Virginia Tech researchers publish revolutionary blueprint to fuse wireless technologies and AI

Virginia Tech

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

“We could actually overcome some of the current network limitations, unleashing a completely new era of wireless networking. It is a win-win strategy for the wireless and AI evolutions," Walid Saad said.

view more 

Credit: Photo by Peter Means for Virginia Tech.

There’s a major difference between humans and current artificial intelligence (AI) capabilities: common sense. According to a new visionary paper by Walid Saad, professor in the College of Engineering and the Next-G Wireless Lead at the Virginia Tech Innovation Campus, a true revolution in wireless technologies is only possible through endowing the system with the next generation of AI that can think, imagine, and plan akin to humans.

Published in the Proceedings of the IEEE Journal's Special Issue on the Road to 6G with Ph.D. student Omar Hashash and postdoctoral associate Christo Thomas, the paper's findings suggest:

• The missing link in the wireless revolution is next-generation AI.
• The missing link in the next generation of AI is wireless technologies.
• The solution is to bring AI closer to human intelligence through common sense.

“We’re looking at least 10 or 15 years down the line before we have a wireless network with artificial general intelligence [AGI] that can think, plan, and imagine,” said Saad, who is a professor in the Bradley Department of Electrical and Computer Engineering. “We have a blueprint and concrete road map. The entire vision might not be immediately deployable, but pieces of it can be implemented now. We're trying to position this paper in a way to tell the community that there is a path to something really revolutionary — step by step we can work toward a living, thinking wireless network.”

Previous generations of wireless networks have been defined by several enhancements to core components, such new antennas and communication technologies that have improved performance. According to the researchers, not even the leap from 5G to 6G, characterized by the addition of an AI-architecture embedded in wireless systems and an open radio access network, will be revolutionary enough to meet future processing and networking needs. 

“That is where things start to become thrilling,” Hashash said. “The next generation of wireless networks and AI are converging hand in hand, but few are seeing how they can actually be seamlessly merged.”

Physical networks endowed with AI

At first, Saad, Hashash, and Thomas were focusing on the metaverse and building on what is currently being explored in 6G: embedding AI across wireless systems, referred to as AI-native networks. 

“The problem is researchers are using classical AI tools that are designed for other tasks such as computer vision, which makes them limited in many ways when it comes to communication networks,” Saad said. “To fuse the real world with a virtual world, you basically have to mirror it. This is not something that old school AI can do.”

Although humans develop common sense through a world model and understanding the intuitive physics of the environment, current AI systems are trained on data. They tend to extract patterns and capture underlying correlative relationships, but these AI systems cannot reasonably navigate unforeseen scenarios. In its next phase, 6G hopes to overcome this narrow, statistical, rule-based AI solution so that the network improves its sustainability, generalizability, trustworthiness, and explainability. To date, we still do not have an AI system that can deal with new and unfamiliar scenarios, because they lack an important human trait: common sense.

“Common sense allows us to deal with new scenarios, learn by analogy, and connect the dots to fill in missing plausible elements when needed,” Saad said. “Simply put, the current level of AI is good at extracting statistical relationships from data, but it’s very bad at reasoning and generalizing to novel, unexpected situations – things that most humans master perfectly.”

To blend the physical, virtual, and digital dimensions seamlessly — for example, to don a virtual reality headset and “travel” across space and time from the comfort of home — the next generation of wireless systems will need extreme wireless quality-of-service requirements for perfectly synchronizing worlds. It also will need a hyper form of AI that can enable the network to seamlessly orchestrate the physical, virtual, and digital dimensions, something that only a real human-like network can achieve. 

In a nutshell, one of the challenges for the next generation of wireless beyond 6G is not only the physical constraints of the wireless technologies, but the limited capacities of current AI technologies. 

The telecom brain

As the research continued, team members were surprised to find they were not only building on their previous wireless studies, but that their research converged with optimistic advancements in AI toward human-level intelligence. 

“On the one hand, the metaverse with its digital world can enable a real-time perception of the physical world, which is an essential factor to enable AGI-native networks,” Hashash said. “On the other hand, the metaverse promises to bring forth novel use cases and applications such as cognitive avatars that require common sense abilities.”

The metaverse, with emerging applications such as digital twins and its ability to have identical digital representations of the physical world, could provide crucial opportunities for networks to acquire perception, hyperdimensional world models, planning abilities, and analogical reasoning. This architecture would provide the missing link that would make it a real “brain,” equipping the network to handle unforeseen obstacles and predict new scenarios outside of its training data. 

“We must create wireless networks with intrinsic abilities to understand the mathematical mechanics behind the designed AI models, the physical properties of real-world objects, and their interactions with each other,” Thomas said. “This requires us to fuse mathematical principles, category theory, and neuroscience to model the physical world and understand the complex operations of the human brain. We are therefore advocating for revisiting the fundamentals in AI, wireless networks, mathematics, and neuroscience.”

Instead of incremental advances to known wireless technologies, the researchers propose a paradigm shift. This shift would surpass the AI-native wireless system anticipated with 6G and aspire to a system that is equipped with human levels of intelligence — intelligence that is learned at the intersection of the digital world  and future wireless networks Then, this AGI-native network is set to endow some of its common sense abilities to digital twins, thereby unleashing a new breed of human-like AI agents.

“The missing link is really the wireless network and its components like digital twins, because we can use a twin that exists as a basis for a world model thereby enabling human-level-like thinking and integrating these 'thought' processes in the wireless network now,” Saad said. “We could actually overcome some of the current network limitations, unleashing a completely new era of wireless networking. It is a win-win strategy for the wireless and AI evolutions.”

Original study DOI: 10.1109/JPROC.2025.3526887

---

Journal

Proceedings of the IEEE

Article Title

Artificial General Intelligence (AGI)-Native Wireless Systems: A Journey Beyond 6G

Article Publication Date

17-Mar-2025

 

Wayne State University research making strides in autonomous vehicle and machine systems to make them safer, more effective

Wayne State University - Office of the Vice President for Research

Facebook

X
BlueskyMessageWhatsAppEmail

 

image: 

Zheng Dong, Ph.D., assistant professor of computer science in Wayne State’s School of Engineering, was recently awarded a prestigious CAREER NSF award.

view more 

Credit: Julie O'Connor, Wayne State University

DETROIT — A grant to Wayne State University from the National Science Foundation (NSF) is opening new doors for researchers and students to explore the future of autonomous vehicles, machines and drones.

Zheng Dong, Ph.D., assistant professor of computer science in Wayne State’s School of Engineering, was awarded a five-year, $595,611 NSF grant for the project, "CAREER: ChronosDrive: Ensuring Timing Correctness in DNN-Driven Autonomous Vehicles with Accelerator-Enhanced Real-Time SoC Integration."

“We are in an age of artificial intelligence,” said Dong. “Deep neural networks and autonomous vehicles are opening new frontiers in real-time systems research, which demands new solutions to ensure these systems are safe and effective.”

Autonomous machines, particularly those powered by deep neural networks (DNNs) in autonomous vehicles, must meet strict timing requirements that necessitate rigorous real-time safety certifications. Such certifications rely on advanced analytical methods combining worst-case execution time analysis with schedulability analysis to ensure operational safety and reliability. However, significant challenges persist in integrating worst-case execution time and schedulability analysis, especially in evaluating the timing accuracy of systems that use computing accelerators for autonomous driving.

Dong’s project aims to develop an integrated architecture that leverages hardware-software co-design to address these complex issues with the aim of significantly enhancing the safety and reliability of autonomous driving systems and other autonomous machines.

“Even though we are talking about artificial intelligence, advancements in this field still rely on human innovation and creativity,” said Dong. “We hope our research lays a solid foundation for the development of safe and effective autonomous vehicle and machine systems.”

“NSF CAREER awards are prestigious honors to rising researchers that emphasize the integration of research and education,” said Ezemenari M. Obasi, Ph.D., Wayne State’s vice president for research & innovation. “Dr. Dong’s important research will incorporate AI to address the complex problems associated with autonomous driving systems and machines. Through this award, he will train computer science and engineering students to create the next generation of autonomous machines and vehicles to be safer and more reliable.”

The grant number for this award from the National Science Foundation is 2441179.

# # #

 

Twenty-two year study: Adolescents engaged in fewer external risky behaviors but some report increasing mental health concerns

Boston College

FacebookX

LinkedInWeChatBlueskyMessageWhatsAppEmail

Between 1999-2021, U.S. adolescents steadily desisted from risky behaviors such as substance use and violence, and from reporting a combination of both risky behaviors and mental health symptoms. Yet a comparatively small but growing proportion of youth demonstrated elevated symptoms of depression, according to a report to be published in the April 2025 issue of Pediatrics.

The study, published online on March 18, and titled “Trends in Mental and Behavioral Health Risks in Adolescents: 1999-2021,” analyzed data from the national biennial Youth Risk Behavior Surveys distributed by the Centers for Disease Control and Prevention.

A total of 178,658 students in the 9th-12th grades nationwide were analyzed across the entire research timeline. Results revealed that most adolescents — and increasing proportions over cohorts — ceased risky behavior such as substance use, unsafe sexual activities, and violence, and did not display signs of mental health problems such as depression, according to researchers at Boston College and San Diego State University.

However, a small group of adolescents — representing less than nine percent of those surveyed — reported heightened mental health concerns such as increased symptoms of hopelessness and suicidality, and an even smaller proportion reported both heightened risky behavior and mental health problems, necessitating additional public health measures to intervene and promote enhanced wellbeing.

The research was led by Rebekah Levine Coley, a professor and the Gabelli Family Faculty Fellow in the Counseling, Developmental and Educational Psychology department at Boston College’s Lynch School of Education and Human Development, in collaboration with Jane Leer, an assistant professor of Psychology at San Diego State University’s Department of Psychology, and Lindsay Lanteri, a Ph.D. candidate at the Lynch School.

“Perhaps the most important finding from this work highlights the dominant and increasing prevalence of adolescents with low levels of internalizing behaviors who are also abstaining from multiple types of behavioral health risks,” said Coley, who also directs BC’s Center for Child and Family Policy. “Simultaneously, the relatively modest but increasing number of youth reporting elevated indications of depression points to target populations for prevention and treatment efforts, which is critical information for policymakers and health practitioners seeking to optimize the well-being of U.S. adolescents.”

For an interview with the primary author, contact Rebekah Levine Coley at coleyre@bc.edu (617-552-6018) or Phil Gloudemans, Boston College University Communications at Philip.gloudemans@bc.edu (401-338-6385).

---

Journal

PEDIATRICS

DOI

10.15442/peds.2024-068774 

Method of Research

Data/statistical analysis

Subject of Research

People

Article Title

Trends in Mental and Behavioral Health Risks in Adolescents: 1999-2021

Article Publication Date

18-Mar-2025

COI Statement

Professors Coley and Leer, and Ms. Lanteri, have no conflicts of interest relevant to this article.

Thorny skates come in snack and party sizes. After a century of guessing, scientists now know why.

Florida Museum of Natural History

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Thorny skates come in two distinct sizes along the Atlantic Coast of North America, but no one could figure out why. Then their numbers began to plummet, and it became imperative for scientists to understand whether their sizes had anything to do with it. Now we have a key part of the answer.

view more 

Credit: Illustration by Jorge Machuski

When Jeff Kneebone was a college student in 2002, his research involved a marine mystery that has stumped curious scientists for the last two decades. That mystery had to do with thorny skates in the North Atlantic. In some parts of their range, individuals of this species come in two distinct sizes, irrespective of sex, and no one could figure out why. At the time, neither could Kneebone.

In a new study, Kneebone and researchers from the Florida Museum of Natural History say they’ve finally found an answer. And it’s all thanks to COVID-19.

People have known about the size discrepancy in thorny skates for nearly a century, but it became critically important beginning in the 1970s, when their numbers took a nosedive. The cause of the decline was thought to be overfishing by humans, and the solution was simple. In 2003, a strict fishing moratorium in the United States was put in place for thorny skates and another species, the barndoor skate, that was also doing poorly.

“The barndoor skate rebounded to the point where they’re now allowed to be harvested again, but for whatever reason, the thorny skate has remained low, despite 20 years of protection,” said Kneebone, who currently works as a senior scientist at the Anderson Cabot Center for Ocean Life at the New England Aquarium.

According to survey data collected by the National Oceanic and Atmospheric Administration, thorny skates have declined by 80% to 95% in some areas, particularly the Gulf of Maine, and they’re also languishing in low numbers in Canadian waters off the Scotian Shelf.

Thorny skates have a large distribution. They can be found from South Carolina up to the Arctic Circle and east through Scotland, Norway and Russia. In the Arctic and European part of their range, thorny skates come in just one size. It’s only along the coast of North America that small and large varieties coexist.

“No one could understand what the deal was with these skates,” said study co-author Gavin Naylor, director of the Florida Program for Shark Research at the Florida Museum of Natural History. Scientists had tried studying thorny skate DNA to see if there were any differences between the large and small sizes, but they came up empty-handed. “The big forms are twice the size, and it takes them 11 years to reach adulthood. The small forms are mature by the time they’re six years old. There’s got to be genetic differences.”

Naylor thought he might be able to crack the code.

The idea was simple. Previous studies had tried to answer the question by analyzing a few short DNA sequences taken from a small number of thorny skates. It was a good strategy, Naylor reasoned, but fell short because researchers hadn’t yet processed nearly enough DNA. Instead, what was needed was a gene capture approach: a labor-intensive method that allows researchers to collect DNA sequence data from thousands of sequences throughout an organism’s genome, the term used to describe DNA stored in the nuclei of cells. Most importantly, they’d do this for hundreds of thorny skates, which would provide them ample data to scour. 

He put the word out to the scientific community, and people sent the team more than 600 tissue samples collected across much of the Northern Hemisphere, and he made the costly preparations to get the lab work underway, with funding from the Lenfest Foundation and the National Science Foundation.

Then the COVID-19 pandemic hit, and the subsequent restrictions that were put in place made it impossible to conduct extensive, in-person lab work, putting the project on indefinite hiatus.

One of Naylor’s postdoctoral researchers at the time, Shannon Corrigan, pulled together a salvage mission. If they couldn’t collect gene capture DNA from hundreds of thorny skates, they could sequence the entire genome of four or five individuals. This would drastically cut down on the amount of in-person work that needed to be done.

It was a risky plan. There was only a small chance they would find what they were looking for by sequencing genomes, and they only had enough funding to do one or the other.

It was a Hail Mary, Naylor said, but one that paid off. Had they used the original gene capture idea, “we would have missed it entirely.”

As it was, they only nearly missed it. The study’s first author, Pierre Lesturgie, was tasked with analyzing the genome — all 2.5 billion base pairs of it — once it had been sequenced. As he was combing through the data, something strange caught his eye.

“There was a large region on chromosome two that we thought was weird. Since it was behaving in a way we didn’t understand, we considered removing it from the analysis,” Lesturgie said. He thought it might be an aberration or potentially an error introduced during the sequencing process, and worried it would reduce the accuracy of their results. He was about to trash it when Naylor mentioned it looked like the sort of thing you’d get from a gene inversion, a natural process in which a sequence of DNA is flipped in the wrong direction.

Most organisms, including humans, have at least a few inversions in their genomes, so they’re not uncommon, but they seldom result in observable differences between individuals. But because it was all the researchers had to go on, they checked to see if the inverted sequence was present in both large and small thorny skates. It wasn’t. Only large thorny skates had the mirrored stretch of DNA. They’d need to do more work to confirm it, but they’d found their answer. Cue the popping bottles of champagne and celebratory good cheer.

Figuring out what caused the size difference is only the first step, Kneebone said. Now researchers can make headway on developing a conservation plan. The next step will involve good old-fashioned observation. Before the discovery of the gene inversion, it was difficult — and in some cases impossible — to distinguish between the large and small types.

“We could identify the large males and females, because they’re bigger than anything else,” Naylor said. At maturity, both large and small males develop long, trailing claspers on either side of their tale, giving them the overall appearance of a kite with streamers. “So when you’ve got a small male with large claspers, we know it’s an adult. But we can’t do anything with the small females, because we don’t know whether they’re just babies on their way to getting big.”

This limitation has hampered research on the species, Kneebone said. “The big question has always been, what do the life histories of the two morphs look like? Currently, they’re not discriminated in the stock assessment, so a thorny skate is a thorny skate is a thorny skate.”

The final step will be figuring out why thorny skates are continuing to decline in parts of their range. Fortunately, scientists already have a few good leads. Current evidence suggests it’s harder for the two sizes to interbreed in places where they’re declining than it is in others. It’s possible this natural and partial barrier to reproduction cold be exacerbated by climate change.

Thorny skates are having the most trouble in the Gulf of Maine, where sea surface temperatures have increased faster than 99% of the world’s oceans over the last several years. This has had all sorts of unpleasant effects, like the collapse of cod fisheries in the region.

Whether climate change is partially responsible for the plight of the thorny skate and, if so, why it has an undue negative influence on this single species compared with other skates that live in the same area, remains to be seen. To determine that, Kneebone said they’ll need more data.

“We’re trying to use the best available science to make decisions about how to best manage and sustain populations.”

The authors published their study in the journal Nature Communications.

The skin of sharks and rays are covered in a layer of denticles, which are essentially microscopic teeth. Thorny skates get their namesake from their sickle-shaped barbs that are made from the same material.

Credit

Florida Museum photo by Kristen Grace

---

Journal

Nature Communications

DOI

10.1038/s41467-025-56126-z 

Article Title

Short-term evolutionary implications of an introgressed size-determining supergene in a vulnerable population

 

Illinois study: Extreme heat impacts dairy production, small farms most vulnerable

University of Illinois College of Agricultural, Consumer and Environmental Sciences

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Milk yield declines when dairy cows are exposed to extreme heat and humidity, University of Illinois researchers found.

view more 

Credit: College of ACES

URBANA, Ill. – Livestock agriculture is bearing the cost of extreme weather events. A new study from the University of Illinois Urbana-Champaign explores how heat stress affects U.S. dairy production, finding that high heat and humidity lead to a 1% decline in annual milk yield. Small farms are hit harder than large farms, which may be able to mitigate some of the effects through management strategies.

“Cows are mammals like us, and they experience heat stress just like we do. When cows are exposed to extreme heat, it can have a range of negative physical effects. There is an increased risk of infection, restlessness, and decreased appetite, which leads to a decline in milk yield. For dairy producers, the heat impact is a direct hit on their revenue,” explained study co-author Marin Skidmore, assistant professor in the Department of Agricultural and Consumer Economics (ACE), part of the College of Agricultural, Consumer and Environmental Sciences at the U. of I. She conducted the study with Jared Hutchins, assistant professor in ACE, and Derek Nolan, Illinois Extension specialist and teaching assistant professor in the Department of Animal Sciences at the U. of I.

Skidmore and her colleagues analyzed milk production data from nine U.S. Midwest states. They included over 56 million cow-level production records from 18,000 dairy farms from 2012 to 2016. They adjusted the milk data for protein and fat content to more accurately estimate milk quality, which determines the price.

“Previous studies have focused on fluid milk yield. But in our dairy marketing system, milk is sold on components. When you calculate revenue, it’s not just about how many gallons of milk, but whether it’s high-quality milk with high protein and fat content,” Skidmore said.

The researchers combined quality-adjusted production data with daily weather data for temperature and humidity. They calculated temperature-humidity index measurements, which most accurately reflect the heat stress a cow experiences, as high heat and humidity make it harder for the cow to cool down through sweating. 

They found that, on average, 1% of annual milk yield is lost to heat stress. This may not sound like a lot, but it amounts to about 1.4 billion pounds of milk (adjusted for energy content) over five years for the 18,000 herds included in the study. Based on average milk prices, this is equivalent to about $245 million in lost revenue.

Most of the losses are due to low- and moderate-stress days because those are more common; however, yield loss per cow due to an extreme-stress day is more than double that of a moderate-stress day.

Heat stress disproportionately affects smaller farms, the study showed. Herds with fewer than 100 cows lost an average of 1.6% of annual yield, and while they supplied less than 20% of total output in the sample, they represented 27% of total damages.

Producers can implement various forms of mitigation strategies, such as open barn sides, fans, and sprinklers. Larger farms are better able to do so, but it’s not possible to fully protect against heat stress.

“There are a number of different adaptive methods, but there is no silver bullet. You can install more sprinklers and sophisticated ventilation systems. You could change the timing of calving to avoid these warmer periods, but that incurs other risks, and it is a complex issue,” Skidmore explained.

“Lower levels of heat intensity are potentially manageable with some of the practices available. At those levels, the largest farms are not really taking on noticeable losses, and that’s where we start to see the difference between small and large farms. But there is a level of heat stress where it’s so hot and humid that you can’t completely manage it.”

The researchers also project potential losses forward to 2050, using the average predictions from 22 different climate models. Under most scenarios, extreme heat days are predicted to be much more frequent, and milk yield losses are expected to increase about 30% by 2050.

If policy makers consider dairy production a priority, small farms will require greater support to remain competitive in the future, Skidmore stated.

“If there's interest in continuing to have a healthy and robust small dairy production presence in the U.S., that probably will require financial incentives to help farmers implement mitigation strategies, as well as investments in further research on how to manage the highest levels of heat stress,” she concluded. 

The paper, “Vulnerability of US dairy farms to extreme heat,” is published in Food Policy [DOI:10.1016/j.foodpol.2025.102821]. Funding was provided by a Center for the Economics of Sustainability Seed Grant. 

---

Journal

Food Policy

DOI

10.1016/j.foodpol.2025.102821 

Method of Research

Data/statistical analysis

Article Title

Vulnerability of US dairy farms to extreme heat