Categories
Hackster.io

Arduino Oplà IoT Kit // Unboxing

We received this kit from the team at Arduino, who have cooked up a very special Arduino carrier board and 8 beginner IoT projects including remote-controlled lighting, a solar system tracker, home security alarm, and more – to get you connected and coding. Adapt these to your own solutions and deploy them right away, as the kit comes with its own enclosure and peripheral sensors, and will accept an 18650 Li-ion battery.

// https://store.arduino.cc/usa/opla-iot-kit
// https://store.arduino.cc/usa/mkr-wifi-1010
// https://www.hackster.io/news/arduino-s-opla-iot-kit-aims-to-give-your-home-appliances-smarts-ef157310d647
// https://www.hackster.io/news/the-10-most-common-hardware-hackathon-projects-8e2fc66ee9b7

Categories
Hackster.io

“Doctor Who” HiFive Inventor Coding Kit

Wow! This new kit from BBC Learning, SiFive, and Tynker comes with lessons narrated by Jodie Whittaker – the newest Doctor Who – herself!

// https://www.hackster.io/news/bbc-announces-risc-v-powered-doctor-who-themed-hifive-inventor-educational-microcontroller-kit-8dbffb7a7adb
// https://www.hifiveinventor.com
// https://www.theregister.com/2020/11/19/bbc_doctor_who_sifive
// https://www.sifive.com/documentation
// https://www.hackster.io/search?i=projects&q=minecraft

Categories
Hackster.io

Nordic nRF9160 DK // Unboxing

We check out this cool kit from Nordic: a multi-sensor cellular IoT prototyping platform for hardware engineers. Easily connect Arduino shields and standalone sensors to the nRF Connect for Cloud platform (nrfcloud.com).

Where the Thingy:91 device (previously: https://www.youtube.com/watch?v=tLUKgDT2V9g) comes with built-in sensors and a consumer-ready interface, the nRF9160 DK empowers you to prototype apps with your own custom hardware.

nRF9160 DK materials:
// https://www.nordicsemi.com/Software-and-tools/Development-Kits/nRF9160-DK
// https://www.nordicsemi.com/Products/Low-power-cellular-IoT/nRF9160

Thingy:91 materials:
// https://www.nordicsemi.com/Software-and-tools/Prototyping-platforms/Nordic-Thingy-91/
// https://www.hackster.io/glowascii/getting-started-with-the-nordic-thingy-91-mac-8d44e5

Categories
ScienceDaily

Carbyne: An unusual form of carbon

Which photophysical properties does carbyne have? This was the subject of research carried out by scientists at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), the University of Alberta, Canada, and the Ecole Polytechnique Fédérale de Lausanne in Switzerland, which has led to a greater understanding of the properties of this unusual form of carbon. Their findings have now been published in the latest edition of the journal Nature Communications.

‘Carbon has a very special status in the periodic table of the elements and forms the basis for all forms of life due to the extremely large number of chemical compounds it can form,’ explains Prof. Dr. Dirk M. Guldi at the Chair of Physical Chemistry I at FAU. ‘The most well-known examples are three-dimensional graphite and diamond. However, two-dimensional graphene, one-dimensional nanotubes and zero-dimensional nanodots also open up new opportunities for electronics applications in the future.’

Material with extraordinary properties

Carbyne is a modification of carbon, known as an allotrope. It is manufactured synthetically, comprises one single and very long chain of carbon atoms, and is regarded as a material with extremely interesting electronic and mechanical properties. ‘However, carbon has a high level of reactivity in this form,’ emphasises Prof. Dr. Clémence Corminboef from EPFL. ‘Such long chains are extremely unstable and thus very difficult to characterise.’

Despite this fact, the international research team successfully characterised the chains using a roundabout route. The scientists led by Prof. Dr. Dirk M. Guldi at FAU, Prof. Dr. Clémence Corminboeuf, Prof. Dr. Holger Frauenrath from EPFL and Prof. Dr. Rik R. Tykwinski from the University of Alberta questioned existing assumptions about the photophysical properties of carbyne and gained new insights.

During their research, the team mainly focused on what are known as oligoynes. ‘We can manufacture carbyne chains of specific lengths and protect them from decomposition by adding a type of bumper made of atoms to the ends of the chains. This class of compound has sufficient chemical stability and is known as an oligoyne,’ explains Prof. Dr. Holger Frauenrath from EPFL.

Using the optical band gap

The researchers specifically manufactured two series of oligoynes with varying symmetries and with up to 24 alternating triple and single bonds. Using spectroscopy, they subsequently tracked the deactivation processes of the relevant molecules from excitation with light up to complete relaxation. ‘We were thus able to determine the mechanism behind the entire deactivation process of the oligoynes from an excited state right back to their original initial state and, thanks to the data we gained, we were able to make a prediction about the properties of carbyne,’ concludes Prof. Dr. Rik R. Tykwinski from the University of Alberta.

One important finding was the fact that the so-called optical band gap is actually much smaller than previously assumed. Band gap is a term from the field of semiconductor physics and describes the electrical conductivity of crystals, metals and semiconductors. ‘This is an enormous advantage,’ says Prof. Guldi. ‘The smaller the band gap, the less energy is required to conduct electricity.’ Silicon, for example, which is used in microchips and solar cells, possesses this important property. Carbyne could be used in conjunction with silicon in the future due to its excellent photophysical properties.

Story Source:

Materials provided by University of Erlangen-Nuremberg. Note: Content may be edited for style and length.

Go to Source
Author:

Categories
ScienceDaily

System brings deep learning to ‘internet of things’ devices

Deep learning is everywhere. This branch of artificial intelligence curates your social media and serves your Google search results. Soon, deep learning could also check your vitals or set your thermostat. MIT researchers have developed a system that could bring deep learning neural networks to new — and much smaller — places, like the tiny computer chips in wearable medical devices, household appliances, and the 250 billion other objects that constitute the “internet of things” (IoT).

The system, called MCUNet, designs compact neural networks that deliver unprecedented speed and accuracy for deep learning on IoT devices, despite limited memory and processing power. The technology could facilitate the expansion of the IoT universe while saving energy and improving data security.

The research will be presented at next month’s Conference on Neural Information Processing Systems. The lead author is Ji Lin, a PhD student in Song Han’s lab in MIT’s Department of Electrical Engineering and Computer Science. Co-authors include Han and Yujun Lin of MIT, Wei-Ming Chen of MIT and National University Taiwan, and John Cohn and Chuang Gan of the MIT-IBM Watson AI Lab.

The Internet of Things

The IoT was born in the early 1980s. Grad students at Carnegie Mellon University, including Mike Kazar ’78, connected a Cola-Cola machine to the internet. The group’s motivation was simple: laziness. They wanted to use their computers to confirm the machine was stocked before trekking from their office to make a purchase. It was the world’s first internet-connected appliance. “This was pretty much treated as the punchline of a joke,” says Kazar, now a Microsoft engineer. “No one expected billions of devices on the internet.”

Since that Coke machine, everyday objects have become increasingly networked into the growing IoT. That includes everything from wearable heart monitors to smart fridges that tell you when you’re low on milk. IoT devices often run on microcontrollers — simple computer chips with no operating system, minimal processing power, and less than one thousandth of the memory of a typical smartphone. So pattern-recognition tasks like deep learning are difficult to run locally on IoT devices. For complex analysis, IoT-collected data is often sent to the cloud, making it vulnerable to hacking.

“How do we deploy neural nets directly on these tiny devices? It’s a new research area that’s getting very hot,” says Han. “Companies like Google and ARM are all working in this direction.” Han is too.

With MCUNet, Han’s group codesigned two components needed for “tiny deep learning” — the operation of neural networks on microcontrollers. One component is TinyEngine, an inference engine that directs resource management, akin to an operating system. TinyEngine is optimized to run a particular neural network structure, which is selected by MCUNet’s other component: TinyNAS, a neural architecture search algorithm.

System-algorithm codesign

Designing a deep network for microcontrollers isn’t easy. Existing neural architecture search techniques start with a big pool of possible network structures based on a predefined template, then they gradually find the one with high accuracy and low cost. While the method works, it’s not the most efficient. “It can work pretty well for GPUs or smartphones,” says Lin. “But it’s been difficult to directly apply these techniques to tiny microcontrollers, because they are too small.”

So Lin developed TinyNAS, a neural architecture search method that creates custom-sized networks. “We have a lot of microcontrollers that come with different power capacities and different memory sizes,” says Lin. “So we developed the algorithm [TinyNAS] to optimize the search space for different microcontrollers.” The customized nature of TinyNAS means it can generate compact neural networks with the best possible performance for a given microcontroller — with no unnecessary parameters. “Then we deliver the final, efficient model to the microcontroller,” say Lin.

To run that tiny neural network, a microcontroller also needs a lean inference engine. A typical inference engine carries some dead weight — instructions for tasks it may rarely run. The extra code poses no problem for a laptop or smartphone, but it could easily overwhelm a microcontroller. “It doesn’t have off-chip memory, and it doesn’t have a disk,” says Han. “Everything put together is just one megabyte of flash, so we have to really carefully manage such a small resource.” Cue TinyEngine.

The researchers developed their inference engine in conjunction with TinyNAS. TinyEngine generates the essential code necessary to run TinyNAS’ customized neural network. Any deadweight code is discarded, which cuts down on compile-time. “We keep only what we need,” says Han. “And since we designed the neural network, we know exactly what we need. That’s the advantage of system-algorithm codesign.” In the group’s tests of TinyEngine, the size of the compiled binary code was between 1.9 and five times smaller than comparable microcontroller inference engines from Google and ARM. TinyEngine also contains innovations that reduce runtime, including in-place depth-wise convolution, which cuts peak memory usage nearly in half. After codesigning TinyNAS and TinyEngine, Han’s team put MCUNet to the test.

MCUNet’s first challenge was image classification. The researchers used the ImageNet database to train the system with labeled images, then to test its ability to classify novel ones. On a commercial microcontroller they tested, MCUNet successfully classified 70.7 percent of the novel images — the previous state-of-the-art neural network and inference engine combo was just 54 percent accurate. “Even a 1 percent improvement is considered significant,” says Lin. “So this is a giant leap for microcontroller settings.”

The team found similar results in ImageNet tests of three other microcontrollers. And on both speed and accuracy, MCUNet beat the competition for audio and visual “wake-word” tasks, where a user initiates an interaction with a computer using vocal cues (think: “Hey, Siri”) or simply by entering a room. The experiments highlight MCUNet’s adaptability to numerous applications.

“Huge potential”

The promising test results give Han hope that it will become the new industry standard for microcontrollers. “It has huge potential,” he says.

The advance “extends the frontier of deep neural network design even farther into the computational domain of small energy-efficient microcontrollers,” says Kurt Keutzer, a computer scientist at the University of California at Berkeley, who was not involved in the work. He adds that MCUNet could “bring intelligent computer-vision capabilities to even the simplest kitchen appliances, or enable more intelligent motion sensors.”

MCUNet could also make IoT devices more secure. “A key advantage is preserving privacy,” says Han. “You don’t need to transmit the data to the cloud.”

Analyzing data locally reduces the risk of personal information being stolen — including personal health data. Han envisions smart watches with MCUNet that don’t just sense users’ heartbeat, blood pressure, and oxygen levels, but also analyze and help them understand that information. MCUNet could also bring deep learning to IoT devices in vehicles and rural areas with limited internet access.

Plus, MCUNet’s slim computing footprint translates into a slim carbon footprint. “Our big dream is for green AI,” says Han, adding that training a large neural network can burn carbon equivalent to the lifetime emissions of five cars. MCUNet on a microcontroller would require a small fraction of that energy. “Our end goal is to enable efficient, tiny AI with less computational resources, less human resources, and less data,” says Han.

Go to Source
Author:

Categories
ScienceDaily

Chemistry: How nitrogen is transferred by a catalyst

Catalysts with a metal-nitrogen bond can transfer nitrogen to organic molecules. In this process short-lived molecular species are formed, whose properties critically determine the course of the reaction and product formation. The key compound in a catalytic nitrogen-atom transfer reaction has now been analysed in detail by chemists. The detailed understanding of this reaction will allow for the design of catalysts tailored for specific reactions.

The development of new drugs or innovative molecular materials with new properties requires specific modification of molecules. Selectivity control in these chemical transformations is one of the main goals of catalysis. This is particularly true for complex molecules with multiple reactive sites in order to avoid unnecessary waste for improved sustainability. The selective insertion of individual nitrogen atoms into carbon-hydrogen bonds of target molecules is, for instance, a particularly interesting goal of chemical synthesis. In the past, these kinds of nitrogen transfer reactions were postulated based on quantum-chemical computer simulations for molecular metal complexes with individual nitrogen atoms bound to the metal. These highly reactive intermediates have, however, previously escaped experimental observation. A closely entangled combination of experimental and theoretical studies is thus indispensable for detailed analysis of these metallonitrene key intermediates and, ultimately, the exploitation of catalytic nitrogen-atom transfer reactions.

Chemists in the groups of Professor Sven Schneider, University of Göttingen, and Professor Max Holthausen, Goethe University Frankfurt, in collaboration with the groups of Professor Joris van Slagern, University of Stuttgart and Professor Bas de Bruin, University of Amsterdam, have now been able for the first time to directly observe such a metallonitrene, measure it spectroscopically and provide a comprehensive quantum-chemical characterization. To this end, a platinum azide complex was transformed photochemically into a metallonitrene and examined both magnetometrically and using photo-crystallography. Together with theoretical modelling, the researchers have now provided a detailed report on a very reactive metallonitrene diradical with a single metal-nitrogen bond. The group was furthermore able to show how the unusual electronic structure of the platinum metallonitrene allows the targeted insertion of the nitrogen atom into, for example, C-H bonds of other molecules.

Professor Max Holthausen explains: “The findings of our work significantly extend the basic understanding of chemical bonding and reactivity of such metal complexes, providing the basis for a rational synthesis planning.” Professor Sven Schneider says: “These insertion reactions allow the use of metallonitrenes for the selective synthesis of organic nitrogen compounds through catalyst nitrogen atom transfer. This work therefore contributes to the development of novel ‘green’ syntheses of nitrogen compounds.”

Story Source:

Materials provided by Goethe University Frankfurt. Note: Content may be edited for style and length.

Go to Source
Author:

Categories
ScienceDaily

Chemists discover the structure of a key coronavirus protein

MIT chemists have determined the molecular structure of a protein found in the SARS-CoV-2 virus. This protein, called the envelope protein E, forms a cation-selective channel and plays a key role in the virus’s ability to replicate itself and stimulate the host cell’s inflammation response.

If researchers could devise ways to block this channel, they may be able to reduce the pathogenicity of the virus and interfere with viral replication, says Mei Hong, an MIT professor of chemistry. In this study, the researchers investigated the binding sites of two drugs that block the channel, but these drugs bind only weakly, so they would not be effective inhibitors of the E protein.

“Our findings could be useful for medicinal chemists to design alternative small molecules that target this channel with high affinity,” says Hong, who is the senior author of the new study.

MIT graduate student Venkata Mandala is the lead author of the paper, which appears in Nature Structural and Molecular Biology. Other authors include MIT postdoc Matthew McKay, MIT graduate students Alexander Shcherbakov and Aurelio Dregni, and Antonios Kolocouris, a professor of pharmaceutical chemistry at the University of Athens.

Structural challenges

Hong’s lab specializes in studying the structures of proteins that are embedded in cell membranes, which are often challenging to analyze because of the disorder of the lipid membrane. Using nuclear magnetic resonance (NMR) spectroscopy, she has previously developed several techniques that allow her to obtain accurate atomic-level structural information about these membrane-embedded proteins.

When the SARS-CoV-2 outbreak began earlier this year, Hong and her students decided to focus their efforts on one of the novel coronavirus proteins. She narrowed in on the E protein partly because it is similar to an influenza protein called the M2 proton channel, which she has previously studied. Both viral proteins are made of bundles of several helical proteins.

“We determined the influenza B M2 structure after about 1.5 years of hard work, which taught us how to clone, express, and purify a virus membrane protein from scratch, and what NMR experimental strategies to take to solve the structure of a homo-oligomeric helical bundle,” Hong says. “That experience turned out to be the perfect training ground for studying SARS-CoV-2 E.”

The researchers were able to clone and purify the E protein in two and half months. To determine its structure, the researchers embedded it into a lipid bilayer, similar to a cell membrane, and then analyzed it with NMR, which uses the magnetic properties of atomic nuclei to reveal the structures of the molecules containing those nuclei. They measured the NMR spectra for two months, nonstop, on the highest-field NMR instrument at MIT, a 900-megahertz spectrometer, as well as on 800- and 600-megahertz spectrometers.

Hong and her colleagues found that the part of the E protein that is embedded in the lipid bilayer, known as the transmembrane domain, assembles into a bundle of five helices. The helices remain largely immobile within this bundle, creating a tight channel that is much more constricted than the influenza M2 channel.

Interestingly, the SARS-CoV-2 E protein looks nothing like the ion channel proteins of influenza and HIV-1 viruses. In flu viruses, the equivalent M2 protein is much more mobile, while in HIV-1, the equivalent Vpu protein has a much shorter transmembrane helix and a wider pore. How these distinct structural features of E affect its functions in the SARS-CoV-2 virus lifecycle is one of the topics that Hong and her colleagues will study in the future.

The researchers also identified several amino acids at one end of the channel that may attract positively charged ions such as calcium into the channel. They believe that the structure they report in this paper is the closed state of the channel, and they now hope to determine the structure of the open state, which should shed light on how the channel opens and closes.

Fundamental research

The researchers also found that two drugs — amantadine, used to treat influenza, and hexamethylene amiloride, used to treat high blood pressure — can block the entrance of the E channel. However, these drugs only bind weakly to the E protein. If stronger inhibitors could be developed, they could be potential drug candidates to treat Covid-19, Hong says.

The study demonstrates that basic scientific research can make important contributions toward solving medical problems, she adds.

“Even when the pandemic is over, it is important that our society recognizes and remembers that fundamental scientific research into virus proteins or bacterial proteins must continue vigorously, so we can preempt pandemics,” Hong says. “The human cost and economic cost of not doing so are just too high.”

The research was funded by the National Institutes of Health and the MIT School of Science Sloan Fund.

Go to Source
Author:

Categories
ProgrammableWeb

Facebook Introduces V9 of the Graph and Marketing APIs

Facebook has announced the release of the Graph API v9.0Track this API and Marketing API v9.0Track this API. These updates included several changes that require developer action in order to ensure that production applications are not interrupted. Additional updates include adjustments to Instagram Follower count metrics and Instagram Ads API enhancements. 

For developers that have applications that are included in the Consumer & Gaming application types, it is important to note that Facebook will now begin requiring that applications include a pathway for users to request the deletion of account information. This can be handled either by a Data Deletion Request Callback or via a URL that provides detailed instructions for how this information can be deleted. Facebook plans to send reminders to developers that have not yet transitioned from Development Mode to Live Mode.

Additionally, Facebook has made changes to the way that third-party applications can access user information and connect CRM systems with Facebook’s APIs. These changes may result in a loss of access to Lead Ads campaign data. When updating to V9 applications that intend to continue receiving access may need to go through app review. 

Facebook also announced improvements to the Instagram Ads API, the announcement noted that:

“We are introducing several technical changes that will now give your users the ability to take existing organic Instagram posts and promote them as ads. We will also begin returning the Instagram Graph API’s user ID on the Instagram Ads API and several other endpoints, offering a more unified developer experience between the Ads and Graph API platforms.”

The company has also adjusted the Instagram follower count metric and noted that users may notice a one-time drop in followers via the API as they align this count with the native app.  

Go to Source
Author: <a href="https://www.programmableweb.com/user/%5Buid%5D">KevinSundstrom</a>

Categories
Hackster.io

nRF9160 Webinar: Your solution for ultra low power cellular applications

LTE-M and NB-IoT enables everything to be connected to the Internet. In this webinar you’ll learn how the nRF9160 System-in-Package realizes this and makes it possible to build very small devices with many years of battery life.

Nordic DevZone: https://devzone.nordicsemi.com/

Categories
Hackster.io

Raspberry Pi Fan SHIM // by Pimoroni

This useful SHIM ("shove hardware in the middle" device) provides active cooling for your Raspberry Pi! While updates to the firmware mean that you no longer *need* cooling (unless you’re trying to overclock your Pi), a little fresh air never hurts. We put together this "whisper-quiet" little peripheral in a couple of minutes, no fuss.

// https://shop.pimoroni.com/products/fan-shim
// https://learn.pimoroni.com/tutorial/sandyj/getting-started-with-fan-shim