Welcome to the September 15, 2021 edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Foster et al. Receive Most Influential Paper Award for Network Programming Language
Cornell Chronicle
David LaRocca
September 10, 2021
This year's International Conference on Functional Programming named research co-authored by Cornell University's Nate Foster the Most Influential Paper for its creation of a network programming language. The awards committee described the 2011 paper on the Frenetic language, designed to address prevailing limits in emergent software-defined networking, as "the first conference paper in a highly influential line of work creating a bridge between programming languages and networking." Frenetic's authors analyzed deficiencies in state-of-the-art languages, resolving them in a new language design, and characterized its deployment and performance on a suite of microbenchmarks. The awards committee said the paper demonstrated "the practical value of improving network programming," which influenced “subsequent advances in the state of the art."
|
DNA-Based Chip Can Be Programmed to Solve Complex Math Problems
Incheon National University (South Korea)
September 14, 2021
Scientists at South Korea's Incheon National University (INU) have automated mathematical DNA calculations via a personal computer (PC)-controlled chip. The three-dimensionally-printed microfluidic processing unit can execute Boolean logic, with a single-stranded DNA template serving as a logic gate while different single-stranded DNA act as inputs. If part of an input DNA possesses a complementary Watson-Crick sequence to the template DNA, it pairs into double-stranded DNA whose size signals the output as true or false. INU's Youngjun Song said, "Our hope is that DNA-based CPUs [central-processing units] will replace electronic CPUs in the future because they consume less power, which will help with global warming. DNA-based CPUs also provide a platform for complex calculations like deep learning solutions and mathematical modeling."
|
Coughs Say Much About Your Health, If Your Smartphone Is Listening
The Wall Street Journal
Betsy McKay
September 8, 2021
Many researchers are trying to train artificial intelligence (AI) to enable smartphones to analyze patterns in a person’s coughs, as a means of diagnosing and treating deadly respiratory diseases like tuberculosis (TB) and COVID-19. Peter Small at acoustic AI developer Hyfe said a cough's sound and frequency contain valuable information, because coughs resulting from different diseases produce audible distinctions. Hyfe and other developers of what Small calls acoustic epidemiology are compiling datasets and training AI to recognize the different types of coughs. Hyfe offers free smartphone applications for consumers and researchers that use AI to detect and track how frequently someone coughs. The apps operate continuously on a smartphone, recording half-second clips when the AI perceives a cough sound. The University of California, San Francisco's Adithya Cattamanchi is using the researcher app to record coughs in several countries, as the first steps toward building a TB cough database.
*May Require Paid Registration
|
Researchers Generated an Entire Virtual Universe; You Can Even Download It
ScienceAlert (Australia)
Brian Koberlein
September 13, 2021
An international team of researchers has generated the most detailed model of the cosmos to date, composed of 2.1 trillion "particles" across 9.6 billion light-years. The Uchuu simulation models 13 billion years of cosmological evolution by focusing solely on the behavior of dark matter within an expanding universe, with over 40,000 computer cores and 20 million computer hours invested in it. The team compressed more than 3 petabytes of data generated by the model into 100 terabytes of storage, which can be retained on a single drive. The team made this raw data freely available on skiesanduniverses.org so researchers can explore the model while scientists datamine its results.
|
New Tool for Analyzing Large Superconducting Circuits
Northwestern Now
Megan Fellman
September 13, 2021
Large superconducting circuits can be analyzed with a new theoretical tool developed by Northwestern University researchers. Northwestern's Daniel Weiss said, "Our framework is inspired by methods originally developed for the study of electrons in crystals, and allows us to obtain quantitative predictions for circuits that were previously hard or impossible to access." The researchers used the tool to extract quantitative information from a protected circuit, and the method complements existing tools with a similar purpose for when they yield suboptimal results. This work dovetails with the missions of the U.S. Department of Energy's Superconducting Quantum Materials and Systems Center and Co-design Center for Quantum Advantage. Their respective goals are building a superior superconducting technologies-based quantum computer, and constructing basic tools for assembling scalable, distributed, and fault-tolerant quantum computers.
|
Don't Fidget! Wi-Fi Will Count You
The Current (University of California, Santa Barbara)
Sonia Fernandez
September 13, 2021
University of California, Santa Barbara (UCSB) researchers used Wi-Fi signals to estimate the number of people in a stationary seated crowd without needing them to carry a device. The method simply requires a wireless transmitter and receiver outside the area of interest. The transmitter emits a signal and the receiver quantifies its received power; the resulting calculation of the number of people in the specified area closely matches the actual figure. UCSB's Yasamin Mostofi said the technique overcomes the problem of counting people who make few major body motions, who often engage in small in-place natural body movements called fidgets. The methodology statistically extracts a stationary crowd's collective fidgeting behavior, and links this to the total number of seated people.
|
Apple Issues Emergency Security Updates to Close Spyware Flaw
The New York Times
Nicole Perlroth
September 13, 2021
Apple has released emergency security patches after researchers at the University of Toronto's Citizen Lab in Canada found a flaw that allows spyware from Israel's NSO Group to infect any Apple product. The researchers discovered a Saudi activist's iPhone had been hacked with NSO's Pegasus spyware using a zero-click remote exploit that can infect targets without their awareness, commandeer device functions, and transmit information to NSO clients. More than 1.65 billion Apple products in use globally have been susceptible to the spyware since at least March. Apple said it intends to launch new security defenses for its iMessage texting application in its next iOS 15 software update, expected later this year.
*May Require Paid Registration
|
Italy Data Authority Asks Facebook for Clarifications on Smart Glasses
Reuters
Elvira Pollina
September 10, 2021
Italian data protection authority Garante has asked Facebook to provide it with clarifications related to its newly launched smart glasses, to determine the product's compliance with that nation’s privacy laws. Developed with Ray-Ban manufacturer EssilorLuxottica, the Ray-Ban Stories glasses allow users to hear music, take calls, or shoot photos and videos and share them across Facebook's services via a companion application. Garante made its request via the Irish Data Protection Commissioner (DPC), which oversees Facebook because the social-media giant's European headquarters are based in Ireland. The regulator said it wanted clarification on measures Facebook has deployed to shield people occasionally filmed, especially children, and on systems adopted to anonymize collected data, and features of the smartglasses' voice assistant.
|
The Beauty of OpenRooms
UC San Diego News Center
Josh Baxt
September 9, 2021
University of California, San Diego (UCSD) computer scientists have released OpenRooms, a dataset and tool suite to help users manipulate objects, materials, lighting, and other properties of indoor three-dimensional (3D) settings, for use in augmented reality and robotic applications. OpenRooms uses synthetic data to furnish ground truth geometry, materials, and lighting accurately and inexpensively. The data can be used to train deep neural networks that calculate those properties in real images, enabling photorealistic object insertion and material editing. Said UCSD's Manmohan Chandraker, "We are creating a framework where users can use their cellphones or 3D scanners for developing datasets that enable their own augmented reality applications. They can simply use scans or sets of photographs."
|
Researchers Demonstrate Extreme Heat Exchanger with Additive Manufacturing
University of Illinois Grainger College of Engineering
September 9, 2021
Extreme high-performance heat exchangers produced via additive manufacturing were designed using shape optimization software developed by researchers at the University of Illinois Urbana Champaign (UIUC) Grainger College of Engineering. UIUC's William King said, "The software allows us to identify 3D [three-dimensional] designs that are significantly different and better than conventional designs." The team combined the software with metal 3D printing to fabricate and test an optimized tube-in-tube heat exchanger. UIUC's Nenad Miljkovic said the device has approximately 20 times the volumetric power density of a state-of-the-art commercial counterpart.
|
Real-Time Intent Monitoring Could Enable Safer Use of Drones
University of Cambridge (U.K.)
September 15, 2021
A system for monitoring drones' intended flight paths could prevent disruptions like the 2018 incident at London's Gatwick Airport. Scientists at the U.K.'s University of Cambridge combined statistics and radar data to predict drone flight paths and intent in real time, and to enable automated decision-making based on that information. The team applied Bayesian statistics to build a system that only flags and prioritizes drones posing a threat, using a stochastic model to determine the aircraft's underlying intent. Tests using real radar data showed the system could identify drones before they reached their next waypoint, and forecast the likelihood of any given drone reaching the next waypoint. Cambridge's Bashar Ahmad said, "It's crucial for future drone surveillance systems to have predictive capabilities for revealing, as early as possible, a drone with malicious intent or anomalous behavior."
|
Portable Analyzer Brings Blood Testing to Rural Areas
IEEE Spectrum
Michelle Hampson
September 14, 2021
Sangeeta Palekar and Jayu Kalambe at India's Shri Ramdeobaba College of Engineering and Management built a portable blood analyzer that can be easily deployed to rural regions. The device features an automated fluid dispenser that adds a controlled amount of reagent to a blood sample, which then is exposed to light while a Raspberry Pi computer analyzes the data. The analyzer can produce results within 30 seconds, and its data almost perfectly matchs data generated by standard laboratory equipment. Palekar said the analyzer "offers the advantages of automation, low cost, portability, simple instrumentation, flexibility, and an easily accessible interface."
|
AI Can Detect Deepfake Face Because Its Pupils Have Jagged Edges
New Scientist
Chris Stokel-Walker
September 10, 2021
A computer model developed by researchers at New York’s University of Albany can determine whether an image of a face is a deepfake by examining its pupils; the model will deem the image a fake if the pupils are not circular or elliptical. If the image passes that test, the model will check whether the pupil has smooth or jagged edges, with the latter indicating a deepfake. University of Albany's Siwei Lyu said, "Even though [generative adversarial network] GAN models are very powerful, they don't really understand human biology very well. A lot of these very fine details won't be represented by the model effectively." Although the shape of one's pupils can be affected by certain diseases and infections, Lyu noted that such cases are rare.
*May Require Paid Registration
|
|
