Welcome to the May 8, 2023, 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."
|
|
|
Algorithm Makes Satellite Signals Act Like GPS
Ohio State News
Tatyana Woodall
May 5, 2023
An algorithm developed by Ohio State University (OSU) researchers can listen in on satellite signals and use them to locate any point on Earth. The algorithm eavesdropped on signals from eight low Earth orbit (LEO) satellites for approximately 10 minutes, after which it was able to locate and zero in on a stationary ground receiver with an error of only about 5.8 meters (19 feet). The researchers required no assistance from satellite operators, noting they only had access to publicly available information relevant to the satellites' downlink transmission frequency and a rough estimate of their locations. OSU's Zak Kassas said the algorithm is LEO constellation-agnostic, meaning it can use practically any satellite signal.
|
Electronic Cochlea Adjusts to Noise Levels Like Human Ears
New Scientist
Alex Wilkins
May 4, 2023
Scientists from Germany's Ilmenau University of Technology (TU Ilmenau) and Kiel University designed a miniature electrical sensor that mimics the human ear's cochlea by automatically adjusting its sensitivity to different frequencies. TU Ilmenau's Claudia Lenk said, "It's the first real sensor which integrates the processing that happens in our inner ear directly into the sensor. That makes it quite efficient and faster than typical set-ups that were developed before." The approximately 350x150-micrometer sensor uses a strip of silicon to render sound waves as electrical signals of different frequencies, while a connected actuator can adjust the sensor's response to sound.
|
Japan Firms Turn to VR for Customer Service Training
The Japan News
May 8, 2023
Increasingly, Japanese companies are using virtual reality (VR) technology to train customer service staff. Employees of convenience store chain operator FamilyMart wear VR goggles and manipulate a controller to learn various tasks, like using cash registers, ordering products, and accepting delivery orders. VR enables trainees to learn on their own, shortening training time from 36 hours to 12. In January, Kyoto College of Hotel, Tourism, and Bridal Management developed a VR-based training program for hotels and inns featuring situations that change according to trainees' responses, so they can manage various scenarios while engaging with different guests. Japan's manufacturing, automobile, and railway sectors are using VR for inspection and repair, while a labor storage has prompted that nation’s food service and lodging industries to roll out VR training to enhance efficiency.
|
Inspection Drone Uses Wind to Lengthen Flight Times
EPFL (Switzerland)
May 5, 2023
Scientists at Swiss Federal Institute of Technology, Lausanne spin-off Elythor created an inspection drone whose wings can adapt to wind conditions and flight position in order to extend its time in the air. The Morpho hybrid unmanned aerial vehicle combines wings and quadcopter design with sensors and cameras, and can fly equally well in enclosed spaces and open air. Morpho is able to take off vertically and rotate into a horizontal attitude, expanding or contracting its wings in response to strong winds. Elythor's Nathan Müller said Morpho's controller automatically chooses when to keep the wings stationary or to let them move freely with the wind, according to trajectory, speed, and wind direction.
|
Tech Increases Equity in Diabetes Care for Kids
Stanford University Institute for Human-Centered AI
Katherine Miller
May 1, 2023
The 4T Program initiated by Stanford University researchers found technology can democratize diabetes care for children. The program offered every T1D-diagnosed child a continuous glucose monitor (CGM) and associated training within one month of diagnosis, as well as remote glucose monitoring services and artificial intelligence-prioritized dose adjustment. Analysis of the results of the program found glycemic control improved for participants across all socioeconomic backgrounds, while remote monitoring also helped (as long as patients had access to continuous cloud connectivity and smart devices). However, Stanford's Ananta Addala cautioned, "Those who are on public health insurance or are members of historically minoritized groups still lag behind their counterparts."
|
3D Mapping of West Arnhem Flood Plain Could Be 'Game Changer'
ABC News (Australia)
Lillian Rangiah
May 4, 2023
Researchers at Australia's Flinders University uncovered a hidden coastal landscape buried under layers of sediment in the inland Red Lily Lagoon flood plain in West Arnhem, Australia, using drones, laser, and subsurface imaging techniques. The researchers created a three-dimensional (3D) model to showcase the landscape. Said Flinders' Jarrad Kowlessar, "The data is one thing, but to then visualize it, model it in 3D, and be able to explore those environments means that we can understand the data in new ways and understand what life might have been like." Flinders' Ian Moffat called the technique a "game changer" for archaeological research, explaining, "Instead of focusing on the archaeological sites—which is the way we normally think about archaeology—we've really stepped out and tried to understand the landscape in a much more holistic way."
|
Researchers Detect, Classify Multiple Objects Without Images
Optica
May 3, 2023
A technique developed by researchers at China's Beijing Institute of Technology can detect, classify, and track multiple objects without requiring images or complex scene reconstruction. Known as image-free single-pixel object detection (SPOD), the technique can perform object recognition, classification, and tracking simultaneously with less computing power than traditional approaches. SPOD involves the use of a small but optimized structured light pattern to scan the scene and take two-dimensional measurements. Important scene features are extracted from those measurements by a transformer-based encoder, after which a multi-scale attention network-based decoder is used to simultaneously output the class, location, and size of all targets in the scene. The researchers found SPOD was faster than performing scene reconstruction followed by object detection.
|
Monitoring Health Through Sweat
University of Hawaii News
May 3, 2023
University of Hawaii (UH) researchers have developed a three-dimensionally (3D)-printed wearable sweat sensor for health monitoring. The "sweatainer" collects perspiration via a "multi-draw" technique, analyzing multiple sweat samples either directly or at a laboratory. This enhances the efficiency of sweat collection while also unlocking potential for at-home testing, storing samples for future experiments, and combining with current health monitoring methods. UH's Tyler Ray said, "3D printing enables an entirely new design mode for wearable sweat sensors by allowing us to create fluidic networks and features with unprecedented complexity. With the sweatainer, we are utilizing 3D printing to showcase the vast opportunities this approach enables for accessible, innovative, and cost-effective prototyping of advanced wearable sweat devices."
|
Nvidia GPU Speeds Past Quantum Computer
The Register (U.K.)
Tobias Mann
May 3, 2023
Scientists from Microsoft and Switzerland's Scalable Parallel Computing Laboratory (SPCL) found commercially sold graphics processing units (GPUs) sometimes surpass cutting-edge quantum computers. SPCL's Torsten Hoefler, Microsoft's Matthias Troyer, and former Microsoft researcher Thomas Häner tested a hypothetical quantum computer with 10,000 error-correcting quantum bits against a conventional computer with a single Nvidia A100 GPU. The results indicated a quadratic speedup is insufficient to achieve superiority over classical computers, which cannot be realized without "super-quadratic or ideally exponential speedups." Troyer said classical computing better serves large dataset-reliant applications because quantum-system bandwidth currently is too low to accommodate them. "Generally, quantum computers will be practical for 'big compute' problems on small data, not big data problems," according to the researchers.
|
Tool Accurately Simulates Complex Systems
MIT News
Adam Zewe
May 4, 2023
A tool developed by researchers at the Massachusetts Institute of Technology (MIT) eliminates bias in trace-driven simulations of new algorithms. Trace-driven simulation involves collecting a small amount of real data (traces) that is replayed while simulating a complex system; the resulting bias could lead to a worse-performing algorithm being selected. CasualSim helps researchers choose the best algorithm of those tested by estimating the underlying functions that produced the trace data collected through a randomized control trial and indicating how a new algorithm would change the outcome under the same underlying conditions. MIT's Mohammad Alizadeh said it is important "to understand what aspects of the behavior we are seeing are intrinsic to the system and how much of what we are observing is based on the actions that were taken. If we can disentangle these two effects, then we can do unbiased simulations."
|
Sick of Bumpy, Delayed Flights? Weather Tech Could Help
The New York Times
Julie Weed
May 5, 2023
The travel industry hopes to reduce the number of turbulent and delayed by using artificial intelligence (AI)-powered sensors, satellites, and data models to produce more accurate weather forecasts. The U.S. Air Force allocated $19 million to weather intelligence company Tomorrow.io to deploy more than 20 satellites by 2025 for hourly global meteorological reporting, which the company's Dan Slagen said will give pilots more comprehensive weather information. Meanwhile, Florida-based WeatherSTEM uses AI to make flight recommendations to airports based on weather data analysis. The company also sets up hyperlocal weather stations that may include mini-Doppler radar systems that show rainfall in greater detail, solar-powered wind speed and direction sensors, and digital video cameras.
*May Require Paid Registration
|
LED, Holographic Microscope Converts Mobile Phone Cameras into High-Resolution Microscopes
Singapore-MIT Alliance for Research and Technology
May 4, 2023
Researchers at the Singapore-MIT Alliance for Research and Technology (SMART) created what they’re calling the world's smallest light emitting diode (LED), and used it to develop what they’re calling the world’s smallest holographic microscope. The researchers said the LED could be used with existing mobile phone cameras to function as high-resolution microscopes. The researchers also said they had developed a neural networking algorithm that can reconstruct objects measured by the holographic microscope. SMART's Rajeev Ram said, "Our LED could be ideal for bio-imaging and bio-sensing applications, including near-field microscopy and implantable CMOS devices. Also, it is possible to integrate this LED with on-chip photodetectors, and it could then find further applications in on-chip communication, NIR proximity sensing, and on-wafer testing of photonics."
|
|
