All the latest quantum computer articles

See the latest stories on quantum computing from eeNews Europe

Monday, October 28, 2019

Renesas moves ThreadX to Azure RTOS for IoT

By Nick Flaherty

Renesas is to support Microsoft's Azure RTOS real time operating system (formerly ThreadX) across its microcontroller (MCU) and microprocessor (MPU) portfolio for the Internet of Things (IoT).

ThreadX is one of the most deployed RTOS worldwide for microcontrollers, and is currently integrated with the Renesas Synergy Software Package (SSP). The new versions, now called Azure RTOS since the purchase in April this year, will be added to the broader set of Renesas RA and RX microcontrollers and RZ/A microprocessors, starting with out-of-the-box integration into the new Renesas RA Flexible Software Package (FSP).
The deal, one of the first for Azure IoT. aims to provide a complete chip-to-cloud IoT solution based on Renesas’ intelligent and secure devices and Microsoft Azure IoT building blocks including Azure RTOS, Azure IoT device SDK for C, IoT Plug and Play, IoT Central and IoT Hub.

“Renesas Synergy customers have already experienced Microsoft Azure RTOS’ out-of-the-box connectivity to the Azure IoT Hub, making it an ideal choice for the billions of connected endpoints on the horizon,” said Sailesh Chittipeddi, Executive Vice President, General manager of Renesas’ IoT and Infrastructure Business Unit. “We are excited to bring Azure to customers working with our MCUs and MPUs. This collaboration will enable customers to bring their innovative connected solutions to market quicker with proven, trusted, and secure IoT building blocks.”
“Making Microsoft Azure RTOS offerings broadly available across Renesas’ product families, including Renesas Synergy and RA MCU devices means customers, will be able to deliver new connected solutions quicker," said Sam George, Corporate Vice President for Azure IoT at Microsoft 
"Our collaboration with Renesas is making enterprise grade IoT development as easy and seamless as possible.”

Wednesday, October 23, 2019

Protocol boosts WiFI range for smart home IoT devices

By Nick Flaherty

A group of researchers in the US has created a protocol that significantly extends the distance a Wi-Fi-enabled device can send and receive signals as noise and could be implemented as a software upgrade.

Test results for the On-Off Noise Power Communication (ONPC) protocol show a boost of over 60m, which could also be used to reduce power consumption and extend battery li.

"That's the really cool thing about this technology: it's all done in software," said Phil Lundrigan, assistant professor of computer engineering at led by a Brigham Young University (BYU). "In theory, we could install this on almost any Wi-Fi-enabled device with a simple software update."

ONPC is programmed right on top of the existing Wi-Fi protocol using the same hardware. While Wi-Fi requires speeds of at least one megabit per second (1 Mbps) to maintain a signal, the "ONPC" protocol Lundrigan and his co-authors created can maintain a signal on as low as 1 bit per second -- one millionth of the data speed required by Wi-Fi.

To do so, Lundrigan, Neal Patwari of Washington University (in St. Louis) and Sneha Kasera of the University of Utah adjusted the transmitter in a Wi-Fi-enabled device to send wireless noise in addition to data. They programmed into the Wi-Fi sensor a series of 1s and 0s, essentially turning the signal on and off in a specific pattern. The Wi-Fi router was able to distinguish this pattern from the surrounding wireless noise (from computers, televisions and cell phones) and therefore know that the sensor was still transmitting something, even if the data wasn't being received.

"If the access point (router) hears this code, it says, 'OK, I know the sensor is still alive and trying to reach me, it's just out of range,'" said Patwari. "It's basically sending one bit of information that says it's alive."

But according to Lundrigran, one bit of information is sufficient for many Wi-Fi enabled devices that simply need an on/off message, such as a garage door sensor, an air quality monitor or even a sprinkler system. During the research, the team successfully implemented their ONPC protocol, along with an application to manage the protocol ("Stayin' Alive"), ultimately extending the range of an off-the-shelf device 67 meters beyond the range of standard Wi-Fi.

The researchers made clear to point out that their ONPC protocol is not meant to replace Wi-Fi or even long-range wireless protocols like LoRa, but is meant to supplement Wi-Fi. Specifically, only when Stayin' Alive detects that the Wi-Fi device has lost its connection, it starts transmitting data using ONPC.

The protocol can also be used with LoRa sub-GHz links. "We can send and receive data regardless of what Wi-Fi is doing; all we need is the ability to transmit energy and then receive noise measurements," Lundrigan said. "We could apply this to cellular or Bluetooth as well," he said.

Tuesday, October 22, 2019

Solid state drives on a roller coaster ride

By Nick Flaherty

Solid state drives (SSDs) and the underlying NAND flash technology have seen a huge swing in shipments over the last six months.

Following a plunge during the first three months of the year, global SSD shipments snapped back to growth in the second quarter, as demand from the client system market boosted the performance of third-party suppliers like Kingston, says the latest report from IHS Markit | Technology.

Worldwide SSD revenue rose to $57.9 billion in the second quarter, up 9.7 percent from $52.8 billion in the first quarter, says IHS, now a part of Informa Tech. Shipments declined by 8.4 percent sequentially in the first quarter. This dramatic shift from contraction to expansion represents a more than 18-percentage-point swing between the first and second quarters.

Despite the vigorous rise in unit shipments, sequential SSD revenue growth in the second quarter was limited to 1 percent. That’s because overall average selling prices (ASPs) fell by 9.7 percent in the second quarter, preventing revenue growth.
“Falling prices are continuing to fuel SSD demand growth in client systems,” said James Zhao, Principal Analyst, SSD and HDD, at IHS Markit | Technology. “The global average selling price (ASP) for client SSDs declined by 6.7 percent sequentially in the second quarter, marking the sixth consecutive quarter of ASP contraction in the client segment. These declines are making SSDs more affordable for a larger number of systems.”

The client SSD segment, which includes devices like mobile and desktop PCs, was the main driver of the rebound, Client SSD shipments increased by 11.9 percent sequentially in the second quarter. In contrast, the enterprise SSD market saw sequential growth of only 1.9 percent.

This surge in the client segment disproportionately benefitted third-party players that don’t manufacture the NAND flash memory microchips used within their products, despite chip maker Samsung being teh largest SSD supplier by far (see chart).

Kingston exemplified this third party trend, with shipments rising by 23.3 percent sequentially in the second quarter. This propelled Kingston to the No. 3 rank in SSD unit market for the first time in three years, after placing fifth in 2017 and 2018 and fourth in 2016. The company saw six consecutive quarters of double-digit revenue growth.

“Third-party players like Kingston are outperforming the overall market because NAND components are available at such low prices compared to a year ago,” Zhao said. “These companies are able to take advantage of these lower costs and gain market share rapidly.”

Intel captured the No. 2 position in SSD revenue share in the second quarter 2019. Because Intel focuses more on the enterprise market, its SSD ASP is far above average. Intel shipped more than 4 million SSDs in the quarter, up 17.5 percent compared with the previous quarter.

The SSD and HDD Storage Market Tracker from IHS Markit | Technology provides coverage of quarterly/annual shipments, ASP, density, revenue and various dimensions of insights for major strorage devices, including HDDs and SSDs.

Low power front end module for IoT designs quadruples range

By Nick Flaherty

Skyworks has launched a a 2.4 GHz front-end module (FEM) designed to boost the range of Zigbee, Thread and Bluetooth Low Energy wireless links in the Internet of Things. The multichip module can be used in applications ranging from sensors, beacons, smart meters and thermostats to wireless cameras, smoke and CO detectors, as well as wearables and medical devices.

The SKY66405-11 measures 1.9 x 1.9 x 0.56 mm and provides increased efficiency as well as more than 4X range extension compared to a standalone system-on-chip (SoC).

The module includes a power amplifier with +13 dBm output power alongside a low-noise amplifier with 2 dB noise figure, low-loss bypass path, single ended transmit and receive switches, and digital controls compatible with CMOS levels.

A wide supply voltage range of 1.7 V to 3.6V allows the device to be used in battery powered IoT applications over a broad spectrum of the battery discharge curve and it covers worldwide unlicensed bands. A fast switch on/off time of under 800ns and 1μa sleep current also help reduce the power consumption, and no external bias resistor is required.

The operating temperature ranges from -40 to 105° C, making the module also suitable for lighting and industrial applications in the IIoT.

Monday, October 21, 2019

Bioprinting living cells with a 3D printer

By Nick Flaherty

Slightly outside the things we usually cover, but the 3D printing in a new process developed at TU Wien in Vienna opens up new types of sensors as living cells can be integrated into fine structures created in a two photon laser-based 3D printer.

Tissue growth and the behaviour of cells can be controlled and investigated particularly well by embedding the cells in a delicate 3D framework. This is achieved using additive 3D printing methods - so called "bioprinting" techniques. However, this involves a number of challenges: Some methods are very imprecise or only allow a very short time window in which the cells can be processed without being damaged. In addition, the materials used must be cell-friendly during and after the 3D biopriting process. This restricts the variety of possible materials.

A high-resolution bioprinting process with completely new materials has now been developed at TU Wien (Vienna). Using a particular "bio ink" for the 3D printer, cells can now be embedded in a 3D matrix printed with micrometer precision - at a printing speed of one meter per second, orders of magnitude faster than previously possible.

"The behaviour of a cell behaves depends crucially on the mechanical, chemical and geometric properties of its environment," said Prof Aleksandr Ovsianikov, head of the 3D Printing and Biofabrication research group at the Institute of Materials Science and Technology at TU Wien. "The structures in which the cells are embedded must be permeable to nutrients so that the cells can survive and multiply. But it is also important whether the structures are stiff or flexible, whether they are stable or degrade over time".

It is possible to first produce suitable structures and then colonise them with living cells - but this approach can make it difficult to place the cells deep inside the scaffold, and it is hardly possible to achieve a homogeneous cell distribution that way. The much better option is to embed the living cells directly into the 3D structure during the production of the structure - this technique is known as "bioprinting".

Printing microscopically fine 3D objects is no longer a problem today. However, the use of living cells presents science with completely new challenges: "Until now, there has simply been a lack of suitable chemical substances," said Ovsianikov. "You need liquids or gels that solidify precisely where you illuminate them with a focused laser beam. However, these materials must not be harmful to the cells, and the whole process has to happen extremely quickly."

In order to achieve an extremely high resolution, two-photon polymerization methods have been used at TU Wien for years. This method uses a chemical reaction that is only initiated when a molecule of the material simultaneously absorbs two photons of the laser beam. This is only possible where the laser beam has a particularly high intensity. At these points the substance hardens, while it remains liquid everywhere else. Therefore, this two-photon method is best suited to produce extremely fine structures with high precision.

However, these high resolution techniques usually have the disadvantage of being very slow - often in the range of micrometers or a few millimeters per second. At TU Wien, however, cell-friendly materials can be processed at a speed of more than one meter per second - a decisive step forward. Only if the entire process can be completed within a few hours is there a good chance of the cells surviving and developing further.

"Our method provides many possibilities to adapt the environment of the cells," said Ovsianikov. Depending on how the structure is built, it can be made stiffer or softer. Even fine, continuous gradients are possible. In this way, it is possible to define exactly how the structure should look in order to allow the desired kind of cell growth and cell migration. The laser intensity can also be used to determine how easily the structure will be degraded over time.

"Using these 3D scaffolds, it is possible to investigate the behaviour of cells with previously unattainable accuracy. It is possible to study the spread of diseases, and if stem cells are used, it is even possible to produce tailor-made tissue in this way."

Quantum tunnelling boost for smaller chips ... Wireless charging coils in Formica laminates ... Quad rotor wind turbines

Power news this week by Nick Flaherty at eeNews Power

. Bizen uses quantum tunnelling for smaller chips
. CATL breaks ground on German battery gigafactory
. Two team for US automotive discrete power device production
. Formica adds wireless charging to its laminates
. Calcium looks to replace lithium battery technology
. Wind turbine boost from modelling
. Single-chip ASIL-D battery monitor for mid-to-large cell counts
. MJD bipolar transistors up to 8A
. Compact form factor for DC link film capacitor

Sunday, October 20, 2019

AI data compaction boosts IoT bandwidth

By Nick Flaherty

AtomBeam Technologies is launching data compaction software that enables intelligent IoT networking by reducing data transmission by 75%, while adding virtually no latency. The patented technology will be on show at Mobile World Congress (MWC) event in Los Angeles this week.

The data compaction system reduces the amount of IoT data transmitted by 70-80%, effectively expanding existing bandwidth by 3-5x. AtomBeam's proprietary AI/machine learning can provide savings of 30 to 70% in hardware, network and cloud costs. The software also provides a built-in security layer.

This means companies can see up to 75% data reduction in their cloud costs from solution providers, while satellite companies and their users having up to 5x more bandwidth at lower latency
Cellular and NB-IoT (LPWAN) service providers can double the transmission range of links with increased security and see a 25% reduction in battery life usage, enabling a similar OPEX savings for companies to maintain their field sensors.

AtomBeam Technologies licenses its software to manufacturers of IoT devices and gateways, edge and data centre servers and cloud storage systems as well as to networks such as satellites, LPWANs, and telcos. 

"AtomBeam is at the intersection of four powerful technology trends," said James Brehm, Founder & CEO of James Brehm and Associates, a leading consulting and market intelligence company. "It is leveraging AI-ML to manage the Big Data that 5G and IoT companies will be forced to deal with as thousands of IoT devices, gateways and sensors contribute data. Much of the data generated in IoT is repetitive. AtomBeam's technology compacts duplicative and oftentimes extraneous information, reducing network load and providing virtual network capacity gains of an additional 75% or more. This company is going to play a vital role in the movement of IoT data."

According to McKinsey, 127 new devices connect to the Internet every second. Statista reports that 75 billion devices are expected to be connected by 2025. IoT companies will need the capability of real time data transmission combined with added security to assure their networks can safely manage the tsunami of data projected from IoT devices.

"With a fast, light code, AtomBeam can be embedded in virtually any network, data centre server or connected IoT device," said Charles Yeomans, CEO of AtomBeam Technologies.

Tuesday, October 15, 2019

Pixel 4 smartphone adds Infineon 60GHz radar chip

By Nick Flaherty

Infineon is supplying a 60GHz radar system for gesture detection in the next generation Pixel 4 smartphone being developed by Google.

An integrated antenna system in the XENSIV BGT60TR13C package allows sensing of the presence and movement of people and objects with high precision or measures distances and speeds. This chip is the base for Google’s Soli technology and has now been integrated for the first time into a smartphone for gesture control.

"With our radar technology devices become 'context-aware’. This means that they can finally understand their environment and react much more purposefully," says Andreas Urschitz, Division President for Power Management and Multimarket at Infineon Technologies. "The precise motion detection by the 60 GHz radar chip turns the Google Pixel 4 smartphone into a gesture control system. This is a revolution in the human-machine-interaction. At Infineon, we are furthermore working on the fusion of multiple sensors to simplify interaction and increase the usefulness of the devices."

Infineon’s radar technology has its roots in the automotive sector. Radar sensors have been effectively measuring distances, speeds and movements while driving for decades. Infineon has further developed these functions for small devices. The 60 GHz chip is a complete radar system with antennas on a very small area (5 x 6.5 mm) coupled with low power consumption. The BGT60 family has a transmit current of 480mA at 3.3V, giving a power consumption in use of 1.5W, which is high for a smartphone but the previous devices were aimed at tethered applications such as smart speakers with a sensing distance from 20cm to 5m. However the overall power usage of the sensor will depend on the software controlling the duty cycle, or how often it is used, as it is only likely to be used in infrequent short bursts. This version will be optimised for shorter distances from 1cm to 1m, which will reduce the power consumption by potentially by a factor of 5. Further circuit optimisations will reduce that further, and we’ll publish the figures when we get them. 

The radar sensor can perceive movements in rooms or measure distances from objects in the millimetre range with utmost precision. With the appropriate software, the motion data is converted into functions, so that control via gestures is possible without touching the device.

 The fusion of multiple sensors in a single device creates new solutions that measure and improve air quality or intelligently control burglary protection, for example. In addition to voice-controlled assistants, 'intelligent' household appliances or wearables, buildings, so-called smart buildings in particular, are becoming more interactive. Sensors detect the number of people in the rooms or can adjust the need for light sources to improve safety and energy efficiency.

Monday, October 07, 2019

NXP ARM multicore processors get Lynx hypervisor

By Nick Flaherty

NXP's QorIQ Arm-based processors can now use the LynxSecure Separation Kernel Hypervisor for to keep applications separate in avionics designs.

Lynx has ported LynxSecure onto the NXP QorIQ Layerscape 1046A (LS1046A) multicore communication processor integrating quad 64-bit Arm Cortex-A72 cores. This joins the existing support for NXP S32V234 MPSoC.
“Multicore architectures are seeing increasing adoption in avionics systems due to performance requirements and the lack of availability of single core processors," said Geoff Waters, Senior Principal Engineer at NXP Semiconductors and Chairman of the NXP- led Multicore for Avionics Working Group. "The porting of LynxSecure separation kernel hypervisor technology is an important step forward. It offers an exceptional technology for managing the complexity associated with multicores, reducing the design and certification time for safety critical systems based on NXP Layerscape SoCs.”

David Beal, Director of Product Marketing at Lynx Software Technologies, added, “Increasingly, avionics, transportation and other safety-critical systems rely on redundant platforms. Now, system architects and engineers who have strict requirements around performance, safety and security can take full advantage of the features and capabilities offered by NXP LS1046A within a straight-forward, secure-by-design software systems architecture that is enabled by LYNX MOSA.ic. The combination results in high performance and immutable security while preserving modularity and providing a natural path for future software-migration as newer, processing platforms become available.”

The LynxSecure separation kernel forms the foundation of the LYNX MOSA.ic framework to address markets such as avionics, transportation, industrial and medical systems. By providing the highest levels of performance, security and safety protections, LYNX MOSA.ic enables complex, systems that are responsible for both critical and non-critical functions. This reduces system design, integration, test and certification efforts.