performance with low frequency noise. A new frequency converter design suppresses low frequency noise by moving the DC current source to the RF side of the conversion switches. Low frequency noise is up-converted to the RF frequency, which is separated from the desired low frequency signal. The result is sufficient switch performance without noise degradation.

The third is a current adder opamp. This amplifies the converted low frequency signal to level a sufficient high to input the analogue signal into the digital converter. As the maximum output signal level is limited by the supply voltage, the opamp is forced to reduce its operating voltage range. The new design suppresses the DC current source from the output stage with a high speed differential current mirror, and provides larger output voltage even in a low voltage environment. .

Integrating these three technologies into a 5GHz wireless LAN receiver has allowed the labs to reach the level of performance essential for next-generation wireless LAN.

Sunday, September 17, 2017

Indegy wins award for industrial IoT security technology

By Nick Flaherty

Industrial networks are increasingly utilizing IIoT devices for collecting data used to improve decision-making and reduce operations management costs. However, these technologies are creating security challenges that can impact the safety and reliability of industrial control systems (ICS) and processes, raising the prospect of industrial systems being hacked to create networks of malware. 

The Indegy Cyber Security Platform protects operational networks used in critical infrastructure including energy, utilities, petrochemicals and manufacturing from both external and internal threats.
By providing comprehensive visibility into the control-plane engineering activities performed in operational technology networks, Indegy’s technology automatically discovers all controllers (PLCs, RTUs, DCSs) on ICS networks, monitors all access and changes in real-time, and validates their integrity ensuring no unauthorized changes go undetected. Indegy enables advanced detection and response to threats that place the safety, reliability and security of industrial networks at risk before damage occurs.

“Indegy is proud to be selected the Overall IoT Security Solution of the Year from such a large number of respected security companies and top startups,” said Barak Perelman, CEO of Indegy. “Industrial IoT products are proliferating and placing critical infrastructure at risk. Indegy’s unique approach ensures that operators have the end-to-end situational awareness and comprehensive visibility needed to protect against cyber threats introduced by IIoT devices.”

The technology provides visibility and security that spans the entire operational (OT) network, with a unique focus on control-plane activities performed over proprietary engineering protocols. This lets Indegy identify threats before damage can occur as well as a comprehensive audit trail of all changes to ICS devices, who made them and when. It is the only solution that can capture access and changes to critical controllers, whether made over the network or by physically connecting to the controllers (via serial cable or USB drive).

Related stories:

Friday, September 15, 2017

Quantum sensor market to top $330m by 2025

By Nick Flaherty at

The global market for sensors using quantum technology is set to reach nearly $330m by 2025 as the technology takes off in embedded applications.

This growth is being driven by increasing adoption of quantum sensors in defence, oil & gas, transportation, and construction markets.

The report by Persistence Market Research sees the market expanding at a moderate rate of 4.3% a year from 2016 as it moves from a surge in demand of atomic clocks to improve the accuracy of time sensitive signals for IoT applications such as autonomous cars and drones.GPS systems are inherently dependent on precise calculations of timing for efficient communication between the GPS using device and the satellite providing GPS co-ordinates.

Quantum gravity sensors will bring considerable accuracy in the process of ground scanning for oil drilling and extraction and help drillers to get a clear picture of what is under the ground helping them discover pockets of crude oil left undiscovered with the usage of classical methods.

The type segment of the global quantum sensors market includes atomic clock, gravity sensor, magnetic sensor, rotation sensors, imaging sensors, and temperature sensors. Among type segments, atomic clock segment is expected to dominate the global quantum sensors market with US$ 127.8 Mn revenue in 2025. The atomic clock segment is expected to gain significant traction throughout the forecast period and gravity sensors will also see a steady growth of 4.7% ayear.

Among the industry verticals, defence applications are anticipated to expand at a CAGR of 5.3% during the forecast period to $101m, with Europe leading the business. It already accounts for the largest share in the global quantum sensors market and is expected to lead the market in future. In 2016, Europe quantum sensors market was valued at $73.7m and is expected to reach $111.8 in 2025 with a CAGR of 4.9%.

Related stories: 

Software Defined FPGAs launch on Amazon Web Services

By Nick Flaherty

FPGA maker Xilinx has ported its software defined development environment, SDAccel, is now available on Amazon Web Services (AWS) for use with Amazon Elastic Compute Cloud.

The EC2 F1 instances provide reconfigurable, custom-hardware acceleration with 16nm Virtex UltraScale+ FPGAs enabling customers to meet the demands of compute-intensive workloads like data analytics, video processing and machine learning. This means the algorithms can be used for IoT implementations without having to programme the FPGAs in Verilog or VHDL.

Using FPGA accelerators in the data centre is an increasingly popular system design technique and this means software developers less familiar with FPGAs can now get up to 50x better performance for their workloads. SDAccel automates the acceleration of software application written in C, C++ or OpenCL by building application-specific FPGA kernels for Amazon EC2 F1. SDAccel also supports traditional hardware designers writing optimized accelerators in lower-level hardware description languages such as VHDL Verilog.

SDAccel is accessible through the AWS FPGA developer AMI which allows developers to quickly build applications on the cloud, without requiring the purchase or installation of any software or hardware. Developers can then deploy and elastically scale their application to meet any customer demand using the full breadth of AWS services.

Related stories:

iPhone X design shows up the problems at Imagination

The launch of the iPhone 8 and iPhone X, which both use the A11 Bionic chip, is showing up the changes in GPU architecture design. Back in 2014 Imagination was relaxed about the relationship, but moves were already underway to develop a new GPU design and now the whole company is up for sale.

You can read the analysis at iPhone X chip shows problems of Imagination

Related stories:

Using blockchain to tackle counterfeit components

An interesting project from Accenture is looking to use the same blockchain technology that is behind bitcoin to stop counterfeit components entering the supply chain.

The demonstrator with Thales uses a range of different sensor and identification techniques to ensure that components are not replaced between the fab and the manufacturer. The first two demos are for diodes and FPGAs.

You can read the rest of my story at Blockchain system tackles component counterfeiting

Thursday, September 14, 2017

Quantenna teams with Technicolor on dual-band Wi-Fi extender for mesh networks

By Nick Flaherty at

Technicolor is using the QV860 chipset from Quantenna Technology in its OWA0130 dual-band extender to extend the coverage of home Wi-Fi gateways via a mesh network.

As service providers offer an increasing array of services to the home, the importance of a high-capacity, reliable Wi-Fi network becomes crucial for the successful deployment of services to multiple devices in the home. The OWA0130 will provide extended Wi-Fi coverage, at higher speed and reliability than existing mesh products says Quantenna.

Related stories:
The OWA0130 supports both 5GHz 802.11ac Wave 2 and 2.4GHz 802.11n networks. It can be easily configured to be a wired extender, which uses a gigabit Ethernet connection as the backhaul, or a fully wireless extender, in which the 5GHz 802.11ac is used as the backhaul. In either configuration, the OWA0130 supports a dual-band, dual-concurrent 4x4 5GHz 802.11ac and a 2x2 2.4GHz 802.11n network to downstream clients.

“A successful connected home experience today depends on reliable and high-quality WiFi infrastructures which can support a growing array of devices requiring wireless access to the Internet,” said Bart Vercammen, VP Product Management at Technicolor. “The collaboration between Technicoor and Quantenna offers network service providers an opportunity to offer subscribers consistent and excellent performance throughout their homes.”

Lionel Bonnot, Senior Vice President of Marketing and Business Development at Quantenna said, “Our line of wireless repeater solutions, which span from 5GHz 4-stream to tri-band 12-stream products, is the most complete in the industry and shows our commitment to this burgeoning market. We believe the combination of our wireless repeater solutions with our award-winning line of 4x4, 8x8 gateways and 802.11ax solutions positions Quantenna as the undisputed leader in Wi-Fi networking for service providers, retail and enterprise customers around the world.”

The OWA0130 also supports Technicolor’s cloud-based Wi-Fi monitoring solution, Wi-Fi Doctor.

Wednesday, September 13, 2017

Zephyr IoT RTOS adds Bluetooth 5 support

By Nick Flaherty at

Zephyr, the open source low footprint real time operating system targeting the Internet of Things (IoT), has added support for Bluetooth 5.

Version 1.9.0 of the kernel also includes a Bluetooth Qualification-ready BLE controller and support for mesh networking, as well as a lightweight Machine to Machine (LwM2M) support. There is also POSIX thread IPC support.

New chip architectures include the STM32F405, STM32F417, STM32F103x8 and TI CC2650 SoCs as well as Nordic's nRF52, and ST's STM32L4, and STM32F3

Various boards from Atmel, ST and Olimex have been added, as well as Intel's Quark boards, even though these are being withdrawn from the market.

JEDEC’s xSPI standard boosts execute-in-place memories

By Nick Flaherty at

Low power memory maker Adesto Technologies has used the latest JEDEC xSPI standard for a new family of eXecute-in-Place (XiP) non-volatile memories (NVM).

The xSPI standard establishes mechanical, electrical and transactional guidelines for developing high-throughput octal devices, such as Adesto’s EcoXiP product, and provides users with assurance of controller compatibility with peripheral devices.

“Divergent products with different solutions to speed communication between the host controller and memory can confuse controller designers as to which direction to follow,” said Gideon Intrater, CTO at Adesto. “This new standard will provide customers, including system developers and controller designers with assured compatibility. This milestone is key to faster and broader deployment of EcoXiP platform and provides the opportunity for new designs to move forward.”

 A growing number of products targeting the internet of things such as Wi-Fi and LTE communications modules, wearables, point-of-sale controllers and other embedded devices need more program memory than what can be implemented economically on-chip using embedded Flash or SRAM.  Together, projections for these devices total production in the hundreds of millions annually. Standalone DRAM devices are also not an attractive solution to be used as program memory in these applications due to standby power requirements and cost. Until now, to hit performance targets, system designers have been required to invest in memory solutions that are expensive, power-hungry and performance limiting.

The EcoXiP family is an eXecute-in Place memory that eliminates the need for on-chip embedded flash, while more than doubling processor performance, lowering system power consumption and reducing system cost compared to quad serial peripheral interface (SPI) devices.

The xSPI standard which was developed by a task force comprised of representatives from most NOR Flash device manufacturers and several PC and microcontroller companies, defines its primary applications as computing, automotive, Internet of Things (IoT), embedded systems and mobile systems.

“While many of our peers will target their devices to accelerate systems’ boot speed, EcoXiP not only provides this functionality, but also delivers the lowest power eXecute-in-Place performance,” said Intrater. “This virtually eliminates the need for on-board flash to store firmware.”

“Realtek’s ICs, such as our controllers for the digital home and Wi-Fi System-on-Chip devices, include a variety of solutions that offer high performance and ultra-low energy consumption,” said Scott Shen, Director, Communications Network, at Realtek. “With the adoption of JEDEC’s new xSPI standard, octal architectures are becoming very appealing to leverage improved SPI device communication. The combination of the octal xSPI interface and low power consumption makes Adesto’s EcoXiP an excellent XiP solution.”

Adesto is sampling a 32Mb device now, with a family of densities planned for the future.

PRQA updates its flagship static code analysis tools

By Nick Flaherty at

PRQA has released new versions of its QA·C and QA·C++ automated static code analysis tools. QA·C 9.3.1 and QA·C++ 4.1 provide improved analysis and an enhanced user interface.

“Automated static code analysis, by removing coding defects before compilation, cuts down on test and debug time, bringing projects to completion faster and with better quality code,” said Paul Blundell, CEO of PRQA. “The new releases of QA·C and QA·C++ are designed to improve still further the analysis phase, with subsequent benefits throughout the development process.”

The two tools share the PRQA Framework 2.2 platform which provides faster detection of coding defects with fewer “false positives”. A Diagnostic Severity Filter, controlled by a slider in the GUI, allows the analysis to focus on the most important code defects for a rapid turnaround, while new capabilities provide the ability to manage analysis configurations for multiple build variants within a single project. There is also a new facility to detect any issues caused by passing function parameters between translations.

Also in the new releases is extended full coverage of MISRA C:2012 coding guidelines and greater support for CERT C security coding standards.

While the tools can be integrated into standard industry IDEs such as Eclipse, a new GUI provides improved management tools, particularly for project configuration and traceability of defects and fixes throughout the product development cycle.

QA·C 9.3.1 and QA·C++ 4.1 are available now from

Related stories:

Renesas ramps up AI development with Codeplay deal

By Nick Flaherty

Renesas has boosted its development capability for machine vision and AI through a deal with development tool maker Codeplay Software.

The two will port ComputeAorta, Codeplay’s OpenCL open standard-based software framework to the Renesas R-Car system-on-chips (SoCs) for self driving cars. The framework will be available first on R-Car H3 as a proof of concept, then focusing on R-Car V3M and other R-Car SoCs of the Renesas autonomyPlatform for advanced driving assistance system (ADAS) and automated driving. The company has several GPU processor core technologies it is using for its chips. 

The new framework is designed to support software development for the R-Car’s latest image recognition IP, the IMP-X5, a multi-threading core optimized for computer vision and cognitive processing. Codeplay will also provide R-Car with ComputeCpp, an implementation of the SYCL open standard, enabling single source C++ software for high level and object-oriented programming.

The collaboration provides developers with standard software development tools and support for a wide range of open source computer vision or open source deep learning software, such as TensorFlow library. Developers can reuse their initial software assets or take advantage of the latest innovations in computer vision and cognitive processing, minimize porting effort and accelerate time-to-market.
“Computing performance is an essential factor in enabling highly-advanced computing platforms for autonomous cars” said Jean-Francois Chouteau, Vice President of Renesas Global ADAS Centre. “Renesas’ exclusive IMP-X5 computer vision IP plays a key role within the R-Car SoCs to deliver these advanced features and enhance safety. Giving access to Renesas’ computer vison hardware with OpenCL and SYCL through collaboration with Codeplay will enable our customers to accelerate time-to-market while achieving an outstanding performance benefit in their ADAS designs. This demonstrates the value of our “Renesas autonomy” platform for ADAS and automated driving.”

“OpenCL is a widely adopted open standard providing an efficient framework to support heterogeneous computing platforms,” said Andrew Richards, CEO of Codeplay. “Our ComputeAorta and ComputeCpp platforms will enable advanced vision and deep learning developers to use the familiar programming standards of OpenCL and SYCL and quickly benefit from R-Car’s advanced IMP-X5 computer vision IP.”

The ComputeAorta framework provides an OpenCL hardware abstraction layer that enables Tier1s and OEMs to take advantage of the full computing performance from Renesas R-Car SoCs, especially IMP-X5 computer vision IP. ComputeCpp extends OpenCL with a SYCL open standard implementation providing an efficient programming environment for multithread execution on heterogeneous cores. By using these frameworks, users can take full benefit of IMP-X5 computing performance from tailored multi-thread processing.

Tuesday, September 12, 2017

India taps former Nokia Siemens technology for 5G IoT

By Nick Flaherty at

India’s leading Telecom Operator Bharat Sanchar Nigam Limited (BSNL) is working with Coriant to chart a path to 5G and Internet of Things (IoT) in the country.

Coriant was formed from the combination of Nokia Siemens Networks (NSN) Optical Networks, Tellabs, and Sycamore Networks and will work with BSNL on new network architectures and services using 5G, IoT, SDN/NFV, and mobile edge computing technologies.

As part of its strategic collaboration, BSNL and Coriant will work together to develop 5G and IoT use cases such as rural connectivity, connected healthcare, industrial automation, public safety, video surveillance, energy, and agriculture. The deal also covers  research programs, knowledge sharing, and workshops focused on the latest networking trends and hyperscale architectures designed to support the scalable, ultra-low latency commercial 5G services and applications.

“We continue to focus on bringing our customers across India the latest innovations in communications technologies that enrich their personal and work lives,” said Shri Anupam Shrivastava at BSNL. “5G represents an enormous leap forward in capacity and throughput speeds, and we are pleased to team with our long-term technology partner Coriant to tap into these capabilities and explore real-world use cases for next-generation services and applications.”
“BSNL’s vision aligns nicely with our strategy of driving innovation in purpose-built 5G solutions and architectures,” said Shaygan Kheradpir, CEO and Chairman of Coriant. “We look forward to our joint work and the opportunity to help BSNL fast track development toward a 5G and IoT future and bring the value of these technologies to India and its citizens.” 

First multi-channel IoT transceiver certified for the Thread protocol

By Nick Flaherty at

Qorvo's two year old GP712 System-on-Chip is the first multi-channel Internet of Things (IoT) transceiver in the industry to be certified for the Thread protocol. 

With concurrent support for Thread and ZigBee on different RF channels, the GP712 enables designers to use a single transceiver on their gateway products to simultaneously serve both IEEE802.15.4 implementations.

The GP712 was developed by GreenPeak Technology in 2015 for long range and low power thorough an RF technology that limits interference. The multi-channel, multi-protocol design helps ensure that gateway devices are not made obsolete by shifting protocol standards. Capabilities can be added with simple software upgrades. Qorvo acquired Greenpeak in 2016.

Related stories:

“The Qorvo multi-channel IoT transceiver addresses the burgeoning demand for Thread-enabled products, and we’re excited to see our members begin to capitalize on this opportunity," said Grant Erickson, president of Thread. "Qorvo has been an active contributor to the Thread Group for several years, and we congratulate them on successfully completing Thread certification.”

Cees Links, general manager of Qorvo’s Wireless Connectivity business unit and the man behind Greenpeak, pointed to a market report from Berg Insight that said 91 million homes in the EU and US will be smart by 2020. "However, no analyst can predict today which networking standard will rule tomorrow. If design engineers want to keep their options open for future directions, they should choose a multi-protocol solution for their gateway,” he said.

The GP712 uses OpenThread, a programme from Google's Nest Labs initiative that aims to provide an open-source implementation for the Thread standard for smart home solutions. Qorvo’s Wireless Connectivity business is actively engaged as a contributor to the OpenThread community with the GP712 development platform.

Qualcomm teams with Nokia for 5G radio deployment

By Nick Flaherty at

Following yesterday's announcement of Intel's 5G development system, Qualcomm Technologies has teamed up with Nokia to conduct interoperability testing and over-the-air field trials based on the latest 5G New Radio (NR) Release-15 specifications. 

The testing and trials intend to demonstrate that the 5G NR technology can be used for commercial network launches in 2019 based on 3GPP standards.

Related stories:
In the testing and trials, the companies will showcase 5G NR technologies to efficiently achieve multi-gigabit per second data rates at latencies as low as 1ms, a key capability for IoT networks, autonomous vehicles, drones and industrial equipment. 

The trials will use Nokia's 5G FIRST solution that combines its AirScale base station transmitting over the 5G NR radio interface to the device prototype from Qualcomm Technologies. The companies will test end-to-end applications over-the-air between the base station and the device to simulate real-world scenarios across a broad set of 5G NR use cases and deployment scenarios. The testing will include 5G NR operation in sub-6 GHz spectrum bands such as 3.5 GHz and 4.5 GHz, as well as millimeter wave (mmWave) spectrum bands such as 28 GHz and 39 GHz, to trial the 5G NR unified design across diverse spectrum bands.

To demonstrate the performance and efficiency enhancements 5G NR will bring, the testing and trials will showcase advanced 3GPP 5G NR technologies including Massive Multiple-Input Multiple-Output (MIMO) antenna technology, beamforming techniques, adaptive self-contained TDD, scalable OFDM-based waveforms to support wider bandwidths, advanced coding and modulation schemes, and a new flexible, low-latency slot structure based design. This will include 5G NR operation in mmWave spectrum, employing advanced 5G NR antenna technology to deliver robust and sustained mobile broadband communications, including non-line-of-sight (NLOS) environments and device mobility.

“Qualcomm Technologies is committed to deliver the most compelling mobile broadband experiences based on 5G NR technology. We are making 5G a reality for mobile devices – including smartphones – from sub-6 GHz to millimeter wave – globally,” said Cristiano Amon, executive vice president of Qualcomm Technologies and president of Qualcomm CDMA Technologies.

“The appetite for 5G technology is clear, amongst both consumers and communications service providers,” said Marc Rouanne, president of mobile networks at Nokia. “We firmly share Qualcomm’s view that the demand for wide-scale mobile 5G is accelerating, and that is why we are so enthusiastic to work jointly with Qualcomm on developing and driving commercial technology and applications based on 5G NR using our 5G FIRST.”

The interoperability testing, which will start in the second half of 2017, are intended to track closely with the first 3GPP 5G NR specification that will be part of Release 15. The two are also working with network operators to conduct 5G NR over-the-air field trials starting in 2018 across various regions including China, Europe, Japan, and the USA. Tracking the 3GPP specification is important because it promotes adherence and validation with the global 5G standard, accelerating the time to standard-compliant devices and infrastructure. It will also ensure forward compatibility to future 3GPP 5G NR releases.

Monday, September 11, 2017

Intel launches 5G development platform ahead of December standard

By Nick Flaherty at

Intel is launching a development kit for 5G new radio (NR), the wireless radio standard that will be the foundation of 5G. In December, the 5G non-standalone new radio (NR) standard will be finalized with a goal to accelerate commercial 5G trials and deployments while the final standalone 5G NR standard is being developed. 

The 5G Mobile Trial Platform (MTP) will begin to support the new non-standalone NR standard in December using a combination of its (formerly Altera) FPGAs and Intel Core processors, says Asha Keddy, the vice president for Client and Internet of Things Businesses and Systems Architecture, and general manager of Next Generation and Standards at Intel.

5G will be a key technology for industrial and IoT applications with low latency connections.

Intel's MTP includes a 5G 28GHz mmWave RFFE (Radio Frequency Front-End) and five Arria 10 FPGAs that can reach data rates up to 3Gbits/s at 100-800MHz bandwidth with 4×4 and 8×8 mmWave MIMO multiple antenna arrays.

The single chip RF IC supports 50MHz, 100MHz, 200MHz, 400MHz and 800MHz transmission and reception at bands from 3.3-4.2GHz to 28GHz with support for 4×4 MIMO sub-6GHz and 2×2 dual-polarization MIMO for mmWave bands. 

Keddy is planning to work with equipment makers Ericsson and Nokia and participating in trials with operators like AT&TKorea TelecomNTT Docomo and Verizon to take non-standalone NR out of the lab and begin testing it in real-world situations in a move to deploy commercial 5G services before 2020.

Now equipment makers can test interoperability with devices sooner, operators can take the platform into real-world situations earlier, and standards bodies can collect data for final specs faster. This will drive results into the full standalone 5G NR standard, which is expected at the end of 2018.

5G can happen only when the network, cloud and device come together into a powerful end-to-end 5G solution, says Keddy, which is where Intel sees its opportunity.

Power news this week

A small selection of my latest power stories on eeNews Europe, from the latest attacks on European utilities and why battery startups fail

. Dragonfly hackers target power grids

. Why battery startups fail

. German electric aircraft startup sees $90 million boost

. Salt water makes lithium batteries safe

. Lightweight chassis changes battery pack design for electric vehicles

. Microfluidic channels cool stacked chips and datacentres

. Promising new material for solid state lithium batteries

. 150V synchronous buck-boost controller eliminates surge protection

. Free DC-DC converter tool boosts power supply designs

. 500V sine-wave driver in small surface mount reflow package for three-phase brushless fan motors

Xilinx boosts Huawei 's cloud server

By Nick Flaherty at

Huawei is using Virtex FPGAs from Xilinx to provide a 10 to 50x boost in performance for machine learning and video and data processing applications in its cloud server.

The Huawei FPGA Accelerated Cloud Server (FACS) is a platform that enables users to develop, deploy and publish new FPGA-based services and applications on Huawei Public Cloud. This is a key design win in the battle against Intel, which bought Altera's FPGA business and is being used in Microsoft's Azure platform.

Related stories:
The inherent ability of an FPGA to be reconfigured and reprogrammed at any time is a critical advantage in today’s data centres. The hardware can be reconfigured in under a second to provide different instantiations for the next workload, giving significant flexibility that is not available with custom ASICs. The ability to manage this re-configuration in teh data centre is a key step forward.

“The Huawei FACS is a fully integrated hardware and software platform offering developer-to-deployment support with best-in-class industry tool chains and access to Huawei’s significant FPGA engineering expertise,” said Steve Langridge, Director, Central Hardware Institute, Huawei Canada Research Centrr. “Our considerable collaboration, both in the labs and in standards groups, has enabled us to create the best value possible for our customers.”

The FPGA Accelerated Cloud Server is available on the Huawei Public Cloud, and developers can register for the public beta at