Tech
How to ensure high-quality synthetic wireless data when real-world data runs dry
To train artificial intelligence (AI) models, researchers need good data and lots of it. However, most real-world data has already been used, leading scientists to generate synthetic data. While the generated data helps solve the issue of quantity, it may not always have good quality, and assessing its quality has been overlooked.
Wei Gao, associate professor of electrical and computer engineering at the University of Pittsburgh Swanson School of Engineering, has collaborated with researchers from Peking University to develop analytical metrics to qualitatively evaluate the quality of synthetic wireless data. The researchers have created a novel framework that significantly improves the task-driven training of AI models using synthetic wireless data.
Their work is detailed on the arXiv preprint server in a study titled “Data Can Speak for Itself: Quality-Guided Utilization of Wireless Synthetic Data,” which received the Best Paper Award in June at the MobiSys 2025 International Conference on Mobile Systems, Applications, and Services.
Assessing affinity and diversity
“Synthetic data is vital for training AI models, but for modalities such as images, video, or sound, and especially wireless signals, generating good data can be difficult,” said Gao, who also directs the Pitt Intelligent Systems Laboratory.
Gao has developed metrics to quantify affinity and diversity, essential qualities for synthetic data to be used for effectively training AI models.
“Generated data shouldn’t be random,” said Gao. “Take human faces. If you’re training an AI model to identify human faces, you need to ensure that the images of faces represent actual faces. They can’t have three eyes or two noses. They must have affinity.”
The images also need diversity. Training an AI model on a million images of an identical face won’t achieve much. While the faces must have affinity, they must also be different, as human faces are. As Gao noted, “AI models learn from variation.”
Different tasks have different requirements for judging affinity and diversity. Recognizing a specific human face is different than distinguishing it from that of a dog or a cat, with each task having unique data requirements. Therefore, in systemically assessing the quality of synthetic data, the team applied a task-specific approach.
“We applied our method to downstream tasks and evaluated the existing work of synthesizing data,” said Gao. “We found that most synthetic data achieved good diversity, but some had problems satisfying affinity, especially wireless signals.”
The challenge of synthetic wireless data
Today, wireless signals are used in technologies such as home and sleep monitoring, interactive gaming, and virtual reality. Cell phone and Wi-Fi signals, as radio waves, hit objects and bounce back toward their source. These signals can be interpreted to indicate everything from sleep patterns to the shape of a person sitting on a couch.
To advance this technology, researchers need more wireless data to train models to recognize human behaviors in the signal patterns. However, as a waveform, the signals are difficult for humans to evaluate.
It’s not like human faces, which can be clearly defined. “Our research found that current synthetic wireless data is limited in its affinity,” said Gao. “This leads to mislabeled data and degraded task performance.”
To improve affinity in wireless signals, the researchers took a semi-supervised learning approach. “We used a small amount of labeled synthetic data, which was verified as legitimate,” Gao said. “We used this data to teach the model what is and isn’t legitimate.”
Gao and his collaborators developed SynCheck, a framework that filters out synthetic wireless samples with low affinity and labels the remaining samples during iterative training of a model.
“We found that our system improves performance by 4.3% whereas a nonselective use of synthetic wireless data degrades performance by 13.4%,” Gao noted.
This research takes an important first step toward ensuring not just an endless stream of data, but of quality data that scientists can use to train more sophisticated AI models.
More information:
Chen Gong et al, Data Can Speak for Itself: Quality-guided Utilization of Wireless Synthetic Data, arXiv (2025). DOI: 10.48550/arxiv.2506.23174
Citation:
How to ensure high-quality synthetic wireless data when real-world data runs dry (2025, September 15)
retrieved 15 September 2025
from https://techxplore.com/news/2025-09-high-quality-synthetic-wireless-real.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
Tech
Tata Communications unveils self-healing network | Computer Weekly
Tata Communications has launched a self-healing network platform called IZO datacentre Dynamic Connectivity, which is designed to eliminate costly datacentre downtime and support the demands of an artificial intelligence (AI)-driven world.
In explaining the rationale for the launch, Tata Communications said that in the current digital economy, disruptions from cable cuts, route failures or sudden AI workload spikes can bring business to a standstill.
Specifically, that is every enterprise depends on the ability to always be connected with an uninterrupted data flow. From financial transactions, information technology-enabled services (IT-ITeS) and manufacturing to streaming platforms and online retail, the connections between datacentres keep the modern world running. Tata Communications added that when those connections are interrupted, businesses do not just slow down, they are brought to a complete standstill.
The company warned that the networks connecting many enterprise datacentres were built for a different era. Traditional datacentre (DC)-to-DC links were designed for predictable workloads and stable traffic patterns. It stressed that the current reality is far more dynamic. In this, enterprises operate across global locations and cloud environments, moving massive volumes of data in real time to support AI workloads and business needs.
In an environment shaped by increasing geopolitical constraints, cable outages, route failures or sudden spikes in demand, these can quickly cascade into service disruption and operational risk, leading to a costly downtime. In such scenarios, the response is often reactive and manual, consuming valuable time when business need certainty and speed.
The IZO datacentre Dynamic Connectivity platform is designed to address these issues by creating an intelligent network that covers key global datacentres across five continents.
Tata Communications, said that unlike conventional architectures, the new platform uses deterministic multi-path routing to deliver predictable latency and performance. It said this transforms resilience from a reactive process into an autonomous capability, changing how enterprises connect their datacentres in an increasing AI-driven and distributed world.
This means the platform is smart enough to automatically re-route traffic within seconds without manual intervention during disruptions. This is said to enable enterprises to achieve >99.99% service availability across mission-critical infrastructure that supports business-critical applications, “turning resilience from a contingency into a default state”.
The platform is also attributed with giving enterprises access over their connectivity. Through a unified digital interface and APIs, enterprises can monitor performance, receive proactive alerts and dynamically scale bandwidth as workloads evolve.
Tata Communications said the result is that business impact is a shift from crisis management to strategic growth with business leaders no longer having to guess their future needs or over-pay for “just in case” bandwidth. Instead, leaders have access to Al-driven predictive insights allowing them to forecast their capacity requirements in advance. If a sudden workload demands more capacity or choice of route, users can instantly scale their bandwidth or add route through self-service feature.
Tata Communications calculates that by moving to a flexible, consumption-based pricing model, enterprises can reduce the need for idle backup capacity and save up to 30% on operational costs. Enterprises can activate resilience and bandwidth when required, helping to optimise costs while maintaining deterministic performance across geographies.
“Datacentres are the core engines of today’s digital economy, and the connections between them must be as resilient as the networks that connect them,” said Genius Wong, chief technology officer and executive vice-president of core and next-gen connectivity services at Tata Communications. “They must be just as dynamic as the applications they support.
“With IZO DC Dynamic Connectivity, we are shifting resilience from a reactive process to an autonomous capability. By combining global reach, deterministic routing and intelligent automation, we are enabling enterprises to build a digital foundation that scales with confidence and operates without disruption.”
Tech
Can a Home Appliance Fix the Problem of Soft-Plastic Waste?
Soft plastics are notorious for jamming sorting machines, slipping through processing lines, and wreaking havoc on the environment. They’re also not accepted in most municipal curbside recycling programs.
Facilities for recycling these types of plastic exist, but getting waste to these locations clean and free of what some call “wishful recycling” items (compostable cups, plastic utensils) is such a challenge that the majority of soft plastics, even the bags recycled at the front of grocery stores, end up in the trash. The SPC is what Arbouzov calls a “pre-recycling device,” designed to simplify this stream and deliver plastic that’s contained, traceable, and more likely to make it through the system.
I tried to envision how the blocks would turn into patio furniture, as advertised, but didn’t learn exactly how until months later, when Arbouzov sent me a video of the blocks at their final destination—a facility in Frankfort, Indiana, that specializes in processing polyethylene and polypropylene films. The blocks get shredded into crumbles resembling, at least on video, handfuls of wet newspaper, which are then compressed into composite decking, chairs, garden edging, and more.
Courtesy of Clear Drop
Courtesy of Clear Drop
“The full cycle from mailing a block to it entering recycling processing typically takes a few weeks,” Arbouzov said, “depending on shipping time and batching schedules.” Right now, the Frankfort location is the only facility processing the blocks, but Arbouzov said he hopes this is only temporary.
“Our goal is to shift more of this processing closer to where the material is generated, so blocks can move in bulk through regional recycling infrastructure rather than through mail-based logistics,” he said. “The mail-back system is essentially a bridge that allows the material to be captured today while that larger infrastructure develops.”
Recycling, Rewired
I found that my household of three was able to produce a block every couple of weeks, which quickly outpaced the provided supply of mailers. As the blocks started piling up on the floor of my office, I found myself wishing the SPC made something useful for consumers. Spoons, straws, 3D-printing filament … anything that could be used at home.
However, a 2023 Greenpeace report found that recycling plastic can actually make it even more toxic than it already is—heating it can not only cause existing chemicals to escape into the air and water supply, but even create new ones, like benzene. Would I want this in my house? Does recycled plastic actually belong in a circular economy? I asked Arbouzov what he thought.
Tech
Can Modular Phone Accessories Finally Evolve Beyond MagSafe?
Predating the launch of Moto Mods in 2016, the first batch of Jolla The Other Half concepts included back covers with an extra E Ink display, an infrared camera, and an Angry Birds tie-in that activated themes and ringtones. But probably the most popular was a Blackberry/Nokia Communicator-style slider keyboard made and sold by two entrepreneurs from the original Jolla community. That trend is back in—at CES 2026, accessory company Clicks showed off a magnetic keyboard accessory you can slap on the back of any Qi2 or MagSafe smartphone, though it uses Bluetooth for connectivity.
Quite a bit has changed in what’s achievable, not least more bandwidth, more capability, and more accessible, high-quality 3D printing. “We have seven pogo pins [on the Jolla Phone] which give you the capability to get power out and power in,” says Jolla CEO Sami Pienimäki. “So you can do maybe wireless charging, and you can power external circuit boards.” Pienimäki imagines E Ink interfaces or low-bandwidth radios on the back of its upcoming phone—it has an I3C interface, which delivers bit rates up to 12 megabits per second, allowing data to flow between the phone and the mod, enabling new kinds of smarter modular accessories.
Jolla has promised to release the final phone specifications by the end of the month, with shipping due for the first preorder customers at the end of June. Pienimäki teases that it’s “tempting” for him to release one of Jolla’s own internal concepts for a TOH back cover even earlier as “a showcase of what you can actually do.” (The Jolla Phone doesn’t have FCC approval in the US, but the company is considering a US launch in the future.)
With more than 10,000 preorders since December 2025, Jolla is back in business but still far from mainstream. So why, despite plenty of internet hype over the years, did truly modular phones never quite take off?
“During the LTE days, there was thinking that these devices would morph into ‘cloud phones,’ where the rest of the phone could be cost-optimized,” Fieldhack says. “Swappable parts and lower costs, as most of the compute would be done in the cloud.”
But things changed as flagship phones went from costing $350 to around $1,000. Both the camera and media production and consumption became much more important: “Great displays, great cameras, multiple cameras, more memory, better sound and mics, as well as more elegant and thin devices—this is not easily done on a modular smartphone,” Fieldhack says. “There are huge compromises, and phones are thicker and heavier with less performance. Then, agentic AI, on-device for lower costs and better security, made modular design even less optimal.”
Repairable Modules
One strong and emerging argument for true hardware modularity is repairability. Another European smartphone maker, Fairphone, has been making that case for over a decade. “It’s about thinking about how do we group the actual phone itself into modules?” says Fairphone chief technology officer Chandler Hatton. The latest FairPhone Gen 6 smartphone is made up of 12 modules. A customer sitting at the kitchen table with a single T5 screwdriver (included) and a guitar pick can repair the phone quickly, easily, and cheaply.
-
Entertainment1 week agoStrategic oil stocks to be released ‘immediately’ in Asia and Oceania: IEA
-
Tech7 days agoJustice Department Says Anthropic Can’t Be Trusted With Warfighting Systems
-
Business1 week agoStocks To Watch: Tata Motors, IndiGo, Jindal Stainless, GMR Airports, Hindalco, And Others
-
Sports1 week agoMen’s March Madness 2026 bracket: Get to know all 68 teams
-
Business1 week agoGas supply crunch a worry for AC makers ahead of peak season – The Times of India
-
Tech1 week agoEarly Deals From the Amazon Spring Sale That Passed Our BS Test
-
Sports6 days agoMarch Madness 2026 – How to watch in SA, start time, schedule, TV channel for NCAA championship basketball tournament
-
Fashion1 week agoTrump signs order to combat fraudulent ‘Made in America’ labels
