In a lab in Luxembourg in 2024, engineers at Flawless Photonics unpacked a spool of glass fiber that had just returned from the International Space Station (ISS). As they threaded it onto their rewinding machine, the machine jammed—as happens sometimes—and they braced for the usual outcome: the fiber would snap. This time though, nothing broke.

It was an early hint that their experiment might have been successful. The Flawless Photonics team had set out to draw fibers of a heavy-metal fluoride glass called ZBLAN—named for its other components of zirconium, barium, lanthanum, aluminum, and sodium (chemical symbol Na)—in microgravity, aiming to achieve superior optical and physical properties. “We realized, wow, that fiber is much stronger than the stuff we had produced on Earth,” says Hubert Moser, the company’s co-founder and CTO.

ZBLAN Fiber’s Microgravity Advantages

Photonics engineers have been intrigued by ZBLAN for decades because it can, in theory, transmit light with a 100x lower loss than silica. It also retains this transparency across a wide wavelength range, from the ultraviolet through to mid-infrared. Because of these characteristics, ZBLAN has been touted as a potential route to achieving “repeaterless transoceanic transmission of light” in submarine telecommunication cables.

But challenges with manufacturing have meant that the current state-of-the-art commercial ZBLAN fibers fall far short of their theoretical performance. One major issue is that ZBLAN is prone to crystallization at both the casting and fiber drawing stages. Tiny crystals form in molten glass as it solidifies, causing light to scatter and reducing ZBLAN’s transparency. Parabolic-flight experiments in the 1990s and early 2000s were the first to suggest that processing glass fibers in reduced gravity could suppress crystallization. That finding led to several high-profile attempts to draw ZBLAN fiber in microgravity environments, including a 2022 payload to the ISS from a start-up called Apsidal. Each of these efforts produced relatively short lengths of fiber, and somewhat mixed results, but excitement over in-orbit manufacturing has remained high.

The latest experiment from Flawless Photonics and Adelaide University focused on drawing long, continuous fibers from a series of ZBLAN preforms—solid cylinders of the glass, made using various formulations. Over the course of a month aboard the ISS, their compact drawing tower, which is about the size of a large suitcase, successfully produced the first long ZBLAN fibers in space, 11.8 kilometers in total, including one continuous strand measuring 1.1 km.

A researcher at Flawless Photonics tests the light conduction capabilities of a strand of ZBLAN fiber.Yordan Prodanov/Luxembourg Space Agency

Once returned to Earth aboard an automated SpaceX capsule, samples of the space-drawn fiber were sent to five labs for optical and mechanical testing, alongside equivalent fibers made in an identical drawing tower at Flawless’ facility in Luxembourg. The results show that for “step-index” fibers, which consist of a high-refractive-index core and lower-index cladding, those made in orbit had distinctly lower attenuation (or loss) than the fiber drawn on the ground.

The space fibers were also 80 percent stronger, as the engineers had serendipitously discovered. “Crystallization typically diminishes the mechanical stability” of glass, says Heike Ebendorff-Heidepriem, the Adelaide University materials scientist who led the glass side of the project. “So here we clearly saw the effect of microgravity.”

But the story was different for the “single-index” fibers, composed of just the lower-index form of ZBLAN. Mechanically, the space-drawn samples were again slightly stronger. But in terms of their optical performance, there was no measurable difference between those drawn in orbit and those drawn on the ground. Evidently something other than gravity influences the quality of ZBLAN fibers.

Tackling ZBLAN Fiber Impurities

The Flawless team calculates that, at best, ZBLAN fibre made in microgravity leads to a factor-of-two reduction in loss compared to conventional fiber—a far cry from the factor-of-100 potential. Closing the rest of the gap to ZBLAN’s theoretical performance will involve tackling two gravity-independent factors, related to the properties of the preforms themselves.

Bubbles and other defects can form when the molten glass is poured into a mold, and are trapped as it hardens. These defects, which exist regardless of whether a fiber is drawn on the ground or in orbit, scatter light and reduce how much of the signal makes it through the fiber.

More significant than structural defects are impurities. Traces of transition metals—iron, cobalt, nickel and copper—from the raw ingredients can remain in the final ZBLAN formulation. Each of these impurities strongly absorbs light at specific wavelengths in the fiber. Impurity-absorption means that even a fiber free of bubbles and crystallization would still lose signal over distance.

John Ballato, a materials scientist at Clemson University who wasn’t involved in the research, argues that the impurity problem undercuts the current rationale for making fiber in space. “We’ve known since the ‘90s that the problem with fluorides with respect to clarity and loss are the impurities in the raw materials,” he says. “Melting those materials in space buys you nothing from that perspective. It’s not attacking the fundamental problem.” Ballato acknowledges that crystallization “may creep back in as a problem” if the goal is to scale up production: “Reduced gravity may become a manufacturing solution in that case, but that’s after a whole bunch of ifs.”

“We are trying to automate things and make them more reliable and cheaper,” Flawless’ Moser says. “We knew before this result that the chemical purity of our ingredients needs to be improved. It’s something we want to work on, but it’s very financially intensive.” Ebendorff-Heidepriem adds achieving a purification “on the order of a thousand times,” the level needed to eliminate most transmission losses, will be a challenge.

Economics of Space-made ZBLAN Fiber

All of the technical issues around making optical fiber in space lead up to the big practical one: Is it worth it? ZBLAN fiber is currently a specialty item that sells at a significant premium over conventional silica, and the price of heavy launches to low-Earth orbit has fallen by 95 percent since 1980. Moser claims the arithmetic will favor Flawless at scale: “As soon as we can use our machine for longer lengths of fiber, the launch cost is no longer the driver.”

Relying on the human operators aboard the ISS is a major limitation, however, especially because the station is set to be retired in 2030. “We need to look at all the alternatives,” Moser says, including possible private space stations or automated orbital facilities. Space manufacturing also has significant environmental downsides. A 2022 study showed that soot from rocket launches has about 500 times as much warming effect on the atmosphere as does soot from aviation and surface sources. Some models suggest that frequent launches could also lead to a thinning of the ozone layer.

Flawless Photonics is planning another experiment for mid-2027, likely the last one aboard the ISS, to help make its case. The company and university teams are testing different methods to make better fiber preforms, including pre-processing ZBLAN in brief bursts of microgravity “using a drop tower from a German university,” Moser says.

Ballato’s advice to anyone hoping to build a fiber optic business in space: “Work with the best fluoride glass possible. Do your space processing on it. Then you’ll know if microgravity actually buys you something.”

Providing wireless connectivity from the sky has been the dream of a number of commercial ventures like the shuttered Google Project Loon. Now, Japanese telecom giant SoftBank has moved a step closer to selling phone service from the sky. In a partnership with an American aerospace startup, the pair recently sent a solar-powered, uncrewed, lighter-than-air aircraft across the Pacific to beam cellphone coverage down over Japan.

Moriarty, N.M.-based Sceye’s first mission to Asia flew from New Mexico to Japan and back in August and September. The craft traveled nearly 30,000 kilometers through the stratosphere over 30 days. While above Japan’s Shikoku Island, the lighter-than-air platform performed station-keeping maneuvers off the island’s southeastern tip and tested its network’s performance.

Sceye’s platform features an antenna called SceyeCell, behaving like a cell tower in the sky. The company says its high-altitude platform system (HAPS) can cover the equivalent area of 500 ground-based cell towers.

The tests simulated a major earthquake in a tectonically active region off Japan’s southeastern coastline called the Nankai Trough.

“We confirmed that communication performance equivalent to terrestrial networks can be achieved while minimizing interference with existing ground base stations,” says Masayuki Kamimura, a senior director at SoftBank’s Technology Unit.

The researchers report successful video, voice, and text-based smartphone communication with the balloon’s 4G LTE-equivalent base station linked to SoftBank’s terrestrial cellphone network. The technicians also tested communication with a drone via the SceyeCell platform. The HAPS-based network transmitted the drone’s control, location, and video feeds—simulating situations involving reconnaissance and provision of supplies in a disaster, and monitoring activities and traffic in the affected areas around a disaster site.

“In the future, we aim to build a three-dimensional communications network that combines terrestrial, HAPS, and satellite layers,” Kamimura says.

What are the Future of High-Altitude Communications Networks?

HAPS are attractive not only for their maneuverability but also their proximity to Earth, compared to satellites. HAPS cut the lag time for a signal’s round-trip travel, compared to satellite connectivity networks like SpaceX’s Starlink, making a possibly critical difference in applications such as real-time robotic control.

SoftBank aims to start selling HAPS services in 2027. It expects demand in Japan, where mountains sometimes make continuous coverage difficult to maintain, while earthquakes can sometimes take down the networks that exist.

SoftBank and Sceye’s flight is the latest round in a race between helium-filled craft and winged aircraft. Helium craft are cheaper to build and can carry more solar panels, batteries, and equipment. But lighter-than-air craft are also harder to steer, and so far winged craft have flown longer. (Within the lighter-than-air category are airships and balloons; airships tend to operate their own propulsion, while balloons tend to drift with the wind.)

By contrast, fixed-wing platforms require energy-draining propulsion to stay aloft. Last year, a Airbus-owned Zephyr fixed-wing uncrewed aircraft reportedly stayed aloft for a record-setting 67 days.

“Our lift capacity is moving towards over 200 kilograms, with today’s fixed wings only carrying around 5 kg,” says Jonathan Wilson, Chief of Mission Operations, at Sceye. “It makes a world of difference in what you can offer, from broadband connectivity to persistent Earth observation.”

Guided from the ground by satellite, Sceye’s solar-powered ST1 craft chose different routes across the Pacific on the way out and the way back.SoftBank

What are the Biggest Challenges for HAPS-Based Telecom?

Tim Farrar, wireless industry consultant, of the Menlo Park, Calif.-based TMF Associates, says each of the competing HAPS platforms must negotiate a delicate balancing act between maintaining sufficient power and battery life on a very light platform and having enough mass and power budget left to usefully operate its payload.

“Improved battery technology—that is, better power to weight—is the most critical aspect here in enabling these platforms to operate outside tropical regions and in wintertime when the amount of sunlight is lower,” Farrar says.

Craig Mascarenhas, head of engineering research at Stratospheric Airborne Climate Observing System at Harvard University, is co-author of a recent ACS Energy Letters study on the battery needs of HAPS. He found that emerging batteries could improve feasibility for fixed-wing solar platforms, but commercialization depends on a wider set of variables, including autonomous reliability, launch and recovery logistics, and of course economics.

“For telecom end users, if HAPS become part of networks, the most visible change may be less about noticing a new technology and more about connectivity being available where it currently is not,” Mascarenhas says. “But the user experience—speed, latency, reliability, and cost—would depend heavily on the payload, spectrum, backhaul, network integration, and business model.”

Sceye, meanwhile, says it’s planning its next flight for Peru, where the company will also test broadband internet connectivity.

“We’ve now completed over 25 flights, and the focus is shifting from innovating the platform to scaling,” Wilson says. “The next chapter is making HAPS a normal part of our infrastructure, as familiar as ships in ports or trains on tracks.”

SoftBank and Sceye’s trial also tested other challenges for HAPS platforms, including edge computing in the stratosphere. When data processing was clocked on a server mounted on the balloon, communication latency was reduced by more than 40 percent compared to processing in the cloud via the internet, according to Kamimura. HAPS edge computing could be used for applications including physical AI and real-time video analytics, he added.

“I don’t understand the importance of that for typical users,” Farrar says. “What communications tasks are mainly being undertaken locally within a single HAPS platform? Maybe it is relevant for local netted battlefield communications or some such, but probably not your average consumer.”

Another test involved using a ground-based laser to track the HAPS and receive signals from an onboard reflector. Lasers would allow a HAPS system to be tracked from the ground or from satellites in space and precisely determine its position, but the technology is also relevant for future optical wireless links such as free-space optical (FSO) communications.

This White Paper shows RF and microwave engineers and technicians how compact analog signal generators support core amplifier and receiver test tasks, from gain and distortion measurements to blocker rejection, power calibration, and local oscillator substitution.

What you will learn about:

• How to measure amplifier gain and gain compression, including the P1dB point that marks the transition from linear operation into saturation.
• How two-tone intermodulation distortion testing reveals unwanted signal products created by amplifier nonlinearities, and why third-order intercept (IP3) is used to compare amplifier linearity.
• How to test receiver performance under blocker and interferer conditions, using high output power and fast switching speed to simulate real-world interference.
• How power calibration and RSSI testing depend on accurate, repeatable signal sources across multiple frequency channels and bands.
• How signal generators can substitute for local oscillator and clock subsystems during early-stage receiver development, before final circuit prototypes exist.

My job is to translate
dry and unrelenting code
into a user interface of surpassing beauty.
With my mouse, I roll one pixel after another
up the vast anthill of the internet.

My dream is to translate
the visions of the holy ones
into a communication protocol
of universal wonderment.
I want to launch shreds of light into the air
to fall like a layer of diamonds
on the endless mountains of the Web.

Don’t imagine these dreams are limited
by the LANs of the software lab.
Between here and the ultimate
unlimited interface of my aspirations
lives a dazzling darkness,
wide as the universe and thin as a hair.

In early August, two subsea cables off the coast of Australia, Indigo West and Indigo Central, failed a few hours apart from one another. What caused the two outages is still uncertain. An emerging technology that allows a cable to sense threats in advance could help pinpoint cause and location of future cable failures like these.

According to a recent International Telecommunication Union report, subsea cables suffer between 150 and 200 accidents a year, with fishing and anchoring as the leading causes of outages. The new cable-sensing capability could turn what is today a weeks-long guessing game into real-time diagnosis, allowing authorities to issue alerts even before a cable outage occurs. The tech could also address a limited availability today of reliable and standardized data on cable accidents and their causes.

The Indigo West and Indigo Central cables land at the city of Perth, where anchoring has been banned inside a designated submarine cable-protection lane since 2007. Subco, the cables’ operator, flagged “suspicious” vessel activity around the time of the outage, although the operator has said it will not know the cause of the failures for certain until later in the repair and recovery phase.

There is a limited fiber-optic-sensing tech in place today, able to detect cable failures over the outer 200 kilometers of a cable, called distributed acoustic sensing (DAS).

However, on long-haul cables, like Indigo West and Indigo Central lines, any portion of a DAS-enabled cable in the middle of the ocean—not in the first or last 200 km of any given fiber line—is today still unable to sense any pending dangers. So for long transoceanic cables, often between 6,000 and 20,000 km long, DAS as it currently exists is insufficient to detect most possible outages along a cable’s length.

Fortunately, the Les Ulis, France–based Alcatel Submarine Networks (ASN) says it is close to launching an upgrade of its DAS that would extend the technology’s supersensing capabilities across the entire length of a transoceanic fiber line.

How Does Distributed Acoustic Sensing Work?

A DAS interrogator box on either end of a transoceanic cable, installed at the cable’s landing point, sends laser light down the fiber and reads faint echoes that come back because of flaws in the glass, originating from the cooling process during the fiber’s manufacturing.

A tiny fraction of laser light traveling down the cable reflects off those flaws. DAS exploits this property (called Rayleigh backscattering), in which each cable behaves a little like an extended succession of tiny mirrors along the cable’s entire length—where each tiny mirror lets nearly the entire signal pass by, but occasionally reflects a small portion back toward the laser that sent it.

“The system works like an optical interferometer” says Jan Kristoffer Brenne, R&D manager for fiber sensing at ASN, “measuring how the phase of the returning light changes at each point along the cable.”

When an anchor or piece of trawl gear drags across the seabed, it sends vibrations through the seabed floor. Those vibrations stretch the fiber optic cable a little bit, subtly shifting the phase of the signals inside the line. Of course, if an anchor or trawl gear ever actually strikes or drags a cable, more substantial phase shifts along the line can pick that up too.

DAS interrogator boxes on shore can detect these phase shifts and issue a warning up to 15 minutes and 2 km before an anchor or trawl gear even reaches the cable.

The new DAS technology, extending sensing capabilities across an entire cable’s span, introduces new hardware components to onshore operations for the line. Instead of one interrogator box covering each end of the cable, ASN’s upgraded system runs multiple interrogators at once at the landing station, each one tuned in to a different wavelength.

The technology underlying the new, enhanced DAS—called wavelength-division multiplexing, a popular throughput-increasing technique also used in 4G and 5G telecom—is in fact already in use for every cable’s repeaters. But now enhanced DAS also relies on the same multiplexing trick, but for the Rayleigh backscattered reflections not just the primary signals.

Installing DAS on an existing cable is quick, Brenne says, typically requiring a few hours of work at the landing station. And installation doesn’t even require shutting down the cable. Shutting down single fibers at a time are all that’s required to set up the enhanced DAS hardware.

“We don’t put anything new in terms of new equipment inside the repeaters,” Brenne says. “But we adjust a feature inside [the repeater] in a way that supports cable-threat monitoring on long-haul cable.”

What Challenges Lie Ahead for DAS?

Signal strength is today the main hurdle to developing enhanced DAS that senses outages across the entire length of a transoceanic cable. Each echo from each tiny flaw in the fiber line is itself extremely faint, weaker by a factor of millions than the main signals the fibers carry. But also, every repeater that amplifies each echo also adds in a little noise that, over multiple repeaters, can drown out the original echo.

Even despite these complications, enhancing DAS sensing across the entire length of a cable is still worth pursuing, because outages in transoceanic cable lines are prohibitively expensive and time-consuming to fix.

“There’s a shortage in the repair ship industry generally with only 20 to 25 ships worldwide dedicated to [subsea-cable] repairs,” said Sandra Maximiano, chairwoman of ANACOM, Portugal’s telecommunications regulator. Brenne says a single repair to a cable line costs about 20 times as much as a DAS unit.

Enhanced DAS could also increase the time and accuracy of warnings, which could help answer the question of responsibility in the case of damage to the line. The ITU subsea cables report notes that pairing DAS with ship-tracking data can identify a vessel near a cable. However, the report added that linking inferred seabed activity to a specific ship in the open ocean “remains challenging.”

When four cables off West Africa were cut in March 2024, Nigeria’s communications minister, Bosun Tijani, had no clear answer to give the public. Operators suspected landslides or earthquakes, but the cause was never confirmed.

“Everyone wanted an answer. I struggled to give one,” says Tijani who now wants a regional redundancy center “to monitor conditions of the cables with solutions like DAS, provide information, predict issues, and give governments early warning.”

Even before France approved legislation banning social media for children under 15 last January, 13-year-old Benjamin was already wondering what life without social media would look like. “If we want to play football, we won’t be able to organize it. What will we do? Send letters instead?” he joked in an interview for Le Monde.

His reaction captured the central challenge behind the growing wave of youth social media bans: Removing access is one thing; understanding what those platforms mean in children’s lives is another.

Within weeks of Australia’s similar ban, the country’s eSafety Commissioner reported that platforms had restricted access to 4.7 million under-16 accounts. Two months later, though, one in five Australian teenagers under 16 was still using TikTok and Snapchat, according to a parental-control data company. But even if all children’s social media accounts were to disappear, do such bans actually make children safer online?

Governments are moving ahead without answering that question as they follow Australia’s lead. Indonesia’s child-safety framework, which took effect in March, bars children under 16 from holding accounts on “high-risk” platforms. The U.K. government has announced plans to ban social media for under-16s, add default overnight social media curfews for 16- and 17-year-olds, and extend child-safety rules to cover risky AI features. And on 17 September, the European Commission proposed the EU KIDS Act, which would bar children under 13 from social media, set 15 as the EU-wide minimum age for opening an account independently, and require platforms to show that their services are age appropriate and safe by design.

But based on my experience working on Child Online Protection initiatives with the International Telecommunication Union (ITU) across Southeast Asia and the Pacific, I know the bans don’t address the real problems. Instead, we should be paying more attention to the systems that generate harm in the first place—namely, recommender algorithms, engagement-maximizing design, opaque moderation, and extractive data practices.

Account removals are not the same as online child safety

My experience working on protecting children’s online safety has taught me three main lessons:

First, the public institutions responsible for child online protection often lack the staff, budget, or technical capacity to enforce complex online safety policies.

Indonesia is illustrative. A 2026 UNICEF evaluation found capacity constraints among service providers, long-term funding uncertainty, and a need for specialized personnel. At the local level, some staff lacked digital skills, while budget constraints left some areas reliant on external support.

Second, many children, and often their parents, lack the digital literacy and critical thinking skills needed to navigate online risks safely. My policy research on child online protection in Indonesia, published earlier this year in Digital Society, found substantial gaps that account removals cannot repair: Many children lacked guidance on navigating the internet safely, and large numbers did not know how to report harmful experiences.

And third, the platforms have limited independent oversight as they identify underage users, design age-verification systems, and report their own compliance. In Indonesia, platforms themselves are responsible for carrying out age verification, while the Ministry of Communication and Digital Affairs oversees compliance. TikTok’s appeals process for users flagged as underage, for instance, can require a government-issued ID and selfies, which is a problem because it involves collecting the additional personal data on an ID card, beyond that needed to confirm age. Will government regulators ensure that TikTok handles that data responsibly?

The privacy paradox of proving age

Every age-based ban creates an engineering problem: How can a platform reliably determine that a user is old enough, without intruding on other information? Governments and companies may use identity documents, parental authorization, app-store checks, or facial age estimation. Each approach has trade-offs among accuracy, privacy, accessibility, and resistance to circumvention.

There are also technical issues. One tool, facial age estimation, draws on enormous databases but it is probabilistic, not exact, because people vary so much. It’s also been shown to misclassify both children and adults.

The challenge should not merely be to “verify age.” It should be to prove that someone is above a threshold, without disclosing their identity, birth date, or other information third parties might use to create a marketing profile. The European Commission’s age-verification blueprint challenges companies to verify ages without collecting all that additional information.

Privacy-preserving technologies offer promising ways to achieve this. Zero-Knowledge Proofs (ZKPs) can confirm that someone meets an age threshold without revealing their identity or exact date of birth. W3C Verifiable Credentials are cryptographically verifiable digital claims that can disclose only the information needed, such as “over 16.” And device-based age signals can allow a phone or app store to share an age range without revealing a user’s exact birth date. But these methods still require rigorous security testing, common standards, independent oversight, and clear limits on data retention. Otherwise, poorly designed child-safety policies risk creating permanent identity infrastructures in which businesses, not people, control personal data.

Where connection goes when a platform closes

Blocking access to a platform redirects some young people, but not always where expected. Early anecdotal reports in Australia pointed to teenagers migrating to smaller, less-regulated platforms like Yope, a pattern the Cato Institute flagged as a “whack-a-mole” problem for regulators. But industry data collected two months later found no broad-based shift of that kind, aside from a small uptick in WhatsApp use. Many teens simply found a way to stay on the banned platforms.

This points to a deeper gap in current society: the erosion of youth “third places“ physical spaces where young people have room to socialize and build identity outside home and school. As those spaces have diminished, commercial communications platforms have absorbed that role.

For many teenagers, social media workarounds are merely inconvenient. But for isolated, marginalized, disabled, or LGBTQ+ youth who depend on online communities for support that’s otherwise unavailable, displacement can mean losing certain kinds of belonging, or having to move to a platform with even weaker oversight.

How to design safer online systems for children

If blanket social media bans don’t work, then what will? The platforms have created many of the conditions that governments are now trying to contain: engagement-optimized recommenders, intrusive data practices, weak safeguards against unwanted contact, and features such as infinite scroll, autoplay, streaks, and persistent notifications.

These design patterns increasingly face regulatory scrutiny, including what’s required under the European Union’s Digital Services Act. A 2026 study from the 5Rights Foundation that tracked children’s device use minute by minute found that the user interfaces shape children’s attention, sleep, and well-being in real time.

A more durable response would regulate those interfaces directly, treating children as legitimate users whose privacy, agency, and well-being are required protections, not afterthoughts. That means designing for safety from the outset. One example would be for children’s apps to have high-privacy defaults, such as private accounts and location sharing switched off for minors. They could also have recommender systems that explain the main factors shaping a feed and give young users more control over personalization. The European Commission has published age-appropriate interaction guidelines that limit unsolicited contact and prevent minors from being added to groups without consent. Rules could also prohibit engagement-maximizing features that demand users’ attention, such as autoplay, infinite scroll, usage streaks, read receipts, and push notifications, by disabling or limiting them by default.

Governments should define measurable outcomes and fund independent evaluation, platforms should give researchers meaningful data access, and engineers should audit age-assurance systems for bias and data leakage. Schools, parents, and children themselves need a seat in designing the technology that’s designed to protect children.

If policymakers still decide to remove an infrastructure for youth connection, they should offer something better in return. Social media bans may reduce some forms of exposure to harmful content and may be justified for particular ages, services, or risks. But they are just one tool, not a comprehensive substitute for safer design, accountable platforms, digital literacy, institutional capacity, and noncommercial digital “third places”—moderated communities, creative spaces, and public-interest platforms designed for youth participation rather than profit.

The first wave of social media restrictions isn’t enough to keep children safe. Governments are still measuring what’s easiest to count, while neglecting harder-to-measure outcomes such as children’s access to safe third places and meaningful social connection, both online and offline. Until governments can show evidence that harm has actually declined, they will keep mistaking account removal for safety.

Shortwave radios offer a way to connect one location on Earth to practically anywhere else with minimal infrastructure. But these radios come with some drawbacks—a significant one being that, unlike satellite communications, their transmission rates for digital data are typically measured in just hundreds of bits per second.

Peter Bloom

Peter Bloom is the founder of Rhizomatica, a nonprofit that works with remote, indigenous, and off-grid communities around the world to build shortwave and cellular-communication infrastructure.

Peter Bloom is the founder of Rhizomatica, a Philadelphia-based nonprofit that has open-sourced a digital shortwave-radio set called the High-frequency Emergency and Rural Multimedia Exchange System, or HERMES. The set operates in the high-frequency (HF) band from 3 to 30 megahertz, as does Mercury, its digital modem. Rhizomatica staff travel around the globe to remote locations in countries like Bangladesh, Brazil, and Ecuador. Wherever they go, they use HERMES to help connect locals to the rest of the world.

Bloom spoke with IEEE Spectrum about how HERMES brings better data rates and encryption to shortwave radios.

How does HERMES connect remote locations?

Peter Bloom: We use the ionosphere as our satellite—or mirror—which helps us move information, voice, and data over really long distances. We’re using small radios that put out about 20 watts of power, and we can pretty reliably do 400- to 600-kilometer links between two radios. We’re talking about places that are not easy to reach, where it’s not simple to put terrestrial infrastructure.

What can HERMES send that a basic voice radio can’t?

Bloom: HERMES is a software stack—it’s a set of different programs that all work together in order to be able to send data over HF. HERMES allows you to send pretty much any file. Depending on what the file is, whether it’s a photo or an email or a voice memo, it just sends it as a file. It’s like a data pipeline over HF.

Why does sending files and data matter more than just voice?

Bloom: In emergency situations, people send their latitude and longitude over HF to say, “Hey, I’m here at this place.” People need to be able to send data over HF if there’s a manifest, a parts list, telemedicine—here’s what we have, here’s what we need. Instead of trying to read that out over the air, it’s much easier to just send the file. Same with a photo—if we need evidence that an area was logged illegally, we can just have someone send that over HF, rather than spending days getting down the river to get the photo where it needs to go.

Why did you build in encryption that amateur-radio regulations in many countries don’t allow?

Bloom: Encryption [regulations] for ham radio operators are different in each country. So it’s all optional—you turn it on, you turn it off. The reason we built the encryption is that some of the partners we work with are in very sensitive areas and don’t want to be sending out information that can be easily captured and used against them.

How has HERMES made an impact?

Bloom: We’ve been working with artisanal fishers in Bangladesh on a pilot project. There’s 10 or 11 boats that have HERMES systems on them. Pretty soon after we installed those, one of the boats had a mechanical issue in the Bay of Bengal, 100 or 200 kilometers offshore. They were able to send their GPS position and an SOS that they were having trouble. They were able to coordinate the rescue of the crew and the boat. So that was a really cool moment of HERMES in action that we’re super happy about.

This article appears in the October 2026 print issue as “Peter Bloom.”

This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore.

In June 2024, two cars casually traveled along a stretch of highway in Basque, Spain, weaving in and out of traffic just like any other vehicles on the road. But this trip was more than just a joyride through green, rolling, scenic mountains.

Scientists were analyzing how well the vehicle could, under natural highway conditions, send and receive data with other vehicles and dedicated infrastructure along the sides of the road—a process known as “collective perception.”

“The primary objective of collective perception is to provide vehicles with an extended situational awareness that reaches beyond the physical limitations of their onboard sensors,” explains Jon Ander Iñiguez de Gordoa, a researcher with the transport and security division of Fundación Vicomtech, who was involved in the study. For example, if an object on the road is partially occluded to a car’s sensors, the vehicle could use data from smart infrastructure and other vehicles positioned alongside the highway to “see” the object more clearly.

This type of communication—called vehicle-to-everything, or V2X, technology—has been a topic of research for several years, and some automakers and cities have already begun deploying it. However, Iñiguez de Gordoa says it has mostly been studied in urban areas, where traffic is slower. “In highway environments, where vehicles travel at high speeds and reaction times are severely limited, the rapid and precise exchange of data is critical,” he emphasizes.

The results of the study, published in the August issue of IEEE Transactions on Vehicular Technology, reveal key latency and distance limitations in high-speed environments.

A high-speed test

The experiment, funded by the Provincial Council of Bizkaia, took place on the council’s Bizkaia Connected Corridor, a stretch of highway in Spain that’s equipped with V2X communication infrastructure to support development and testing of connected mobility systems. Along the winding highway are dedicated roadside units, which exchange collective perception messages (CPMs) with passing vehicles. These data packets may contain information such as vehicles’ positions and velocities, as well as environmental data, which alerts vehicles to issues such as road obstacles or slippery conditions.

The researchers drove a Toyota Prius, equipped with a chipset designed to transmit and receive CPM messages, through traffic at speeds between 80 kilometers per hour and 120 km/h. During the drive, they studied the car’s ability to detect objects, process the data into CPMs, and transmit the data to the roadside units.

A second car driving alongside the Prius was equipped with a highly accurate monitoring system, to confirm the first car’s ground truth positioning throughout the experiment.

The results reveal several challenges with exchanging CPMs under highway conditions, particularly when it came to detecting objects further away. “The main limitation that we saw was that we were unable to detect objects that were farther than 50 meters,” Iñiguez de Gordoa explains.

While the system achieved high precision in detecting the distance to other vehicles, it frequently failed to detect them at all. For example, when a vehicle passed within 30 meters of the primary car, the system successfully detected it 65.7 percent of the time. This success rate dropped to 20.5 percent at distances between 30 and 50 meters and plunged to just 0.1 percent for objects between 50 and 100 meters away.

Latency challenges

The researchers also pinpointed when latencies were occurring, both during object detection and data transmission. They found that the perception system of the car could take up to 132 milliseconds to detect objects. Iñiguez de Gordoa notes this delay could be reduced somewhat by more sophisticated sensors—but these sensors would come with higher costs and computing demands.

Separately, the researchers identified significant delays in communication between the vehicles and roadside units, in part due to European data transmission standards. For V2X communication, transmission frequency is restricted to a maximum of 10 messages per second to avoid “message congestion.”

In this experimental setup, the vehicle’s perception system and data transmission system also operated asynchronously, meaning messages were sent at a fixed rate independent of when the perception system finished its analysis. This resulted in worst-case buffering delays up to 100 milliseconds. The researchers estimate that these end-to-end delays could be reduced by as much as 33 percent by synchronizing the systems, ensuring that data transmission is triggered the exact moment object detection is complete.

Iñiguez de Gordoa notes that while 100-millisecond delays may not seem so critical in an urban environment, such a delay can have more significant consequences in high-speed scenarios.

Next, Iñiguez de Gordoa says the research team is interested in exploring ways to ensure that the data vehicles are receiving during collective perception communications are trustworthy. For example, incorrect calibration of sensors or a failing GNSS system could result in inaccurate data being exchanged with surrounding smart vehicles and infrastructure.

Quantum networks could protect messages with theoretically hack-proof encryption and also help quantum devices operate in unison to boost their capabilities. For the most part, scientists have sought to upgrade these networks by increasing the distance or fidelity of their communications. Now researchers in China have a new way to dramatically scale up the bandwidth of quantum networks by increasing the number of simultaneous quantum communication channels.

Quantum networks share data that encodes the key details, or quantum states, of particles.This kind of communication could connect quantum computers to allow them to behave as more-powerful processors, or link quantum sensors together in arrays for even greater precision.

However, the data encoding quantum states is fragile and easily lost over great enough lengths of fiber optics. One way to overcome this problem is using quantum teleportation, in which the data essentially disappears from one place and reappears someplace else. Because the information does not travel across the intervening space, there is no chance it will get lost.

Scaling Quantum Teleportation Channels

Quantum teleportation relies on quantum entanglement to connect, for example, two photons, so that they can influence each other instantly no matter how far apart they are. Then, one of the two photons would stay in one location while the other photon would be moved to whatever destination is desired.

Next, a third photon is prepared for teleportation. This third photon is entangled with the first photon and then both are analyzed for their quantum states, which in the process destroys those states. That state data is sent to the destination, typically using light pulses, where it can be used on the other photon to re-create the third photon, so that it is indistinguishable from the original and functionally teleported. (Because the data is sent using signals such as light pulses, quantum teleportation can proceed no faster than the speed of light.)

Conventional quantum teleportation sends quantum states one at a time. This makes it challenging to transfer complex information, such as images. A large quantum network “requires many quantum channels to be generated, matched, and manipulated at the same time,” says Jietai Jing, a professor of physics at East China Normal University, in Shanghai.

Previous attempts to transmit multiple quantum states in parallel usually depended on multiplexing techniques that encoded multiple channels in a single beam of light. However, such techniques require decoding schemes that make it difficult to manipulate each channel independently.

The researchers in China instead devised a way to greatly expand the bandwidth of quantum networks. “We have scaled quantum teleportation up to 100 channels,” Jing says. “This architecture may contribute to high-capacity quantum communication.” They detailed their findings on 20 August in Physical Review Letters.

The new technique encodes a programmable, reconfigurable computer-generated hologram onto a device known as a spatial light modulator. The modulator passes a laser beam through this hologram in the form of a 10-by-10 grid pattern of distinct spatial modes that function as pixels. The modulator can control each spatial mode independently so they can each carry their own stream of data.

Separately, the sender and receiver share a pair of entangled light fields that constitute all of the light rays traveling in all directions. The sender entangles its light field with the hologram’s light field and transmits the resulting signal. The receiver combines this message with its entangled light field to enable quantum teleportation of all the original data at once.

In experiments, the researchers teleported a 100-pixel image of the letter Q. The new technique can support many channels simultaneously without requiring a complex set of devices to handle each channel, Jing notes. He adds that the team is now working to achieve a 1,000-channel system by using higher-power lasers and better amplifiers.

There are 17 rare earth elements. While almost all of them have vital industrial uses, you could be forgiven for believing, based on news coverage, that there are only four that matter: neodymium, praseodymium, dysprosium, and terbium.

To be sure, those are the essential ingredients in the powerful permanent-magnet motors that go into electric vehicles, heating and cooling systems, appliances, tools, fighter jets, bombs, and many other systems. But the industrial and national-security importance of rare earth elements goes far beyond magnets. Among the 13 other rare earths are two—erbium and yttirum—whose importance is no less vital, and which are now facing a looming geopolitical deadline this November. Depending on how negotiations play out, the post-deadline framework could erode U.S. national security—or redraw U.S.–China trade relations.

Erbium is used in the amplifiers that boost optical signals in the fiber-optic cables that crisscross the planet. Yttrium is a key component of the thermal barrier coatings that protect turbine blades and other structures from blistering heat in jet engines and in the combustion turbines so in demand to power new data centers.

Yttrium is more broadly used than erbium; besides thermal barrier coatings, it is also used in white LED lights and displays, microwave filters, and solid-state lasers, among other applications. Also, “small quantities of yttrium feed into trillions of multilayer ceramic capacitors that keep our electronics world going, including AI data centers,” notes Thomas Kruemmer, a rare-earths consultant based in Singapore.

Erbium and yttrium, Kruemmer writes in an email, are good examples of how “tiny quantities of rare earths can have an outsized impact by enabling core functionality in their respective applications.” Taken together, all of these rare earths “literally affect almost anything we can switch on and off.”

Chinese dominance of the markets for these two rare earths is essentially 100 percent. Take erbium: “There is no one that I’m aware of, outside of China, that makes erbium oxide on a commercial scale,” says Eric Bender, vice president of strategy and corporate development at Tronox, a mining, processing, and chemicals company headquartered in Stamford, Conn. Erbium oxide is the molecular form that is typically produced by separation facilities and bought by industrial users of the element.

For yttrium, too, industrialized economies are entirely dependent on China. China mines 90 percent of the world’s yttrium, but it processes essentially all of the world’s mined ore into an industrially usable form, such as yttrium oxide.

Yttrium coatings help turbine blades in generators and jet engines survive the heat. Sergii Zhmurchak/iStock

China’s Monopoly and the Yttrium-Erbium Supply Squeeze

Twelve rare earth elements, including erbium and yttrium, are subject to global export restrictions instituted by China in April and October 2025. The restrictions for some of them, including erbium but not yttrium, were suspended for one year in November 2025. Exports of yttrium, and a few other rare earths, have become erratic and sharply reduced. In March 2026, China approved a single, 60-tonne shipment of yttrium oxide to the United States, followed by a 10-tonne shipment in April. But overall, shipments of yttrium to the U.S. are down around 75 percent this year in comparison with 2025.

In November, China is expected to either extend or remove the suspensions for some or all of the 12 rare earths currently under export controls. And yet there is little movement apparent on the part of U.S. and European governments to cope with the possible cutoff of yttrium and erbium supplies. “Very few folks outside of China are talking yet about ‘How do we produce erbium, how do we produce yttrium?’” says Gareth Hatch, managing director of Strategic Minerals Advisory. “We’re lagging behind the geopolitics, the realities, of the day.”

The most important thermal barrier coating is yttria-stabilized zirconia. GE Vernova is one of the world’s largest manufacturers of turbines and a major user of the compound. At an investor day meeting last December, GE Vernova’s CEO, Scott Strazik, said the company had enough yttrium to last into 2026, but did not specify how far into 2026. An emailed request for elaboration, sent to GE Vernova’s media organization, did not elicit a response by press time.

Nevertheless, there are clear signs that the reduced availability of yttrium is having an effect on the industry. Oerlikon Metco is one of the largest of the small group of companies that produce yttria-stabilized zirconia and the equipment that applies it to the blades and other parts of jet-engine and combustion turbines. The erratic availability of yttrium has “thrown us upside down in the last year or less,” says Riston Rocchio-Heller, an engineer and lab manager at the company. Demand for combustion turbines in the United States is now at an all-time high, driven mainly by the need to power new data centers, including huge “hyperscale” facilities for AI training and inference and general cloud computing.

Yttria-stabilized zirconia is a ceramic produced by mixing yttrium oxide with zirconium dioxide. Alternatives to it exist, but some unique advantages of the yttrium-containing version have made it the standard choice, according to Roccio-Heller. Researchers are trying to find better alternatives, he says: “A lot of collaborative work is going on right now,” he reports. But it’s unlikely to come to fruition soon.

The erbium market is dominated by just a couple of applications: the coloring of consumer glass and erbium-doped fiber amplifiers. These amplifiers are spaced along long-haul fiber-optic lines, typically every 80 to 120 kilometers on a terrestrial cable, and every 50 to 80 kilometers for a submarine cable.

The amplifiers consist of a piece of erbium-doped optical fiber, usually between 10 and 30 meters long. The doped fiber is illuminated, or “pumped,” with laser light at 980 or 1,480 nanometers. That laser light excites the erbium atoms into a higher energy state. When a weak optical signal enters the length of fiber, it triggers the excited erbium atoms, which fall to lower energy states and release photons in the exact phase, direction, and wavelength—1,550 nm—of the incoming signal. That wavelength corresponds to the spectral “window” in which silica optical-glass fibers cause the least loss of the signal. That’s why nothing but erbium will do here: It naturally re-radiates at the necessary wavelength.

Engineers are starting to use erbium-doped amplifiers in the free-space optical communication systems that connect low-Earth-orbit (LEO) communications satellites, such as the Starlink satellites, with each other and with ground stations. Groups of these LEO satellites are being launched into orbit on an almost daily basis. Because their signals don’t go through glass, satellites could use a much wider variety of wavelengths. But engineers are using the erbium amplifiers anyway to take advantage of their relatively low cost and the large existing industrial base for the 1550-nm hardware, says Michel Corriveau, a researcher at MPB Communications, in Pointe Claire, Quebec, Canada, a leading provider of erbium-doped fiber amplifiers.

Corriveau says MPB obtains erbium-doped fiber for its amplifiers from a couple of companies, including Lightera (formerly OFS) in Denmark, which is now a business unit of Japan-based Furukawa Electric. An executive at Lightera, Annette Lundby, declined to comment on the company’s ability to secure stocks of erbium.

Project Vault and a geopolitical deadline

Erbium and yttrium, along with 14 other rare earths and a couple of dozen critical minerals, are stockpiled by the U.S. government. This National Defense Stockpile is managed by the Defense Logistics Agency, which does not comment on its specific contents. In February 2026, the Trump administration announced a US $12 billion public-private initiative called Project Vault to stockpile rare earths and other critical materials. The funding includes $10 billion in financing from the Export-Import Bank of the United States—the largest single financing in the bank’s history.

Curtis Stough, a “defense stockpile liaison” at the Defense Logistics Agency, did not respond to a request for an interview. The U.S. Geological Survey, which tracks availability of critical minerals, also declined to make anyone available for an interview. The White House press office, too, did not make anyone available. But spokesman Kush Desai did provide a statement, which read, in part: “The Administration continues to take a multifaceted and nimble approach to reshore critical mineral supply chains back to the United States.”

That approach will surely be tested in the event that China’s 2025 export restrictions take full effect, which could happen on 10 November. But a lot will probably happen before then, says Hatch, making the outcome hard to predict. “This whole thing is ostensibly about dual-use control of materials” that have essential military uses. “But in reality, it’s part of the bigger trade picture.” So Hatch expects a wider discussion, leading up to November, involving leading-edge GPU chips, extreme-ultraviolet lithographic equipment, high-bandwidth memory, design-automation software, and other tech goodies that the United States and its allies are currently denying China.

How do I touch you
across the ocean,
across cold depths
where light travels through glass.

Not copper—fibers. Optical.

Through liquid glass, through flickering light
that carries you in fragments. Light broken into pulses.

You say: it’s easier this way. What are we missing like this? You smile.
Safe distance.

I say: network.

Signals slide beneath the sea, through cables thinner than trust, faster than touch,
slower than longing.

We stand alone, together. Synchronous, yet apart. Icons replace skin, latency replaces breath.

This distance protects us. Silence that feels intentional.

Everything is under control as long as nothing truly hurts.

And we choose it
because it shields us
from what we might become if we actually met.

You are my counterpoint. My response.
My reflection
at a safe distance.

Beneath the ocean, nodes remember paths.

Packets shake hands without bodies.

If we get lost,
we resend everything, with error,
with noise,
with hope.

Large language model (LLM) hallucinations are becoming a profitable side business for cybercriminals.

They prod the AI models to answer prompts with nonexistent but plausible-looking URLs, registering the domain names that recur often. Then the criminals set up malicious websites on the fake domains.

In June, the Santa Clara, Calif.–based security firm Palo Alto Networks reported that it had prompted LLMs to generate 2.1 million URLs that resemble the names of major brands. Of those, more than 13,000 pointed to domains already registered and confirmed malicious, while roughly 250,000 commonly hallucinated domain names remained unregistered—sitting available for scammers to claim.

Last year, other security specialists detected a similar threat, from scammers who prompted LLMs to hallucinate plausible URLs for existing software packages. These AI “slop” links took users to counterfeit packages with scammers’ code—ready to be added to a project with a click.

Scammers who engage in this “slop squatting” have plenty to work with: Last year, a study by Joseph Spracklen—a Ph.D. student at the University of Texas, San Antonio—and colleagues found that nearly 20 percent of LLM-recommended software packages were fictitious.

Seth Michael Larson, who coined the term “slop squatting” last year, is a security developer–in-residence at the Python Software Foundation, which oversees the programming language that is used worldwide by some 23 million people. Larson spoke with IEEE Spectrum about “slop squatting,” “phantom squatting,” and how security work is changing in response.

Slop Squatting in LLM-Generated URLs

What’s your definition of “slop squatting”?

Seth Michael Larson: It’s targeting a pattern in what an LLM is recommending. For lots of users asking for a particular resource, the model will answer with the same hallucination some percentage of the time. Users implicitly will take what an AI tells them to be authoritative. They’ll assume some security trust decisions have happened. But they haven’t.

What made you decide to call this slop squatting?

Larson: That term betrays my feelings about these systems and their marketing. We’re not giving enough warnings to users that the output is not something you should just blindly accept. People need to understand: The trust boundary between this information and your computer is you.

What percentage of malicious packages are sitting on slop-squatted URLs?

Larson: As people operating a package repository, we don’t map a name back to see if an LLM generated it. We could, but it’s expensive to do, and where it comes from matters less to us than taking it down.

But we can do the correlation. Say [an erroneous] URL didn’t exist before, and now suddenly, in the past two years, its traffic is huge. If the graph went up two years ago, it’s probably due to LLMs.

Slop squatting is aimed at coders looking for packages. But phantom squatting targets anyone looking for advice about stores, banks, etc. So is phantom squatting a bigger problem for society?

Larson: Probably. Due to the fact that it targets users who are not necessarily expected to have any sort of security knowledge.

The cybersecurity firm Palo Alto Networks recently studied websites at domain names whose URLs are frequently hallucinated by AI chatbots, discovering at some of these sites a range of malware and other malicious code.Source: Palo Alto Networks

Protecting Yourself from AI Phishing

What should AI users do to protect themselves?

Larson: My recommendation with any AI output is: You should not be trusting this. As in any engineering field, you can’t just go off of vibes. You have to have a way to verify. How do we know that this result is correct? Have that in your mind when you’re using these tools.

That said, I don’t look to individual action for this. I think it is more about infrastructure. People can’t hold everything in their head all at once. In medicine, engineering, lots of areas, experts handle the hard cases for people. If you don’t do that, society just doesn’t function.

I try to push people in security work to actively help users. Don’t cross your fingers and just hope users are doing the right thing. You have to set up infrastructure so that users can’t not do the right thing. We can’t expect every single person on the planet to try to keep up with this rate of change. We need to be there to help them through that transition.

So what new security infrastructure needs to be put in place?

Larson: We haven’t been thinking about authentication in relation to LLMs, which make it so cheap and easy to create perfect phishing websites for pennies. We’re just at the very beginning of phishless or phish-resistant authentication technologies. These are things like WebAuthn. Those are promising. But there’s still this long, long tail of adoption ahead.

Right now, even things like password managers are a completely foreign concept to 90 percent of people. It’s still difficult to get people to even use a two-factor authentication method like TOTP [a one-time password] or even a text SMS code. People really don’t like the technology because it makes using digital services harder. So we just haven’t gotten there. Hopefully safety plans will get accelerated now.

How have these new uses of LLMs for phishing changed your work day?

Larson: The work of getting a vulnerability report and triaging it and potentially fixing it and rolling out fixes to everyone—for that, the numbers are three to four times what they were just a year ago. On a medium-size Python project, you might have processes set up to deal with a vulnerability eight or 10 times a year. And now suddenly you’re dealing with 40 or 50 or 60. You have to think, Oh boy, we need to fix some of these processes here. It has been a really quick ramp-up. Almost every open-source Python project is dealing with this at some scale. It’s a lot of extra pressure.

Evolving Security Processes with LLMs

So, how should security processes change?

Larson: It would be great if security work was more like other sorts of open-source contribution, where basically anyone can participate who is interested and has the ability and the time. Right now, this isn’t possible. Security has involved secrecy. But if a publicly available LLM can discover a vulnerability in an open-source codebase, the value of secrecy is very diminished. It means anyone who has $5 and access to a public model could have found that vulnerability.

So as LLMs make it easy for people to scam, they might also make it easier for people to be a white hat?

Larson: Maybe. This is all very new. I know the Linux kernel is experimenting with this idea, and at Python we’re also thinking about what this means for security teams. Right now in security work, you’re in, or you’re out. There’s no in between. There’s no step-by-step where you do the work one time and then decide whether you like it, whether you want to do more of it. Maybe we can give you the first rung on a ladder toward these types of contributions.

Will AI agents make this problem worse? After all, an agent could accept a URL and download a malicious package without its human user even knowing.

Larson: I’ve never used an agent. And the number of times I’ve asked an LLM to generate code for me is probably in the single digits. I’m not exactly a fan. I would definitely say to be careful with agents. I know that that’s not going to jibe with the pace people are expected to deliver code at these days. But you need to always have output that you evaluate. Keep yourself as the driver.

Maybe we can learn some things from other engineering focuses. For example in [rail and airplane] transit, there is the practice of saying out loud the thing you are doing, to keep yourself in the loop. In software now, we need practices that keep you engaged, keep you thinking critically about what’s actually happening, and not just clicking the accept button.

The history of networking is full of tools that repurposed solutions to very different kinds of problems first. Wi-Fi’s origins trace back, in part, to a team of Australian radio astronomers trying to detect signals from evaporating black holes. But the data-processing tools they’d developed also proved capable at extracting clean messages from any chaotic, echoing signal environment. Echoes are echoes, after all, whether from distant star systems or from the far corner of the house.

I research vehicle communications networks, connecting cars to cars and to transportation infrastructure like traffic lights—for tomorrow’s vehicle-to-everthing (V2X) networks.

V2X research has long relied on models that assume “perfect” or “ideal” network conditions, which is a simplifying assumption that makes the math tractable. But this assumption doesn’t reflect how real wireless signals behave in a moving, obstructed, high-density environment. That gap is exactly the kind of real-world unpredictability that open radio access networks (a.k.a. O-RAN)—an open, programmable architecture behind some 4G and 5G cellular networks—were built to manage.

So why has the O-RAN standard—which is open and available to be applied well beyond 5G telecom—never been used for vehicle communications?

Solutions to the vehicle-to-everything (V2X) problem have to date relied on new networking protocols built from scratch—only to discover chicken-and-egg problems, thorny standards wars, and real signal congestion challenges at scale.

By contrast, O-RAN allows V2X engineers to reuse the networking protocols already developed for cellular communications. O-RAN was developed assuming cellphone towers are generally fixed in place. But, as can be seen below, O-RAN accommodates mobile “towers”—cars and trucks, in this case—with little additional effort.

Imagining a New Way to Connect Vehicles

Self-driving vehicle technology has largely been an each-car-for-itself endeavor. Tesla’s approach, for instance, relies heavily on powerful on-board banks of computers and suites of sensors spread around the car.

However, as an alternative to the “data center on wheels” model, this new O-RAN approach to V2X relies on each car’s nearby neighbors, wherever they are on the road. Each O-RAN–connected vehicle can then use a diversity of cars’ sensors and viewing angles for better group coordination and decision-making.

There is, to be clear, no O-RAN V2X test network operating in the world. Not yet.

It was just 10 years ago that the Third-Generation Partnership Project (3GPP) released its initial cellular V2X standard. The 3GPP have refined V2X over three major releases since. In the U.S. and the EU, the FCC and related European agencies have put forward other standards for short-range wireless V2X communication protocols.

However, no consensus standard has yet emerged. So, lacking any clear, unambiguous guidance on the future of V2X networks, autonomous-car makers—like Waymo, Tesla, Zoox, and Cruise—have leaned more on self-reliance, bulking up each vehicle with as many sensors and GPUs as possible.

Here, though, is where O-RAN might be able to help.

A little like APIs (a.k.a. application program interfaces) connect one app to another on your smartphone, O-RAN serves as an API for the network itself. And because of O-RAN’s open standards, a wireless network becomes programmable, vendor-neutral, and open to custom applications called xApps.

To test our proof-of-concept framework, I have been part of a team simulating five minutes of O-RAN V2X network traffic over one square kilometer of urban area, using real buildings and real-world road layouts from OpenStreetMap and traffic patterns generated by the modeling package SUMO. The simulations assumed a traffic density of 50-70 vehicles per kilometer—not rush hour but not light traffic either. In our simulation, we assumed vehicles communicated via a millimeter-wave frequency of 28 gigahertz and that each component of our O-RAN V2X system had its own dedicated xApp.

Taken together, these inputs—real geometry, real traffic, and each vehicle’s live GPS position—constitute what network researchers call a digital twin of the urban environment. That’s a virtual replica detailed enough for the network to reason about the physical world in real time.

This virtual world gave us a real result, too.

The simulations, published recently in IEEE Network, revealed that existing V2X standards—in which cars uncoordinatedly spit out messages into the network—result in signals “talking” over each other some 80-100 percent of the time. However, using O-RAN signal coordination, the message “collision” rate dropped to near zero.

And that matters because a seized-up V2X network doesn’t just fail quietly. It can fail in ways that might make a road turn treacherous.

How O-RAN Can Coordinate V2X Traffic

High-frequency data links between cars are already difficult to maintain, even on a clear day with no buildings or city infrastructure getting in the way.

Yet, in this situation, existing V2X networks leave a car to conduct blind searches for each dropped signal beam. Traveling at highway speeds, that search takes long enough for the surrounding world to change completely.

An O-RAN network continuously tracks signal conditions across the network, and in O-RAN V2X simulations, we also gave the network access to a detailed map of the urban environment—building positions, road geometry, intersection layouts—combined with each vehicle’s GPS trajectory. Together, these parameters let the network’s control layer predict where and when a signal link is about to fail and instruct each car’s antenna to adjust before the connection drops.

Signal pointing is one failure mode. Losing the connection entirely—because no direct path exists at all—is another.

Consider, for instance, a crossroads of two busy streets, with a few alleys and parking lots adding to the list of potential dangers.

If a signal from car A cannot reach car B directly, or if the path length is too far for an individual beam to travel, the signal must find an intermediary car or stationary sensor nearby that can pass along the message. And existing V2X standards are slow and reactive—polling potential relay vehicles one-by-one: Are you available? Can you redirect this message?

By contrast, O-RAN keeps a running graph of optimized message routes, accounting for a range of real-world constraints. So when an O-RAN link fails (whether that link is direct from sender to receiver—or indirect), the system already has a reroute mapped out.

This is partly why we included “multi-hop routing” in the O-RAN V2X simulations.

Multi-hop V2X O-RAN routing complicated three separate elements of the simulation: for each signal’s middleman (some cars may be ideally positioned to relay a signal from car A to car B, but we made the simulation neglect any cars that were also overwhelmed with their own signals and signal-processing needs); for each signal’s strength (we required that every intermediate link be able to maintain a stable network connection, factoring in distance and traffic conditions); and for each signal’s latency (we required a realistic accounting for added signal latency time for each additional hop in a multi-hop routing).

And with each added complication, O-RAN V2X multi-hop routing continued to extend the network’s capacity from 25 percent of nearby cars connected (without multi-hop) to nearly 100 percent (with multi-hop).

These complications, at least at the simulation level, did not slow down the V2X network.

How Could O-RAN Ever Be Scaled Up for the Real World?

We are in touch with potential collaborators and institutions to develop testbeds, prototype hardware, and tester vehicles for potential proving grounds. The Institute of Science Tokyo, for instance, has already expressed interest in working on some of these early-stage problems.

To date, our published research on O-RAN V2X has centered around a computer simulation only. Real-world hardware will undoubtedly surface challenges our simulation could not. So, questions of network latency and the computational overhead needed for O-RAN V2X signaling remain as yet unresolved.

Plus, concerns about full interoperability and realistic security will each demand their own investigations. After all, no one will trust a V2X network to do anything if that network’s cyber vulnerabilities haven’t been anticipated and patched in advance.

Realizing the O-RAN V2X vision will require progress on multiple fronts simultaneously. On the standards side, O-RAN’s vehicular extensions—the interfaces that allow vehicles to participate in the network as managed elements rather than passive users—would ultimately need to be formally adopted by the O-RAN Alliance and recognized by 3GPP’s V2X specifications. That process takes years.

On the industry side, there is a more immediate problem that our architecture is already positioned to solve: interoperability.

Today, a car made by one manufacturer cannot necessarily parse V2X sensor data sent from a car made by another. Firmware is proprietary; data formats differ. But an O-RAN control layer would act as a universal translator—normalizing each vehicle’s data into a common format and accelerating a push toward true multi-platform vehicle-to-vehicle communications. A more widespread and truly universal standard would, by itself, represent a substantial step forward for V2X.

Weather forecasting today runs on rivers of data. But there’s almost no data available from vast stretches of the planet—open ocean, polar regions, and remote sections of the world. That’s a problem WindBorne Systems, based in Redwood Shores, Calif., aims to tackle with its long-duration weather balloons. The company’s CEO, John Dean, wrote for IEEE Spectrum last year about the AI models (called WeatherMesh) that WindBorne is training based on ongoing observations from some 600 balloons the company keeps aloft worldwide.

WindBorne Systems’ CEO, John Dean, says the company traces its origins to a Stanford University project that pushed the limits of weather-balloon design to maximize time aloft. WindBorne Systems

The company’s fleet of balloons generate thousands of hours of fresh weather readings each day, including from regions of the world that might otherwise go unsampled. One balloon, called Elder, recently set a company record by staying airborne for 345 days. Other balloons have demonstrated something different: falling to the ocean and continuing their mission now as floating weather buoys. WindBorne has provided Spectrum an exclusive look at this new “sky-to-sea” capability below, including detailed flight tracking and recovery photos.

Looking ahead, WindBorne’s near-term plans are ambitious. Dean told Spectrum that the company expects that “in a year from now we should have thousands of buoys floating over the ocean, reporting data.”

IEEE Spectrum spoke with Dean about how WindBorne’s balloons will soon be pulling off a dual life between sky and sea, what made the record-setting Elder flight possible, and the fine art of reading the atmosphere from scratch.

From Weather Balloon to Weather Buoy in One Device

IEEE Spectrum: What actually happened on the weather-balloon-to-ocean-buoy flight?

John Dean, CEO of WindBorne Systems: We have a couple of balloons now that fly as balloons and land as buoys. The main difference in hardware is just the ability to be fully waterproof and then still be able to communicate to satellites while on the ocean surface. We’ve been playing with this concept for a while. And now we have roughly half a dozen prototypes that have this ability.

WindBorne Systems has launched half a dozen weather balloons that have transitioned to ocean buoys mid-mission. The first such flight was in March 2025. WindBorne Systems

Was making your weather balloons able to land in water always the plan?

Dean: Yes. We have balloons that land all over the globe. We have now over a hundred flights that have landed on land and then work as ground stations. Basically if the balloon lands, and there’s no moisture on the ground, it can talk for many weeks after landing. But in the buoy case, every flight we’ve done has been intentional because it requires waterproofing. But in a couple months from now, every balloon that we fly will be capable of this.

So if a balloon gets damaged flying into a dangerous storm, that maybe isn’t a flaw—it’s part of the design?

Dean: Yeah. We’ve flown into a number of tropical cyclones now over both the Atlantic Ocean and the Pacific Ocean—where we’ll fly a balloon over the storm. And then we’ll either drop what’s called a “dropsonde”—a small sensor from the balloon into the storm which gives you like one sounding of data—or we will basically sacrifice the balloon.

[Because] the balloon can’t fly very long inside a hurricane. You can fly overhead, but you can get one to two soundings inside the storm system with the balloon. And so for severe weather events like tropical cyclones, you’re able to use our balloons to get additional data without having to send in a crewed mission. We’re able to lower the cost per data point dramatically.

And we’ve now proved out both the ability to collect data throughout the entire troposphere and then land on the ocean surface and collect data on the ocean surface.

WindBorne CTO Andrey Sushko holds up an ocean buoy that was once the payload of a WindBorne weather balloon. WindBorne Systems

Is data from the ocean surface actually useful for your weather-forecasting models?

Dean: We collect wave height, wave information, ocean pressure, and ocean temperature. And if you want to know the weather one week from now, because the sky and the ocean are very coupled together, it’s actually one of the most useful assets you can have.

Elder’s Record Flight and What Broke the Bottleneck

WindBorne’s Elder balloon flew for 11 months, covering more than 300,000 kilometers. What if anything surprised your team about the mission?

Dean: That mission happened after we made a design change to our vent mechanism. We used to have like very small leaks that would limit the duration of a mission to a couple months max. Roughly a year ago we remade the valve not to leak at very low temperatures. And [Elder] was the first balloon with that change. That’s what unlocked this almost-one-year flight.

WindBorne Systems’ weather balloons are designed to maximize time aloft, in one case logging a record-setting 345-day flight. WindBorne Systems

Do these balloons use helium?

Dean: Actually, most of our flights are hydrogen. It’s cheaper, and there’s a helium shortage. Also, we have a launch site on Cape Verde, an island chain off the west coast of Africa. It’s pretty hard to source helium there. Hydrogen is much easier to source.

What finally brought Elder down?

Dean: Ballast exhaustion, which is what takes out most balloons.

Every balloon has a fixed amount of gas and a fixed amount of ballast. You drop ballast to go up, you vent gas to go down. Eventually, you run out of ballast, and you lose the ability to rise again. You’ll eventually fall from the sky. And so flight was a normal termination just due to lack of ballast. But what was unique with that flight is the rate of ballast use was so low that it was able to fly for 11 months.

Building Better AI Models to Predict the World’s Weather

Walk me through how balloon data actually becomes a better weather forecast.

Dean: All these balloons, they talk to Iridium satellites. Then it goes from the satellites back to the ground. On the ground, we do quality control. Then we sell that data to customers and also run it through our own AI-based weather model. We’ll use our balloon data, [other] weather balloons, ocean buoys, ground stations that NOAA operates, and satellites. [It’s] all the freely available weather data plus our balloons.

Our balloons are the only source of ground truth for temperature, pressure, humidity, wind speed, and direction over remote areas. So for many storm events, basically the only ground truth we have is from our balloons. And then WeatherMesh ingests all this data that tells WeatherMesh what the current state of the weather is. And that allows it to make better weather forecasts over the next week.

What kind of accuracy gain does that produce?

Dean: At the scale of our constellation today, we see roughly a 10 to 20 percent accuracy improvement from assimilating our balloon observations into it. So if you look at a temperature forecast over the United States on a three-to-five-day time horizon, we’ll see a roughly 10 percent improvement in forecast accuracy.

Also NOAA found that for hurricanes, our balloon observations improve hurricane ground tracking by about 10 to 20 percent as well. Ten to 20 percent is not a huge number, but for improving forecasts actually it’s quite a big deal. And our constellation today is a factor of a hundred smaller than what it will be in the future. When you fly balloons all over the globe and collect data in these remote places, that moves a needle on forecast accuracy.

WindBorne Systems’ weather balloons operate on hydrogen gas, rather than helium, both for cost savings and the potential to scale up company operations. WindBorne Systems

Are you feeding balloon data into an existing AI weather model, or did you build your own?

Dean: We train AI weather models from scratch—Earth-foundation models, completely designed by us and trained by us. Because there’s a lot of room to improve just in AI techniques for weather modeling. Also most AI weather models you might have heard about—for example, DeepMind’s GraphCast—don’t really do data assimilation. That’s where you take all the measurements you can find over the last couple days and then use that to make the best guess of the current weather conditions. Then you forecast that forward in time.

Most models don’t do data assimilation to actually build the best guess of reality. But WeatherMesh is fully end-to-end. It takes observations in, and it makes forecasts out. That’s important because AI-based data simulation can be much more accurate than traditional data simulation.

Where Does a Weather Balloon Startup Go From Here?

Briefly put, what makes WindBorne’s approach different from 50 years of weather ballooning?

Dean: Rather than flying for 2 hours, we can fly for over two weeks. And rather than the hardware landing and becoming waste, a few more changes to make it waterproof mean you can get more value out of the hardware. That’s why I started the company, to take this traditional weather balloon and rather than giving up after 2 hours, you make the hardware last longer. Then whether it’s in the sky, in the ocean, or on land, it allows you to collect more weather observations for much lower cost and less waste.

How did this novel approach originally develop?

Dean: All the cofounders met in college at Stanford, and we were part of this group called the Stanford Student Space Initiative. Actually most of the founders worked at SpaceX as interns back like 10 years ago in college. We’re all very interested in space.

But if if you’re interested in in space and you want to do hands-on engineering, you can’t really do anything fun with rockets. Because everything cool with rockets becomes controlled. If you make the rocket controllable, it becomes classified as a guided missile. And satellites are very expensive, with long lead times. But like, balloons are almost in space, and you can do them on a student budget.

So Andrey [Sushko], our CTO, he was actually the person who founded the project that became WindBorne. The whole point of the project was to make the balloon last for a long time. And we worked on that for basically four years of college, just purely for fun. Not for any company, just out of passion. And then by the time a bunch of us were graduating, we’re like, hmmm, this platform is actually pretty useful.

What are the company’s biggest technical goals in the next 12 to 18 months?

Dean: Six months from now, we should be over a thousand [balloons]. A year from now, we’ll likely be 2,000 or so. And then 18 months from now we’ll be at 4,000. We target doubling every every six months.

Dean’s law of balloon scaling?

Yeah—shorter than Moore’s Law.

Today’s advanced 5G networks, and tomorrow’s 6G networks, are now moving toward an increasingly AI-centered future.

This September in Madrid, the global telecommunications standards body, the 3GPP, will meet to decide how present-day 5G networks migrate to the emerging 6G standard, slated to begin rolling out by 2029. On the docket for Madrid are important decisions that will set the direction for how, and how deeply, AI gets built into 6G’s radio equipment.

The stakes are high. If the industry gets it wrong, network operators could end up paying a premium for new AI-enhanced wireless infrastructure that doesn’t lower costs, hasn’t been proven to survive years of real outdoor conditions, and may not do much beyond what existing, cheaper technology can already do.

John Strand—an industry analyst based in Copenhagen who has consulted with telecom network providers around the world for three decades—says the clearest example of AI’s inroads is a US $1 billion deal last October in which GPU titan Nvidia invested in Nokia, the telecom infrastructure company based in Espoo, Finland.

As described by Nvidia’s press release, Nvidia wants the chips inside Advanced 5G and 6G radio access networks (RANs) to look less like specialized telecom hardware and more like the servers running ChatGPT.

Nokia and Nvidia—together with T-Mobile—have done proof-of-concept trials of their proposed system. Earlier this year at T-Mobile’s innovation lab in Seattle, a radio and a Nvidia server successfully handled both a live 5G connection and AI tasks like video streaming and captioning at the same time. It was a working demonstration—although a controlled one, on one radio, at one site. Nokia’s timetable calls for broader commercial trials starting later this year, with commercial rollout expected next year.

The deal could be a win for Nokia, because the company “can get a higher price for their RAN, which have declined in price for the last 25 years,” Strand says. By adding top-tier AI hardware to its systems, he adds, Nokia can “increase the entry barrier for competitors to move into this market.”

Nokia isn’t alone in seeking an AI-fueled future for 6G. Ericsson, the Stockholm-based telecom infrastructure maker, has been running its own advanced AI-in-the-network trials with T-Mobile since early 2025, and in June began selling a software upgrade that adds AI directly into existing radios and base stations—no new hardware required. Huawei, the Chinese equipment giant, has made similar moves, unveiling AI tools this year that let networks diagnose and fix themselves automatically.

What Is AI-RAN Needed For?

Despite the AI test cases, Strand says he hasn’t yet heard a convincing case for what’s called “AI-RAN”—a case that would explain why cellphone network operators would want to “increase [their] spending on RAN and, at the same time, maybe also increase their energy consumption.”

Wireless engineers have spent decades squeezing radio networks close to their theoretical limits of efficiency. “It’s hard to see the economic justification” for using AI algorithms broadly in 6G network RAN, says Kim Kyllesbech Larsen, chief technology and information officer at the Hoofddorp, Netherlands–based telecom operator United Group.

Companies like Seattle-based Opanga Networks, Larsen says, already deliver some of AI-RAN’s strived-for efficiency gains using today’s network architecture—no new AI-RAN hardware needed.

Larsen says the trials run so far by Nvidia, Nokia, and others “are important engineering milestones and show that the concept works,” he says. But they still fall short in some ways: The trials prove that AI computing workloads can run alongside the RAN—sharing space and power with it—not necessarily that AI needs to be deeply integrated into the RAN’s core architecture. The latter is a different, harder problem, he says, which demos conducted to date haven’t resolved.

“AI-RAN will ultimately have to demonstrate that it delivers capabilities and/or economics that cannot be achieved by simply making existing RAN architectures smarter,” he says. “Until that is proven, operators should evaluate AI-RAN pragmatically...rather than assuming that a new architecture is automatically a better one.”

The risk isn’t just unproven hardware, Larsen says. It’s also coordination. Picture several capable managers all trying to improve the same business, he says. Each judges on a different goal. One minimizes energy use, another maximizes speed, a third maximizes coverage. Individually, each makes reasonable decisions. Together, however, they can end up working against each other. The same risk, Larsen says, applies to a network run by multiple independent AI systems with no clear chain of command.

Where Should 6G’s Intelligence Live?

Larsen says the ultimate question isn’t whether or not AI belongs in 6G. Rather, it’s where inside the network should AI algorithms actually run?

“It’s important to distinguish between AI hardware and AI algorithms,” he says. He adds that he’s seen compact machine learning algorithms that increase wireless network efficiencies and work without requiring heavy-hitting GPUs. “These algorithms execute…in microseconds to a few milliseconds and are well suited” to more modest chips, he says.

Larsen ran his own simulations of possible 6G AI functions to determine how quickly each one needs to react—some in microseconds, others with more breathing room. Only the fastest-reacting functions, he found, need to be executed on hardware running at the tower itself.

So some of the speedups that AI might help wireless networks achieve can, in fact, be computed off-site, “rather than assuming every [cellphone] tower will become an AI data center,” he says.

Merouane Debbah, a researcher at Khalifa University’s Digital Future Institute in Abu Dhabi, has run some of the only live trials testing AI hardware in real radio conditions. He says that the fight over 6G’s AI future is not just about whether to place AI compute at the tower or instead to rely on far-away cloud-computing data centers.

“My expectation is not that 6G will place one giant AI model inside every tower,” Debbah says. “A more credible architecture is hierarchical and heterogeneous: very small models inside radios and basebands for hard, real-time decisions; more capable models at…edge sites; and large foundation or agentic models at regional or central levels for reasoning, planning, and coordination.”

Larsen says as 6G standards begin to mature, so too will the expectations placed on them by network designers.

“Ultimately, I trust good architecture more than individual hardware components,” he says. “AI-native RAN will succeed by placing the right intelligence in the right place, with the appropriate authority, latency, and safeguards, rather than trying to make every part of the network equally intelligent.”

In February 2024, a young man lay somewhere on the frozen shore of James Bay, Canada, surrounded by snow and darkness, succumbing to hypothermia. When he failed to get home on time, his frantic mother sent a Facebook message to Elizabeth Kataquapit, then chief of the indigenous community Fort Albany First Nation in northeastern Ontario. Kataquapit used Facebook to alert the community’s search-and-rescue squad, who jumped onto their snowmobiles and drove into the night. Before dawn, they returned with the dazed man, who told rescuers he’d given up until wolves nudged his hypothermic body. “The wolves told him to wake up,” Kataquapit says. “I really believe they saved his life.”

A fiber-optic network also played a key role.

Not long before the young man’s mishap, the indigenous-owned Western James Bay Telecom Network (WJBTN) had built its own fiber-to-the-home network in this remote Cree community, some 975 kilometers north of Toronto. Before the network, the rescue squad relied on a few handheld radios to pass information along. This time, a Facebook message to the full list of volunteers triggered the search.

An aerial view shows the remote community of Fort Albany in northern Ontario.Gavin John

Building Canada’s first fully Indigenous-owned-and-operated fiber-optic network was an uphill battle for Brian Nakogee, WJBTN’s finance officer, who had to secure capital from agencies less familiar with the challenges of remote northern life. For the people of Fort Albany First Nation, accessing many vital supplies and services means traveling about 500 km to the regional city of Timmins. By land, the trip is possible for only a few weeks each winter, when the swampy tundra freezes hard enough to construct a temporary road. “The southern way of doing things is very different than how we here in remote areas piece things together,” says Nakogee.

Telecommunications giants long saw little profit in serving the subarctic coast of James Bay. But even as satellite internet started to become available in remote communities, WJBTN staff saw the value in building and owning its own hard-wired internet service instead of relying on outside companies. Today, the nonprofit operates one of the fastest networks in Canada, while keeping both infrastructure ownership and revenue within the First Nations communities it serves.

WJBTN is part of a broader movement among Indigenous and remote communities seeking more control over their telecommunications infrastructure. In the United States, 30 tribes now operate fiber-to-the-home networks, many launched during COVID. Canada has since created a dedicated Indigenous broadband funding stream. And as more communities pursue the expertise and funding to build their own networks, WJBTN’s experience offers a blueprint for what locally owned connectivity can look like in some of the hardest places to serve.

How a Power Line Became a Broadband Backbone

While many towns in North America were connecting to optical fiber in the early 2000s, the subarctic communities were left out.

The residents of Fort Albany saw the earliest sign of improvement in 2008. That’s when their locally owned power company, Five Nations Energy Inc. (FNEI), strung fiber-optic cable on the utility poles that were delivering electricity from the dusty railway town of Moosonee, 135 km away across the peat bogs. On the shore of the Moose River, Moosonee is the last stop for Ontario’s telecommunications providers. Its only link to the province’s highways is a 5-hour train ride that shuttles passengers, freight, and vehicles through the Boreal forest.

Elizabeth Kataquapit [top], former chief of Fort Albany First Nation, says high-speed internet has transformed her community. Search & Rescue volunteers now coordinate emergency responses through a Facebook group [bottom]. Gavin John

Soon after, regional Cree leaders formed WJBTN to provide high-speed telecommunications in Moosonee and the three Indigenous communities to the north. WJBTN would lease the fiber-optic backbone from the power company, with just 1 gigabit per second of total capacity for the entire population of about 6,000 people.

But the new fiber backbone did not mean fast internet for residents. One of WJBTN’s first commercial clients was a telecom company called Xittel, which used microwave links and local access points to beam wireless internet to homes across each town. It wasn’t what subscribers were hoping for. This system frustrated users with delays, glitches, and strict data caps. Kataquapit calls it a “turtle.”

Everybody complained about the wireless. Xittel advertised a 10 megabit-per-second connection, but speed tests consistently showed 3 Mb/s for both uploads and downloads. And customers paid dearly if they ever exceeded their data limit. “You were billed close to CA $7 per megabit,” Nakogee recalls.

WJBTN’s dream had always been to connect everyone’s home to fiber optic and make it affordable. Without a technical team, roads, or any major funding, the company just had to figure out how.

Designing a Fiber Network for the James Bay Coast

When Nakogee joined WJBTN in 2014, the telecom company was still in its infancy. It was “a department huddled in the corner, trying to latch onto the services of FNEI,” he says.

Nakogee was asked to prepare a proposal to deliver 40 Mb/s fiber connections to each of the roughly 1,000 homes and businesses spread across 300 km of the James Bay coast. Back then, WJBTN operated on revenue from its early commercial and institutional customers—including Xittel, local government offices, hospitals, air navigation facilities, and family service centers that were connected to the first few strands of fiber. To make a residential fiber network possible, WJBTN first had to build both revenue and trust within the communities it hoped to serve; it had to convince people that the small new organization would follow through on its plan.

WJBTN finance officer Brian Nakogee played a key role in financing and planning the community-owned fiber network. Gavin John

Nakogee especially needed support from Moosonee and Attawapiskat, the communities at the start and the end of the line. “They’re the bread that holds this sandwich together,” he says. Through years of diplomacy, the budget grew enough to support a business proposal for a fiber-to-the-home network. Together with engineer Dirk MacLeod, in 2015 Nakogee began hashing out the details of a bare-bones version of the system. The plan required both upgrading the network’s long-distance fiber backbone—known in telecom as the backhaul—while also building the local infrastructure that would connect individual homes to the internet.

Andrew MacLeod, WJBTN’s field project manager, reviews a map of the fiber network.Gavin John

The first step was upgrading the network’s backhaul using newer “coherent optics” technology, which can transmit much larger amounts of data over long distances, explains Dirk’s brother, Andrew MacLeod, another network engineer, who joined as a consultant. Using equipment from telecom supplier Infinera, the new backhaul would deliver 100 Gb/s to distribution points in each town, with redundancy in case a fiber line failed.

Next came the challenge of connecting individual homes. Rather than extend a direct line from the central office for every customer, the engineers designed the network around a telecom architecture called GPON (Gigabit Passive Optical Network), which reduced the fiber needed to connect each customer to the network. Fiber from the backbone would run to networking equipment at each town’s electrical substation, where passive optical splitters would distribute the connection among many households without requiring powered equipment at every junction. In remote communities where maintenance and repair are difficult, reducing the amount of active infrastructure was a necessity.

The total estimated cost was CA $4.7 million, or CA $4,700 per household. By comparison, the Fiber Broadband Association estimates that urban fiber-to-home construction in the United States costs the equivalent of about CA $1,400 to CA $1,800 per household—a stark contrast in expense.

Fort Albany resident Thomas Scott says high-speed internet has improved his work as a mental health counselor. Gavin John

A radio broadcast in 2018 heralded the good news: The fiber-to-home construction project was a go. Fort Albany mental health counselor Thomas Scott says he “couldn’t wait.” People seeking guidance for addiction or grief typically called him on landlines, and it was hard to help them over the phone without seeing their faces. Businesses, too, rejoiced—including the Kataquapit family store, which relied on the phone system for transactions.

Testing and Deploying a Remote Fiber Network

Winning approval for the project was one thing; building it across hundreds of kilometers of remote subarctic terrain was another.

Dirk MacLeod started with a schematic documenting the GPS position of every house, pole, and length of cable that would ultimately form the network. When WJBTN hired Montreal-based Fonex Data Systems to upgrade the backhaul, the detailed plan made it easy for contractor Tasso Varvarikos to design the deployment. Still, tuning the optical system to operate reliably across transmission lines stretching hundreds of kilometers required extra care. “Once you go up north, there’s no fiber store,” he says,

To minimize surprises in the field, Varvarikos traveled to telecom supplier Infinera’s laboratory in Stockholm, where he and other engineers assembled and tested the network before shipping it to the installation site. The Stockholm lab gave the team access to testing tools and technical specialists who helped configure the system before deployment in the remote fly-in communities. Using Infinera’s simulation software, Varvarikos says he and his colleagues “kicked the crap out of it in the lab” until the network performed reliably.

The Western James Bay Telecom Network connects remote communities along the western shore of James Bay in northern Ontario.Chris Philpot

Finally, in the spring of 2019, it was time to pack up and head to the sites. First, more than two pallets’ worth of Infinera equipment were squeezed onto two charter aircraft in Timmins. Bush pilots, unfazed by the stringent logistics, made sure that one box reached Moosonee; the other one, Attawapiskat. Varvarikos says they packed extra fiber-optic transceivers, patch cables, tools, and “the kitchen sink and then an extra sink just in case.” By May, the team was ready to install the new backhaul—a months-long process that necessarily preceded household connections. Once the equipment was in place, the engineers tested the network to make sure the systems in each community could communicate reliably and that data moved correctly across the backbone.

During the switchover to the new system, the MacLeod brothers worked from substations in different communities along the coast while Varvarikos coordinated from Fort Albany. Communicating over a spotty phone line, they started moving connections from the old network to the upgraded backhaul, carefully reconnecting cables and verifying that traffic still flowed correctly between communities. Within days, the anchor customers—including hospitals, schools, and air-navigation systems—were hardwired to a 100-Gb/s backbone.

However, households were still relying on the slow legacy network. WJBTN’s next step was flying huge reels of fiber-optic cable into the remote communities at enormous cost. Nakogee recalls cutting predetermined lengths of cable, then repackaging it onto spools to load onto cargo aircraft. The backbone was finally in place; now WJBTN had to bring fiber to every home.

Training a Local Fiber Crew

Then, in a turn of events that rocked the communities, lead engineer Dirk MacLeod had a heart attack and died in July 2019. “My chest was ripped open,” Nakogee says. The grieving company shut down for the summer.

WJBTN had intended to spend three summers training local crews from each community’s power company to maintain and manage the network. After MacLeod’s death, one line worker quietly took it upon himself to finish extending fiber to distribution points in Fort Albany. The team made a plan to begin connecting homes the following spring.

Utility poles carry power and fiber-optic lines through Fort Albany First Nation.Gavin John

Then COVID hit. Fearful and with limited medical facilities, the communities closed themselves off. Nobody was allowed in.

“COVID really screwed things up,” says Andrew MacLeod. People were screaming for better internet, he recalls, but they wouldn’t let outsiders in. So instead of training local crews and building each town’s network simultaneously as planned, WJBTN made a tantalizing offer: The first community to allow MacLeod in would have its fiber-to-the-home network completed first. Fort Albany jumped on it.

Peyton Reuben, a recent computer systems graduate in Fort Albany, was seeking a new job when his cousin mentioned that there was a “fiber guy” in town. Reuben’s coursework covered coding and how routers and ISPs communicate—a distant relative to Andrew MacLeod’s hands-on infrastructure work.

WJBTN network coordinator Peyton Reuben looks up at the overhead fiber network in Fort Albany [top]. A splice enclosure joins fiber-optic cables at a distribution point [bottom]. Gavin John

“They were looking for helpers to splice fiber,” says Reuben, something he knew nothing about. Nonetheless, he started work the day he met MacLeod and got a crash course in building an aerial fiber network, running cables from utility poles into neighborhood connection boxes that ultimately served individual homes. The fun part was reaching every pole in every neighborhood in a town crisscrossed by tributaries of the Albany River. “We didn’t have a bucket truck, so we had to climb up the poles,” says Reuben. “It was quite the experience, going through thick brush and thigh-deep water.”

The hard part was splicing the fiber—fusing together hair-thin glass strands that connected homes to the larger network. Inside each connection box, distribution fibers had to be joined to cables leading to individual houses, one strand at a time—around 150 in each box. Reuben’s first attempt “looked like a plate of spaghetti,” says MacLeod. More splicing waited back at the substation, where optical splitters connected neighborhood lines to the town’s central GPON equipment. “It took me four times as long as Andrew to complete a splice tray back then,” says Reuben.

A worker installs a new fiber-optic line for the Western James Bay Telecom Network.Gavin John

Slowly but surely, every building in Fort Albany saw a strand of fiber drop from overhead and come through a box drilled into the wall, ready to connect to a router. By springtime, aerial cables linked every house—and future building sites—to the substation. Reuben calls it “a spider web that goes everywhere across town.”

Without much fanfare, on a night in April 2022, WJBTN flipped the switch in Fort Albany. The next morning, Kataquapit woke up to a different world. With 250 Mb/s download and 30 Mb/s upload speeds, she found herself just a click away from her children in faraway cities.

Why Not Starlink?

MacLeod and Reuben continued working up the coast, splicing cables and adding connection boxes. As COVID eased, they hired more local help, and the power company lent a hand with bucket trucks. “We were working 12 hours a day, 7 days a week,” says MacLeod. “Guys were sitting in the heat, in bucket trucks, learning on the fly how to do splicing.” By late 2022, the network reached every home.

Fiber-optic cables and networking equipment inside the Fort Albany substation distribute internet service throughout the community [top, middle]. Large spools of fiber-optic cable were flown in to connect homes [bottom]. Gavin John

Around that time, Starlink’s satellite internet service was becoming widely available. Many of WJBTN’s potential clients asked why they shouldn’t just purchase that instead of a fiber-optic connection that involved drilling into their homes. But speed tests comparing Xittel, Starlink, and the new WJBTN connections showed a major difference in latency: In applications like video calls, Xittel and Starlink’s round-trip signal delays became painfully obvious. Unlike satellite systems, fiber networks don’t need to send signals hundreds of kilometers into orbit and back. That gave WJBTN a significant performance advantage.

MacLeod recalls a conversation with one line worker who wanted to turn to Starlink. “We told him our latency is so much better,” MacLeod says, explaining that ISPs measure performance by both bandwidth and latency. From Attawapiskat, a data packet traveling over WJBTN’s fiber network reached Toronto in 20 milliseconds. Comparable satellite connections took closer to 60 milliseconds. And Xittel’s latency was much worse, at several hundred milliseconds. (Since then, WJBTN has reduced latency further, to about 12 milliseconds.)

The team’s success has bolstered other Indigenous broadband companies. WJBTN representatives have shared their experiences at Indigenous Connectivity Summits since 2017, and how-to workshops have sprung up across Canada (and the United States). And Canada has since created a dedicated broadband funding stream for Indigenous communities. Within Fonex Data Systems, the company hired to do the backbone work, the WJBTN implementation is now considered the model for a successful installation in a remote location.

Nakogee says ownership remains the key advantage. Unlike earlier telecom providers that leased infrastructure or delivered service wirelessly, WJBTN owns the backbone fiber, the right-of-way, and the poles carrying the network. That makes it far harder for outside telecom companies to displace the service. “That’s our secret weapon,” he says.

Fundamentally, it’s about sovereignty. By controlling the infrastructure that carries internet traffic, the communities can govern and maintain the network according to their own priorities rather than the financial goals of distant providers.

How Fiber Changed Daily Life

Shortly after the fiber network was installed, Elizabeth Kataquapit became chief of Fort Albany First Nation. During her term as chief, she kicked the community’s digital era into full gear, encouraging individuals who couldn’t attend community meetings in person to join over Zoom.

Counselor Scott now conducts grief and addiction sessions both in person and via video calls and can connect immediately in a crisis even if he’s away. And as a search-and-rescue volunteer, he responded to the Facebook message when the young man got lost in a winter snowstorm on James Bay. “If we were a half hour later, the person would have [been] gone,” says Scott.

Broadband has also changed everyday life in quieter ways. Residents run small businesses from their homes. Telehealth links patients to specialists in southern cities. More people are working remotely and taking classes online. “Quality of life is so much better,” Reuben says.

But the network has also brought to these northern communities the more isolating side of fast internet. Community members stream more movies and play more video games, and they meet face-to-face less frequently. Standing beside an enormous canoe in his front yard, Scott explains that while he’s grateful for Fort Albany’s new connection to the outside world, he’s equally intent on preserving close connections within the community. Today, he says, as he starts loading fishing nets into the canoe, “I’m taking the kids out after school to harvest whitefish.”

This article appears in the August 2026 print issue as “Wiring the North.”

In 2005, Nokia sold its billionth mobile phone, a budget-friendly device that went to a customer in Nigeria. By then, the company, based in Espoo, Finland, was making one of every three cellphones globally.

But just nine years later, the mobile-device maker offloaded its entire handset division to Microsoft for pennies on the dollar, compared to what it had been worth at its peak.

Nokia had risen from obscurity in the 1990s to become a worldwide cultural phenomenon by the turn of the millennium, its signature devices featured in TV shows and movies, announcing their presence with instantly recognizable Nokia ringtones.

As Nokia was becoming comfortable in the spotlight, the smartphone era arrived. And what came next was swift and brutal. But, as revealed in Nokia internal documents recently made public and interviews with key Nokia engineers from that era, the company saw it coming. Within 24 hours of Apple CEO Steve Jobs’s iPhone unveiling in 2007, Nokia was already weighing its options. They’d immediately recognized the threat. However, outrunning it was another matter.

What follows is Nokia’s story over 14 years, from 1998 to 2012, as the world’s top cellphone maker—how its devices defined their time, how the tech reshaped what phones could be and do, and how the company’s good fortunes in the handset business came to an end.

Nokia Was Once Unbeatable

The centerpiece Nokia devices, the ones that people probably think of when they see the words “Nokia phone,” were the 3210 and its cousin, the 3310. TechRadar has called the 3310 “the greatest phone of all time.”

Nokia’s 3210 phone, released in 1999, was an inexpensive device aimed at younger users. Colin McPherson/Alamy

Released in 1999 and 2000, respectively, the two devices sold more than 280 million units worldwide. Their most innovative hardware feature was the internal antenna—the first mass-market phone without even a stub or retractable aerial. “Consumers had the perception that it could not work well without an external antenna,” said Peter Røpke, a former Nokia senior vice president, in a 2016 interview with Slate.

The phones shipped with games, including the legendary Snake, one of the most popular pre-smartphone mobile games—in which a pixelated serpent eats and grows with every morsel consumed.

Nokia introduced no small portion of the world to texting. At the time of the 3210 and 3310, the prevailing texting standard was SMS (short message service), which allowed up to 160 characters per message. Nokia appended its own Nokia smart-messaging service to SMS, which allowed the sending of small bitmapped images across an otherwise text-only system. A rich-text messaging system that allowed visual images, audio, and video followed in 2002, leading to a multimedia messaging service (MMS) standard that remains in place today.

Nokia also enabled users to easily create and share ringtones on their devices. By 2000, Nokia’s custom-ringtone Composer app had popularized a new, short-form musical medium that the ringtone industry, at its peak, would transform into a billion-dollar marketplace in the United States.

Nokia introduced its 1100 phone in 2003 and ultimately sold half a billion units, making it the most popular cellphone in history. Paul Chesne/Donaldson Collection/Getty Images

A few years later, Nokia reimagined its mobile handsets, releasing the 1100 in 2003. The 1100 sold a half a billion units, more than any cellphone in history. It remains one of the best-selling consumer products ever. Much of the 1100’s success was due to its price tag—in the neighborhood of US $100, making it at the time Nokia’s most affordable device.

Also contributing to the 1100’s popularity were features designed for longevity and tough environments, including dust resistance, nonslip sides for better handling in rainy conditions, and a 400-hour standby battery life. The 1100 introduced a flashlight as well, which the user turned on and off by holding down the “C” key.

Where most device makers at the time were worried about camera megapixels and color screens, Nokia had leapfrogged its competition with a back-to-basics phone that could survive the rain, endure unreliable power grids, and light the way home.

Apple Launched the iPhone, Nokia Scrambled

On 9 January 2007, at the Macworld conference in San Francisco, Steve Jobs made a characteristically bold claim. “Today, Apple is reinventing the phone,” he said, soon pulling one of the first iPhones out of his pocket.

Apple CEO Steve Jobs famously launched the iPhone at the Macworld Conference in San Francisco on 9 January 2007. Nokia held a rapid-response meeting to the event the following day. Tony Avelar/AFP/Getty Images

Rumors of Apple entering the phone market had swirled since the iPod’s debut in 2001, but nobody had really reckoned with what that might mean.

“Executive summary: Apple iPhone is a serious high-end contender,” read a slide from a Nokia internal meeting held the day after Jobs’s keynote. (That slide is now in the company’s online archives, opened to the public last year.)

“User interface has been a big strength for Nokia,” it continued. “Nokia needs to develop touch [user interface] to fight back.”

Peter Bryer, at the time Nokia’s manager of strategic foresight, was part of that 10 January meeting, and he recalls that Jobs’s announcement wasn’t unexpected. But the iPhone’s extensive reliance on multitouch—save for a single home button on the front—did surprise the team.

Nokia was already aware of multitouch technology, Bryer notes. In 2006, the U.S. computer scientist Jeff Han had given a celebrated TED talk about it, demonstrating a multitouch screen, which could sense multiple fingers on the screen at a time, not just one. Bryer remembers his colleague Timo Partanen, then Nokia’s director of market and competitor analysis, getting excited about Han’s demo.

In 2006, the NYU research scientist Jeff Han showed off a new multitouch interface technology as part of a popular TED talk. By the end of the decade, multitouch—in which multiple fingers can interact with a touchscreen at once—would play a key role in smartphones from Apple, HTC, and Palm. Steve Jurvetson/Flickr

“Timo burst into the room, saying, ‘You’ve got to see this TED video of this guy using multitouch,’” Bryer recalls. “We both thought that was cool and that’s the future. Then I looked at the sponsors of the presenter’s research, and among them were Nokia and Microsoft.”

And yet it took Nokia years to develop a phone that used multitouch. “Remember, Nokia is based in Finland,” he says. “It’s very cold in Finland. They wear gloves for six months of the year, including the executives. They didn’t think a device like that would work.”

Winter gloves were no obstacle to operating the chunky buttons on Nokia phones, a design priority perhaps stemming from the company’s Finnish culture and headquarters. Erol Gurian/laif/Redux

Partanen was also at Nokia’s post-iPhone launch meeting, and recalls that there was little concern in the room. “We felt okay,” he says. “This is yet another competitor launching a great product. But we had no doubt that, if it’s successful, we would do the same. We will launch similar products.”

In November 2008, Nokia released the 5800 XpressMusic, a year and a half after Apple had launched its iPhone. Shaun Curry/AFP/Getty Images

That similar product ended up being the Nokia 5800 XpressMusic, known as the Tube, released in 2008. “The idea was to focus on streaming videos and television,” Partanen says. “So we made a phone with a similar form factor to the iPhone [that was] optimized for streaming content.” But the 5800 was “delayed, delayed, delayed, delayed,” he says. “It didn’t materialize in the way it was planned. It was released as a watered-down version.”

Critics skewered the 5800’s “outdated” feature set and “ancient” S60 operating system, which ran on top of Symbian OS, an open-source mobile platform Nokia had recently acquired. The 5800 sold reasonably well for its time, reaching around 8 million units in its first year alone. But it did not feature multitouch.

“I think that started to be the point when everybody realized that, hey, this is by far more difficult than earlier competitive issues we’ve had,” Partanen says.

Nokia finally released its first device with multitouch in 2010, three years after Jobs’s splashy iPhone announcement and four years after Han’s TED talk demo.

How Android Ate Up the Low-End Market

Nokia had long owned the low end of the cellphone market, with its sturdy, no-frills devices suited for that segment. So the years immediately following the iPhone’s launch saw the Finnish firm continue to thrive as it kept turning out simple, rugged devices.

As one review of the Nokia 1200—successor to the 1100—put it in October 2007, “This handset chucks away all the fancy features you’ve come to expect on a modern mobile, leaving you with a pared-down feature set that’s easy for tech novices to get their heads around.”

Two cellphone users in Nairobi, Kenya in 2013 exchange a payment on a Nokia 1200 phone via the M-Pesa Mobile Money Market, a popular online banking service. Trevor Snapp/Bloomberg/Getty Images

The 1200 kept the 1100’s dust-proofing, flashlight, and long-lasting battery, and added features aimed squarely at the developing world. The 1200 was the first to include call-time tracking and a multiuser phone book, allowing owners who planned to lend their device to set up call limits based on time or cost. This feature helped enable what Nokia researchers called kiosks—informal pay-per-call services, in which an enterprising phone subscriber charged neighbors and family members by the minute for use of the device.

In 2006, Nokia studied how Ugandans used their Nokia phones in rural and remote areas. An internal company slide deck from the time reveals just how keyed-in Nokia was to its lowest-income users. “Village phone operators are often women,” the slide deck notes. “And there tend to be a lot of children around. (Phones need to suffer considerable abuse from chewing, dust, sweat, etc.)”

“A unit of phone time is 60 seconds,” another slide states. “But to avoid accidentally going over that time and incurring extra costs, kiosk operators shorten the unit to 57 seconds, allowing a three-second margin of error. Shared mobile used as phone kiosk must show call time.”

Nokia’s familiarity with its market couldn’t protect the company forever, though.

Nokia sought out user input around the world for the company’s device designs, including hosting “Open Studio” contests soliciting users’ sketches of their dream cellphone. Shaul Schwarz/Getty Images

That’s because the iPhone wasn’t Nokia’s only looming smartphone competitor. In September 2008, the first Android phone went on sale—the HTC Dream, which was also sold as the T-Mobile G1.

While the iPhone was aimed mostly at early adopters and affluent users who could afford to drop hundreds of dollars on a new phone, Android phones were, within a couple of years, aiming at the same low-cost, global user base Nokia was selling to.

“I think it’s fair to say Android is the one that disrupted the market more for Nokia,” Bryer says. “Most of Nokia’s successful devices were not on the high-end market. But then, when Android came along, it started to fill that lower end and eventually took that market away from us.”

An executive from Nokia India in 2010 holds the company’s 5530 XpressMusic and 5230 phones, both of which had touchscreens, although only the 5530 had Wi-Fi. Sam Panthaky/AFP/Getty Images

With two emerging competitors in the low end and high end, the Finnish device maker responded with a device that split the difference—and satisfied neither camp.

Released in 2009, the Nokia 5230 attempted to be a low-priced, touchscreen (though not multitouch) competitor to both the iPhone and Android. It sold an impressive 150 million units, doing especially well in developing countries.

But the 5230 didn’t have Wi-Fi—one of the biggest complaints at the time. In the developing world, Wi-Fi connections were still rare, so the lack of Wi-Fi made some sense. But the rest of the world was not pleased.

“We had such a big gap and dominant position,” Bryer says. “Which does maybe create a level of comfort which you should never get.”

How Nokia Lost the Smartphone Race

By the beginning of the 2010s, Nokia could have still drawn from the company’s labs, which were regularly spinning out new technologies and innovations. However, the Finnish handset maker ultimately failed to turn its R&D into viable new product lines in response to the emerging smartphone threat.

Nokia’s predicament had precedent—Kodak, dominant in film photography, had actually invented the digital camera in 1975 but failed to commercialize it before digital imaging made its core business obsolete.

“The technology coming from our R&D teams was cutting edge,” says Gordon Murray-Smith, director of services and ecosystems intelligence from 2008 to 2011. He recalls attending annual R&D innovation days that showcased work on self-healing materials and flexible screens, long before those technologies were seen elsewhere. “But why was Nokia not able to commercialize some of that really interesting and innovative activity more than it did?”

Nokia desperately needed an injection of life to change its fortunes. The company’s first non-Finnish CEO, Stephen Elop (a Canadian fresh off a two-year stint on Microsoft’s leadership team), did not mince words.

In an internal memo from February 2011 that was soon leaked to the media, Elop wrote, “The first iPhone shipped in 2007, and we still don’t have a product that is close to their experience. Android came on the scene just over two years ago, and this week they took our leadership position in smartphone volumes. Unbelievable.”

In 2011, Nokia released the N9, a smartphone with a Linux-derived operating system. Within a year, Nokia had pivoted toward its Windows Phone-powered line of Lumia devices. Munshi Ahmed/Bloomberg/Getty Images

Elop oversaw the 2011 launch of a Linux-based smartphone, the Nokia N9. The N9 ran on a distribution of Linux called MeeGo. Reviewers at the time praised the new smartphone direction the Finnish phone maker had taken. “Possibly the most beautiful phone ever made,” wrote one reviewer about the N9 for Engadget.

But the N9’s accolades did not ultimately carry the day. Nokia announced its Lumia line of phones the same year—a direct pivot away from MeeGo toward the Windows Phone. It would be the last major strategic turn Nokia would take as a cellphone manufacturer. From this point forward, a succession of C-suite decisions all but sealed the fate of Nokia’s iconic line of phones.

In 2013, Microsoft announced its bid to acquire Nokia’s handset operations. After the sale went through the following year, it rebranded the division Microsoft Mobile. But the year after that, Microsoft decided it had made a costly mistake, writing down $7.6 billion—nearly what it paid for Nokia’s handset division—and laying off nearly half of the former Nokia staff it had inherited.

In 2016, Microsoft sold its feature phone assets to HMD Global. The latter still sells Nokia-branded phones—budget-friendly devices as well as nostalgia reproductions of models from Nokia’s glory days. What remained was a brand name, some intellectual property, and two decades of hard-won lessons about what it takes to stay on top—and what it costs when you can’t.

“When you look at the players in the world of smartphones today, any of those players would struggle ever to achieve 14 consecutive years of being No. 1,” says Murray-Smith.

Partanen says there was a downside to Nokia’s mobile-phone dominance. “Often, being the first mover is not necessarily the best position,” he says. “Being a quick follower is the best position.”

The company itself ultimately survived, even if the transition wasn’t painless. Nokia’s revenues, which peaked in 2007, fell sharply through the mid-2010s before the company refocused on a decades-old business line—telecom infrastructure—that many had forgotten Nokia was even in. Nokia now ranks among the world’s top three suppliers of 5G network equipment, serving carriers across more than 125 countries, alongside Ericsson and Huawei. Although the company could never quite crack the smartphone, it now plays a key role in providing the network backbone those smartphones run on.

This article appears in the August 2026 print issue as “How the iPhone and Android Killed the Feature Phone.”

Modern battlefields are bathed in radio waves for communications, radar, remote control of drones, and signal jamming to disrupt those functions. The ability to rapidly locate their source could provide considerable tactical advantage, and a new quantum sensor the size of a paper clip promises to work as an “electromagnetic compass” to determine the 3D direction of a wide-range of radio signals.

Pinpointing where a signal comes from normally requires multiple, bulky antennas to triangulate the direction. Each setup is also tuned for a specific frequency, so getting a full picture of the local radio environment requires a considerable amount of hardware. That’s a major limitation on a battlefield, where being large and conspicuous is a significant risk.

Researchers at the U.S. Army Combat Capabilities Development Command Army Research Laboratory (DEVCOM ARL) have developed a novel quantum sensor that shrinks these capabilities into a single compact device. It relies on a 2.5-centimeter glass chamber filled with excited atoms that can detect a broad range of frequencies and determine the direction of incoming radio waves to within two degrees. The researchers are now working with quantum technology company Infleqtion, in Louisville, Colo., to create a working prototype for use in the field.

“You can imagine having a device that provides you a level of spectrum awareness that’s very difficult to get in a single platform,” says ARL research physicist David Meyer, who is leading the research. “The goal is to get more information, and get that information in the hands of soldiers.”

How quantum sensors detect radio waves

The sensor, described in Physical Review Applied, relies on Rydberg atoms. These are atoms that have had one of their electrons excited to an extremely high energy state, normally by firing a specially tuned laser at them. This forces the negatively-charged electron into a wide orbit of the nucleus, and because it sits so far from the atom’s positively charged core, it is easily perturbed by incoming electric fields, Meyer says. This causes shifts in the atom’s energy levels that can be detected by monitoring the power of a laser shining into the chamber.

Using these atoms to figure out the direction the field is coming from is a little more complicated. Normally, this is done by spacing at least two sensors a significant distance apart and comparing what they pick up, or using an antenna that physically scans around to locate the source. Think of a radar dish in an old World War II movie, Meyer says.

Instead, the new sensor measures the incoming signal’s polarization—essentially the direction in which its waves oscillate. Many radio signals don’t wiggle up and down neatly on a single plane and instead propagate in a corkscrew-like pattern. This traces either a circle or an ellipse, which sits perpendicular to the direction the wave is traveling. This means figuring out the wave’s polarization can tell you where its source is.

To measure this, the researchers beamed three “reference” radio signals into the device that oscillate along the X, Y, and Z axes. When an external signal hits the device, it boosts the electric field along each axis by a different amount depending on where it’s coming from. This causes an interaction between the reference signals and the incoming signals that is picked up by the atoms and can be used to map out the shape and orientation of the signal’s polarization—and in turn work out where it is coming from.

Instead of using multiple sensors spaced apart to detect where a radio signal is coming from, this device uses a single atom to determine its polarization. DEVCOM Army Research Laboratory

Shrinking the sensor

This ability to measure the direction of a signal using a single, static, and tiny sensor, combined with the broad frequency range of Rydberg atoms, presents a significant advance over existing approaches, Meyer says. Conventional antennae have to be a similar size to the wavelengths they monitor, making them bulky.

You also need several to cover the entire spectrum, but placing them too close together interferes with the signals, because any metal object within an antenna’s wavelength essentially becomes part of it, says Meyer. “So if you want to measure lots of different bands, you need lots of different antennas,” he says. “But if you don’t want to separate them by huge amounts, the colocation problem is pretty significant.”

One limitation of the Army researchers’ device is that it only works with circular or elliptical polarization. However, Meyer notes that many signals of interest are already polarized this way, and in the wild, linear signals often become slightly elliptical through interaction with the atmosphere. For smaller wavelength signals in the microwave and terahertz range, researchers can also 3D-print a precisely patterned plastic structure known as a wave plate that can polarize incoming signals to make them elliptical, he adds.

A bigger challenge is getting the sensor out of the lab. Experiments so far have run on an optical table in a controlled environment, and the lasers that probe the Rydberg atoms are especially finicky. “The stability required is very high,” Meyer says. “And lasers that are that stable don’t generally like being outside in high humidity and high heat.”

Making a field-ready device with Infleqtion

This is why ARL announced in May a partnership with Infleqtion to develop a field-ready version. Solving the laser stability problem is the biggest challenge, says Seth Caliga, the company’s director of R&D for quantum RF sensing. The gold standard involves optical cavities (arrangements of mirrors often used to focus beams of light) housed in thermally controlled vacuum chambers to protect the lasers from environmental noise, he says, which is clearly impractical in the field.

Instead, Infleqtion plans to piggyback on their optical atomic clock technology, which relies on technology similar to the ARL sensor. In optical atomic clocks, the oscillations of a light wave are used for timekeeping, comparable to the swinging pendulum in a grandfather clock. To keep these oscillations stable, they are locked to the precise energy levels in a rubidium atom. (This is possible because electrons only transition between levels when hit with light at a very specific frequency.) The clock is tuned by firing a laser at a cloud of rubidium atoms and adjusting its frequency until the maximum number of atoms jumps an energy level, which signals that the laser’s frequency is locked onto the atom’s transition frequency.

Caliga says the same ultra-stable frequency can act as a kind of “ruler” to stabilize the lasers probing a Rydberg sensor. And because all rubidium atoms are identical, environmental noise affects the clock and the sensor the same way, unlike a classical optical cavity. “It’s not a game of telephone, where there’s frequency errors that go down the chain,” he says. “They’re going to drift together.”

Dealing with that drift remains a significant challenge. By its very nature, a sensor must be exposed to the environment to detect signals, Caliga says. But the highly sensitive atoms can be easily disturbed by heat, humidity, or vibration. This means the device must be continuously recalibrated, a complex optimization problem Caliga calls “still quite nascent.”

There is also the more prosaic work of miniaturizing the optical and electronic components so it can be packaged more portably, says Meyer. “A lot of this stuff can be extremely shrunk down into absolutely tiny package sizes, so there’s certainly a path there,” he says. “It’s just a very long path with a lot of dollar signs associated with it.”

An examination of how satellite vulnerabilities, modern wideband waveforms, and automatic link establishment are driving renewed military and government investment in HF communications.

What Attendees will Learn

1. Why HF (High Frequency) declined — and what has changed — How satellites overtook HF for global communications from the 1970s onward, and why growing awareness of satellite vulnerabilities to anti-satellite weapons, jamming, solar storms, and coverage gaps is reviving interest in skywave propagation as a resilient alternative.
2. How the ionosphere enables and limits global HF communication — Understand the roles of the D, E, and F ionospheric layers in refracting and absorbing signals, the concepts of maximum usable frequency (MUF) and lowest usable frequency (LUF), and how sunspot number, solar flux index, and A/K geomagnetic indices are used to quantify and predict propagation conditions.
3. How automatic link establishment transforms HF operability — Trace the evolution from proprietary first-generation ALE through interoperable second- and third-generation standards to fourth-generation wideband ALE, which automates frequency selection, link setup, and adaptation to changing channel conditions — removing the dependency on highly skilled operators.
4. How wideband HF is closing the throughput.

A practical educational guide to common and uncommon VHF propagation modes, covering the physics, range implications, and real-world behaviors engineers need to understand.

What Attendees will Learn

1. Why “line of sight” fails as a practical VHF planning model.

2. How refraction, reflection, diffraction, and scattering deliver or destroy signals where geometry alone cannot predict.3. How tropospheric refraction extends the VHF radio horizon roughly one-third beyond optical line of sight.

4. How temperature inversions form ducts that can carry VHF signals over 1,500 km.5. How sporadic E, meteor burst, and EME propagate VHF signals across hundreds to thousands of kilometers.

6. What frequency limits, distance ranges, and environmental triggers apply to each propagation mode.

7. How to apply this knowledge to link budgeting, interference prediction, and contingency planning.

Download this free whitepaper now!

Global demand for high-capacity, low-latency wireless connectivity continues to grow, and the 6-gigahertz band of radio spectrum has emerged as a particularly vital chunk of wireless real estate.

Both the cellular and Wi-Fi industries consider this contiguous block, which covers frequencies from 5,925 to 7,125 megahertz, critical. The cellular industry believes the spectrum can provide anchor capacity for urban 6G cellular networks, while Wi-Fi advocates want it for low-latency, multigigabit-per-second streaming, augmented reality, and virtual reality applications over Wi-Fi 7 and Wi-Fi 8 devices.

Unlicensed Wi-Fi advocates, backed by companies like Amazon, Apple, Google, and Microsoft, want the spectrum made available as public frequencies that are open to anyone. They are vying with licensed cellular providers who believe the airwaves should be exclusive and purchased from regulatory bodies to guarantee coverage and provide seamless mobility.

Most major global economies have decided to split 6 GHz between unlicensed Wi-Fi access on the lower 500-MHz portion of the band (spanning from 5,925 to 6,425 MHz) and licensed cellular use on the larger 700-MHz upper portion.

The battle lines have now shifted to the European Union, India, and the United Kingdom, as well as other fence sitters that have yet to decide how the critical upper 6-GHz airwaves will be assigned, with Wi-Fi and cellular camps fiercely lobbying their cases with regulators.

6-GHz Spectrum and Economic Growth

“Any country that chooses not to allocate the entire 6-GHz band to unlicensed use is putting itself at significant innovation disadvantage and limiting its capacity for economic growth,” says Mary Brown, executive director of Wifi Forward, a Wi-Fi advocacy group whose members include Broadcom, Comcast, Google, and Microsoft.

United States

The U.S. has assigned the full 6-GHz spectrum band to unlicensed wireless use, such as Wi-Fi 6E, 7, and 8. Mid-band frequencies such as the 7-GHz band and the 2.7-to-2.9-GHz band are being considered for potential 6G cellular use. Canada, Saudi Arabia, and South Korea have taken a similar license-exempt approach to 6 GHz.

Wifi Forward published a report in 2025 revealing that, in the United States, restricting Wi-Fi to the lower 500 MHz of the 6-GHz band could cost over US $1.74 trillion in lost economic value over the following two years. [Editor’s note: The IEEE creates and maintains the set of technical standards behind Wi-Fi.]

“Wi-Fi is already carrying as much as 90 percent of internet traffic in developed economies and that will only continue to increase with AI poised to become the most significant driver of network demand since the emergence of the smartphone,” says Brown.

On the flip side, the GSM Association (GSMA), a trade association representing over 800 mobile-network operators worldwide, says it wants unfettered access to the upper 6-GHz band to avoid fragmentation of the global 6G ecosystem and to maintain mobile’s impact on global GDP growth, which it has forecast at 8.4 percent by 2030.

Wi-Fi has historically lived almost exclusively on the 2.4-GHz and 5-GHz bands. The recent move into the lower 6-GHz band in certain countries has, according to Ross Bateson, a senior spectrum advisor at the GSMA, resulted in very low adoption. This makes Bateson question the need for additional access to the upper 6-GHz band.

“In most cities around the world, there is between 0 percent and 5 percent Wi-Fi use in the lower 6-GHz band,” says Bateson, “For Wi-Fi boxes to have this added complexity without any guarantee that they’ll be able to use the spectrum they’re built for is a little bit confusing to us.”

How Different Countries Are Using 6-GHz Spectrum

India is currently in a regulatory holding position on the upper 6-GHz spectrum, leaving questions around future arrangements to keep the band exclusively for mobile use or to develop strategies for coexistence.

India

India has formally designated the lower portion of the 6-GHz band for unlicensed wireless use, and the 6,425-to-7,125-MHz range for licensed cellular. However, the Telecom Regulatory Authority of India (TRAI) stepped in to recommend delaying license auctions for the upper band until after the World Radiocommunication Conference in 2027 (WRC-27). Tech organizations continue to lobby against granting the upper band entirely to the mobile industry.

Elsewhere, the European Commission appears keen to avoid a winner-takes-all outcome. Its Radio Spectrum Policy Group has recommended that 540 MHz of the upper band is designated for priority mobile use with decisions on the additional 160 MHz (or potentially 125 MHz) delayed until the ITU World Radiocommunication Conference 2027 (WRC-27).

The prospect of a split upper band in EU member states raises technical challenges around coexistence and potential interference. The GSMA’s head of spectrum, Luciana Camargos, has said the approach will not work because mobile and Wi-Fi networks use fundamentally different standards.

“Where licensed mobile and unlicensed Wi-Fi sit adjacent in the same band, as Europe is now engineering, you need guard bands, strict out-of-band emission limits, and power restrictions,” says Adlane Fellah, founder and chief analyst at Florida-based wireless consultancy Maravedis. “The size of Europe’s guard band is itself contingent on what is decided at WRC-27 regarding adjacent 7-GHz spectrum, which tells you how delicate the engineering is.”

The European Union

The EU is taking a hybrid approach to the 6-GHz band. The lower portion of the band (5,925 to 6,425 MHz) is designated for Wi-Fi and the upper band is proposed to be split, with 540-MHz handed over to licensed mobile and a decision on the remaining 160-MHz frozen until WRC-27.

The European Conference of Postal and Telecommunications Administrations, the EU’s spectrum coordination body, is examining the feasibility of a hybrid/prioritized sharing model, which would allow the nonpriority technology to access priority airwaves, provided it can coexist without causing harmful interference.

The stakes here are high, says Fellah. “Real-time licensed/unlicensed sharing at scale has never been proven in a band this commercially important.”

EU decision makers will be closely watching developments in the U.K., where the Office of Communications, the government regulator for communications services, is exploring a proposal for a dynamic spectrum-sharing framework for the upper 6-GHz band.

Through a phased approach, this would see Wi-Fi access points gain primary access to the bottom 160-MHz chunk of the band, as well as early access to the upper 540-MHz “mobile priority” chunk, with the caveat of that portion being under automated frequency coordination control.

AFC dynamically checks the airwaves and then either allocates or denies Wi-Fi access points connection to specific frequencies, depending on the current level of congestion. The solution could help maintain the flow of mobile traffic in high-density areas, particularly when 6G networks start being deployed around the end of the decade.

Canada and the U.S. already operate similar AFC systems, potentially allowing the same Wi-Fi hardware to operate in multiple regions without separate configurations.

Global 6-GHz Spectrum Policy Divide

The world is essentially split three ways on 6 GHz: the United States is dedicating the entire 1,200-MHz band to unlicensed use, China is dedicating it entirely to mobile, and Europe is taking a middle path. That has raised concerns that the next era of wireless communications—headlined by 6G and Wi-Fi 8—will result in the world’s three largest economic blocs adopting three incompatible wireless spectrum models.

China

China is the only large nation to reject Wi-Fi operations in the 6-GHz band. It was also the first to authorize the entire upper part of the band (6,425 to 7,125 MHz) for licensed cellular, demonstrating a commitment to 5G and, in the future, 6G.

This scenario bolsters a traditional geographical split between device original equipment manufacturers (OEMs) with, for example, China’s Huawei and ZTE focused on 5G-Advanced and 6G, and the U.S. anchored in domestic silicon and Wi-Fi infrastructure players like Broadcom, Cisco, and Qualcomm. Ericsson and Nokia lobbied alongside the GSMA for licensed spectrum in Europe, where they have the most customers.

Uneven 6-GHz policies can make Wi-Fi and smartphone devices more difficult and more expensive to design and develop, says Fellah, and Wi-Fi vendors are already being forced to ship multiple hardware variants, raising manufacturing costs.

“Some suppliers are releasing Wi-Fi 7 access points that either use software-defined radios to toggle between 5 GHz and 6 GHz or omit 6 GHz entirely to keep costs down in markets where the band isn’t available,” Fellah says.

6G smartphones are at an early hardware-prototyping phase ahead of an expected launch in 2029 or 2030. According to John Kuzin, the senior vice president of spectrum policy and regulatory counsel at Qualcomm, the company’s chipsets for 6G handsets are likely to remain uniform across markets.

“There aren’t major challenges to, for example, supporting unlicensed Wi-Fi in the full 1,200 MHz in the 6-GHz band, as permitted in the U.S. and Canada, and supporting in other regions of the world unlicensed Wi-Fi in the lower 500-MHz portion of the 6-GHz band and licensed mobile services in the upper 6 GHz,” says Kuzin.

Brazil, Mexico, Japan, and the United Arab Emirates

These countries are among those that have agreed on a direct two-way split between unlicensed Wi-Fi in the lower band and licensed cellular in the upper band.

6G base stations are expected to require larger antenna arrays to counter high atmospheric propagation loss and to match the coverage footprint of older spectrum bands.

“For all global frequency variants, we look at how many antenna elements it makes sense to install on a reasonably sized radio unit that doesn’t weigh too much,” says Per Beming, the head of standards and industry initiatives at Ericsson. Beming adds that base-station filter characteristics and power amplifiers must be tailored to suit the specific band, to prevent cochannel interference, and to protect other services or spectrum outside the band.

Ericsson’s suppliers and sub-suppliers are already working on solutions for markets including China and India, with any compromise “always about judging what the specific market wants and needs rather than any technical difficulties,” says Beming.

As manufacturers and network operators work up solutions for the 6-GHz band, the challenge will be in charting a path that doesn’t undermine the economies of scale needed for wireless innovation, lower hardware costs, and ultimately technology adoption.

This story was updated on 26 June, 2026 to correct a detail about Ericsson’s suppliers and sub-suppliers. They are working on solutions for China and India markets, not based in those countries, as originally stated.

This article appears in the September 2026 print issue.

Take a moment and try to imagine your life without the wireless advances of the past three decades.

Have you lost your luggage? What a shame AirTags have not been invented. The airline representative has promised to call with updates, so settle in for a long wait by the kitchen telephone, because there are no affordable cellphones. You’ll be stuck listening to whatever is on the radio while you wait, because there are no streaming services. That’s not even to speak of all the movie plots that would have been ruined.

This is just a tiny sliver of how wireless technology makes itself felt in your day-to-day existence. The effects it has had on supply chains, infrastructure, and how the economy runs have been world-altering.

None of it would be possible without the radio-frequency integrated circuits that allow all our devices to unobtrusively send and receive information.

Now imagine what the further evolution of this technology will bring: Wide-spread autonomous vehicles, quantum communications, 6G mobile service and satellite communications. Continued momentum will depend on newer and more advanced versions of today’s RF chips.

But there’s the rub. Whereas the design of most of the world’s computing chips has been standardized into its own science, RF design has remained stubbornly in the realm of art. A dark art, even, that is mastered only through years of experience. As any sorcerer will tell you, the dark arts keep their own schedule. And that schedule is impeding progress not just in RF chip design but in every other technology that depends on it.

About seven years ago, in the wake of AlphaGo’s victory over world Go champion Lee Sedol, my students at Princeton and I began to wonder: Could AI be taught this art as well? Recent successes suggest that, to a large extent, it can. Over the last few years, our group and other leaders in the field have started to develop machine-learning-driven algorithmic methods for designing RFICs. Some of the resulting chips look more like modern art than circuit layouts. Yet in many cases, the physical prototypes bested state-of-the art circuits in terms of performance. The real achievement, however, is that it took the AI orders of magnitude less time to conceive a working design than it would a human designer.

This is not about one or two RF chips. AI-enabled design could be the future of all RF design, and maybe much more.

The Dark Art of RFIC Design

So why do these chips all have to be crafted by hand? Why aren’t RFICs designed with an algorithmic synthesis process, much as CPUs and GPUs are?

The design of RFICs is an exercise in engineering across multiple physical domains. Maxwell’s equations, operating across different spatial and temporal scales, govern how electromagnetic fields interact with active and passive devices that must be carefully codesigned for the chip to function. Alongside these are the laws of thermodynamics, which determine how heat is generated and removed during operation, as well as the mechanics of thermal expansion and contraction that dictate how reliably the chip and its packaging survive temperature changes.

Simultaneously accounting for all the physical constraints these impose makes the design space almost impossibly large. Every decision involves complex priorities that often compete with one another, preventing the optimization of any of them.

To better understand the issue, let’s walk through the steps involved, after which you’ll better understand why a single new chip design takes years and tens to hundreds of millions of dollars.

Most of the area of radio-frequency integrated circuits is dominated by complex electromagnetic structures. Human-designed RFICs, like this broadband power amplifier [1], start with templates and follow a symmetric, understandable pattern. But freed from the constraints of human-designed templates and the need for humans to even understand the rationale of electromagnetic structures, power amplifier ICs [2–5] and low-noise amplifiers [6] can take on truly wild-looking yet efficient designs. SENGUPTA LAB

Let’s say you’re an engineer assigned to design a new 28-gigahertz power amplifier for a 5G-millimeter-wave handset. (This is the type of RFIC that boosts the 5G signals on your phone and transmits them to the antenna where they can be picked up by a distant base station). Where do you start?

RFIC design has some features in common with house building. Just as the blueprint for a house dictates the number of bedrooms and bathrooms to be built and the hallways connecting them, the blueprint for an RFIC—called the architecture—establishes the kinds of elements the RFIC needs to fulfill its intended function. Instead of rooms, the architecture includes, for example, the number of stages of amplification your power amplifier needs. Instead of hallways, it shows the paths that signals must take to get through those stages.

The blueprint for RFICs is actually mostly hallway; passive elements, like inductors and transmission lines, take up far more real estate than active elements like transistors.

Here’s why. As you have probably experienced yourself, a typical CPU’s transistors overheat when faced with operating frequencies of just a few gigahertz. The frequencies RFICs can operate at are higher by an order of magnitude—28 and 39 GHz for 5G signals, 26.5 to 40 GHz and even higher for satellite communications, and 77 GHz for automotive radar. Under this onslaught, a CPU’s transistors would fail.

RFIC transistors avoid this fate because these chips cleverly manage the signal’s energy with careful electromagnetic design. This takes the form of byzantine networks of metal elements that dominate the chip’s real estate. These structures are geometrically regular, often symmetrical, and so intricately constructed they sometimes resemble lacelike filigree. But while they may look decorative, they are essential to the chip’s functioning.

Electrically speaking, these “hallways” work more like the chip’s plumbing. Like plumbing, this extensive labyrinth of passives confines electromagnetic energy only to the places it should be traveling around the chip.

The major challenge in RFIC design is putting all these elements together to ensure they work, just as constructing a house from its blueprints demands exact specs for load-bearing beams, pipes, and external walls. On an RFIC, the architecture needs to be realized with physically fabricable transistors and passive components that are connected just so, to permit the signal to travel through the chip and be processed. The way these devices are connected locally is what we call the circuit’s topology.

The RFIC Design Process

To make that power amplifier, then, your first step is to identify a candidate circuit template: The combination of structures that will meet the goals of a particular architecture with a specific circuit topology. Over the years, researchers have eased your burden by developing reusable design templates for specific functions. For example, templates suggest how many amplification stages a circuit needs (because sometimes, combining the output of two smaller amplifiers will result in better bandwidth and efficiency than you would get from a single larger one). And they suggest what the general configuration of the passive structures should be. Today there is an extensive library of such templates.

However, these can’t simply be used off-the-shelf, because each comes with trade-offs. Some have better gain at the expense of stability; some better bandwidth at the expense of efficiency; still others are more energy efficient at the expense of output power, and so on. There is rarely a clear best choice.

To arrive at the “sweet spot” where all these different parameters are balanced into optimal harmony, designers will typically lay out several different versions of the circuit, using intuitions and methods they have picked up in their years of training.

The challenge is that the decision around the architecture, circuit topology, or the electromagnetic passives cannot be done separately. One decision influences the others. So, designing an RF circuit can often feel like trying to fit an oversized carpet into too small a room—press down one corner, and another pops up.

At microwave and millimeter-wave frequencies, even the smallest misstep is the difference between a chip that works and one that doesn’t, and any number of things can go wrong. For example, when an electromagnetic wave encounters a transistor—or any other component —the path it travels must be properly “matched” to what comes next. If it isn’t, some of the energy reflects backward instead of flowing forward. Imagine trying to connect a high-pressure fire hose directly to a narrow garden hose. Without the right adapter, water will splash backward at the junction. Very little will make it through. In electronics, this is called the impedance-matching problem.

To prevent those reflections, engineers design special transitions, essentially microscopic adapters, that smooth the handoff between components. On a chip, these adapters can be surprisingly intricate. They don’t just pass the signal along; they can also split it, combine it, or distribute it across multiple paths with carefully controlled timing and strength.

Once you’ve done the architecture, plumbing, and everything in between comes the moment of truth. Have all the choices you have navigated through the enormous design space resulted in an RFIC that meets its specifications? If the specifications are not met, you will have to go back, either redoing the topology or the entire architecture, and repeat the whole process. So get ready for months of time- and resource-heavy simulation and iteration. Perhaps you now see why, for decades, a core belief has persisted in the RFIC community: “RF design is an art.” It was said that only an experienced designer—with an artisanal understanding of how the pieces make up the whole—could master the subtleties of analog and RF design. Unfortunately, this entrenched notion has long held back algorithmic innovations in the field just when we need them most. Traditional, artisanal RFIC design is hitting its limits as the complexity of these systems inexorably grows.

AI for RFIC Design

While RFIC designers continued their battle against their “oversized carpet” problem, a series of interesting developments emerged in allied disciplines. Across a range of other previously intractable problems like protein folding and climate modeling, AI has been able to successfully navigate multidimensional complex spaces. This gave us the incentive to look deeper into AI for RF. After all, the combinatorial complexity of protein folding is not that different from the nature of the design space in our domain.

We were not the first to think of using artificial intelligence to speed up parts of RFIC design. Researchers had previously trained machine learning algorithms on circuit templates in the hope of speeding up the normal optimization processes. While this approach was undoubtedly faster than humans at optimizing templates, it still relied fundamentally on libraries of existing designs invented by humans.

We didn’t want that. We wanted to break free from the restrictions of prefabricated topologies. Because while a designer’s experience and hard-won heuristics are crucial to building a working design, they also place fundamental limits on it. Furthermore, such an approach would necessarily require simulation steps as part of the optimization cycle, and even the fastest simulations use a lot of computing resources. Worse still, in many advanced cases, such as for broadband designs, there are no existing templates.

But if we didn’t start with templates, where could we start?

The goal here was to allow algorithms to determine—entirely from scratch—every parameter for architecture, constituent circuits, and electromagnetic passives. This approach differs fundamentally from conventional optimization, which is limited to determining the parameters—like transistor dimensions and passive component geometries—that optimize structures originally devised by humans.

In our new approach, the architecture begins essentially from nothing and is progressively assembled through successive iterations. The system explores the design space by generating myriad candidate circuit combinations and mapping the resulting performance trade-offs as it navigates this landscape. Because the process is not biased by prior human design choices, it can produce completely novel circuit topologies that look markedly different from those created by human designers.

In some ways, the approach echoes AI systems such as AlphaGo Zero, which achieved superhuman performance not because it was trained on games played by humans but because it explored the rules by playing against itself. Similarly, our algorithm develops new circuit architectures by exploring and evaluating its own design strategies. In so doing, it learns to understand circuits, electromagnetics, and the close codesign they need to achieve the end-to-end design of RFIC.

Inverse Design for RFICs

To realize this capability, we proceeded in two stages. First, we developed a reinforcement-learning (RL) framework that determines the optimal system architecture, circuit topology, device parameters, and even the properties of the electromagnetic interfaces that connect different circuit elements. In this stage, the algorithm effectively defines how signals should propagate and interact across the system.

The algorithm trains very similarly to how a computer learns to play a game. If you let it play enough times, it can learn to play better by observing the relationship between the actions it took and the score it achieves. In a similar way, the RL agent here learns to design effective circuits by playing with a set of combinations, and over time, it can map the space between the circuit performance to its architecture, topology, and parameters. This training takes a few days to a week, but once trained, the agent can design circuits very quickly

The next step was to determine the physical structure of the IC’s electromagnetics—the plumbing—that can create the desired properties of the passive elements, which are characterized by a set of metrics called scattering parameters. These measure if a signal entering a component actually moves forward—or is reflecting backward, being wasted, as in our previous example with the fire hose and the garden hose.

Deriving the structure from the desired scattering parameters is an example of an approach called inverse design, which appears across many areas of engineering. In structural engineering, for example, one might collaborate with an architect on a physical goal—such as creating large interior spaces with high ceilings—and then determine the arrangement of arches or buttresses that can support it.

Generative AI for Electromagnetic Networks

But RF integrated crcuits pose a particular challenge for inverse design: The process must account simultaneously for circuit behavior and the electromagnetic responses of the interconnects and passive elements that link them together. But it has to figure that out without doing a lot of artisanal iterating.

So we replaced our RF circuit simulator with an AI-based emulator. This AI model can predict the behavior of electromagnetic fields going through any structure—even totally arbitrary two-dimensional shapes—without having to compute the underlying physics from scratch, as simulation tools do. It would predict the solution of Maxwell’s equations and tell you the scattering parameters for any structure you showed it, without actually doing the math. With such an AI in hand, what a time-consuming electromagnetic solver normally takes minutes or hours to accomplish is reduced to milliseconds.

We chose to build our emulator around a convolutional neural network—a machine learning model that has been remarkably successful for image processing. Such networks can extract spatial features from any structure, and it turns out that the image of a structure contains a lot of spatial information that can accurately predict its electromagnetic performance. Then we trained it on a vast number of random pixelated structures whose scattering parameters had been labeled.

Once we had our inverse-design RL and suitable AI emulator, we essentially had an end-to-end AI designer. So we asked it to design us a power amplifier.

Unconventional RF Architectures

In 2023, we published this proof of concept—a power amplifier targeting the millimeter-wave band, specifically spanning 30 to 100 GHz, which covers most of the relevant 5G and radar frequencies. The final design achieved the best combination of wide bandwidth, output power, and efficiency then reported for a silicon-based power amplifier—meaning it could amplify a large amount of data across a wide swath of frequencies—while maintaining record efficiency.

The structure of the IC’s electromagnetic pathways was unlike anything any human would ever consider. Since the AI is not trained on human designs, the layout that emerged looked more like an arbitrary pattern or perhaps a QR code than the regular symmetrical structures we are used to seeing.

One unexpected insight revealed by this prototype, and our research generally, is that there’s no evidence that the templates we’ve historically relied on are even close to optimal for modern design goals. It’s not that a human designer can never come up with a better design. But with the removal of the templates and the time to synthesize cycle upon cycle of optimized circuits, it is now clear that AI-driven synthesis could break traditional design barriers and push the limits of RFIC capabilities.

Our 5G amplifier had only one input port and one output port. Adding more inputs and outputs to a design is not straightforward. Every port electromagnetically couples to every other port, so the scattering parameters quickly add up. Two ports give you four scattering parameters. Four ports, 16 scattering parameters. The math gets ugly fast. Could our model keep up?

We next trained our model on larger classes of electromagnetic structures with many input and output ports. In 2024, we published work showing that multiport integrated circuits are no problem for these AI algorithms either. Where previously multiport electromagnetic simulation required days or weeks of toil, this model evolved new structures in minutes. Since then, a plethora of work in the space by research communities across the globe have demonstrated the power of inverse design in RFIC.

Combining the reinforcement learning framework with the inverse design, we now had the ability to create an RFIC from specifications all the way to a fabrication-ready layout. We’ve so far shown this is true for RFICs ranging from low-noise amplifiers to subterahertz and broadband power amplifiers. The hope is that this will work just as well for other circuits.

Making AI Designs Interpretable

Our goal was to make RFIC design better and easier, but we didn’t want to make it beyond human understanding. Chip testing and debugging is a long, arduous process, sometimes even more so than design. Engineers often prefer ICs to have interpretable structures, so that if a problem crops up, they can understand how the chip works well enough to debug it.

To create structures that are more interpretable, we turned to diffusion models, which you may know from their remarkable ability to generate realistic images from text prompts.

AI-driven synthesis could break traditional design barriers and push the limits of RFIC capabilities.

Imagine you go to your favorite image-generation engine and ask it to create a painting of the sky in the style of Picasso, Van Gogh, or Michelangelo. You will get images that capture the essence of their brushstrokes, their use of colors, and their framing. All are pictures of the sky nonetheless, but in different styles.

Electromagnetic design is similar in that multiple structures can have very similar electromagnetic responses. Instead of using text input, we used scattering parameters as our input, and the electromagnetic structure of an RFIC chip as our output. As part of the inputs to the diffusion model, we created a dial that sets the spatial frequency of the final structure. By turning the dial, a designer can direct the model to synthesize structures with low (classical-looking and interpretable), medium (mazelike structures), or high (pixelated or arbitrarily-shaped) spatial frequency.

From prompts to output, the entire process took about 6 minutes. With this diffusion model, algorithms can now both discover novel architectures and accelerate the creation of conventional, so-called classical ones.

All an RFIC designer needs to do is specify virtually any valid set of scattering parameters. As long as they are physically realizable under Maxwell’s equations, the model pops out a corresponding structure as if it were a vending machine.

The Future of AI-Driven RFIC Design

The results of our investigations have drawn the attention of the RF community. The traditional bottom-up design process is clearly beginning to reverse.

But there are still questions: How generalizable are these methods? Can they consistently deliver truly high performance? Can we get to a place where AI produces designs that maximize every conceivable trade-off, holistically optimizing every parameter to its most ideal physical state? We want to take this strategy beyond RFIC design and invent other kinds of circuits that are different from anything humans have ever done.

These are exciting and ambitious prospects, but we are not there yet. AI can hallucinate a design that creates bad circuits that don’t work. This means verification methods need to remain under human oversight. And, while hallucinations are rare, it would still be good to reduce their occurrence.

History suggests that meeting these dreams of the future will take much more data than we’ve been using. Before the creation of the ImageNet repository—a repository of 14 million varied, human-annotated images—image-recognition models didn’t function well in the real world. The datasets they had been trained on were too tiny to be effective. ImageNet’s massive amounts of training data ushered in a revolution that led to AI that can generalize and recognize images in the wild. The rest was history.

If the goal for RFIC and analog design is a universal foundational model—something that learns the governing laws of electromagnetics and circuit behavior—then we also need data.

The good news is that this data is plentiful. Around the world, countless engineers at companies and academic labs simulate nearly identical RF circuits and passive structures every day. The bad news is that it’s all locked away behind nondisclosure agreements.

Open ecosystems have propelled other areas, and we think the RFIC community should do the same. There had been some movement toward this. Natcast, the operator of the U.S. CHIPS and Science Act’s R&D program, would have bolstered shared infrastructure and innovation for the next generation of wireless, sensing, and defense technologies. Unfortunately, both the organization and the program it ran specifically for machine learning and RFICs have been closed.

But the momentum Natcast’s effort sparked hasn’t died out. Building on our early work, groups across the community have already demonstrated remarkable advances. AI-driven IC design is part of a much broader technological shift. From biology and materials science to automotive and aerospace engineering, AI is reshaping how complex systems are conceived and optimized. Deeper collaboration between AI researchers and chip designers will unlock the field’s full potential. It’s by no means a foregone conclusion, but if we get this right, this genie won’t stay in its bottle.

5G telecommunications, according to industry hype when 5G first launched in 2019, was going to be all about buzzy applications like mobile augmented reality and autonomous vehicles. But the surprise plot twist came when replacing home cable internet turned into 5G’s most widely adopted new application.

Fixed wireless access (FWA) now serves over 14 million U.S. customers, and contributes 28 percent of worldwide wireless traffic. Fixed wireless access is what the term sounds like: broadband internet delivered over a cellular radio link to a stationary location—no cable, no fiber, no trenching, no satellite broadband antenna pointed at the sky. What makes FWA distinctive is that it repurposes the same towers, spectrum, and 5G infrastructure that was built for mobile devices.

One U.S. Federal Communications Commission commissioner has called FWA 5G’s killer app. And that’s true not just in the United States either. Jio, India’s largest carrier, is also one of the world’s largest FWA providers, with over 9 million customers as of last year.

Carriers discovered they could repurpose surplus 5G capacity, while also exploiting a usage pattern quirk: mobile traffic starts to drop after 8 p.m., just when home internet usage peaks. The result is broadband, delivered via traditional cellphone towers, at a lower cost than fiber deployment. For these reasons, FWA provides real price competition to cable broadband, while reaching underserved rural and suburban communities.

Fixed Wireless Access Repurposes Ambitious 5G Infrastructure

FWA is cheaper to deploy than fiber, and for most homes and small businesses, fiber’s gigabit speeds are overkill anyway. And since FWA uses the same wireless networks built for cellular service, FWA works anywhere that receives a steady cellular signal.

As cellular networks extend into areas with minimal service, FWA’s coverage map expands with them. In these remote locales, the other main viable broadband alternative typically comes from satellite services like Starlink—which are, compared to FWA, more expensive, with higher delays, and lower bandwidth.

While most FWA deployments use currently underused microwave bands, some FWA deployments use electromagnetic spectrum that 5G launched but that mostly failed with mobile users. Millimeter waves operate at frequencies 10 to 40 times higher than 4G’s spectrum, offering high data rates from their wide available bandwidth.

However, there are good reasons 5G mobile users today don’t generally use millimeter-wave spectrum. Millimeter waves can’t penetrate buildings. Plus, they lose signal strength within a kilometer or two of the transmitter. Millimeter-wave antennas are also a real drain on cellphone batteries compared to microwave and radio-wave tech.

Yet none of these challenges applies to a fixed station with a clear line of sight to a nearby tower. FWA home units (called customer premise equipment or CPEs) outperform 5G handsets by a significant margin. That’s mostly because of hardware. CPEs carry larger, more sensitive antennas than a typical cellphone, paired with more capable transceivers. CPEs also tend to be plugged into wall outlets, making battery concerns a nonissue.

Another 5G technology that did not gain traction in mobile wireless is multi-user multiple-input multiple-output (MU-MIMO). A base station with MU-MIMO uses an array of antennas to serve multiple users on the same frequency simultaneously.

However, maintaining a MU-MIMO signal involves tracking each user individually—a problem that quickly becomes overwhelming with enough mobile users. FWA is different, however. Static CPEs, with their steadier downlink traffic loads, are an ideal match for MU-MIMO technology.

So, FWA internet service not only uses mostly fallow spectrum but also uses 5G spectrum more efficiently than do 5G mobile users—for whom, of course, these 5G technologies were originally designed!

How FWA Became 5G’s Surprise Killer App

Not long ago, the high-bandwidth use cases for 5G made for an impressive list: millisecond latency for autonomous vehicles, mobile augmented reality headsets with extensive high-speed data needs, and massive machine connectivity for an expanding internet of things (IoT).

These applications have all stalled. Autonomous vehicles pose challenging—and still unsolved—problems unrelated to spectrum allocation. Augmented and virtual reality technologies have yet to create meaningful spikes in bandwidth demand. And the IoT has, to date at least, fragmented across an array of competing standards.

Mobile carriers had built dense 5G networks for mobile customers whose needs rarely saturated the network’s capacity. Home broadband usage peaks in the evening hours, precisely when cellular networks are quietest.

FWA sits at cellular networks’ crossroads of supply and demand.

The Advent of 6G Will Only Expand FWA’s Reach

In December, the telecom standards body, the Third Generation Partnership Project (3GPP), issued its latest 5G specification—Release 20, the final “5G only” update. So, although 6G is still years away (its first specifications are expected in early 2029), engineering decisions that will define 6G are being made today. And FWA is not on the margins of that conversation; FWA is currently considered an established day-one use case.

6G wireless technology promises to expand FWA’s reach—not only via spectrum but also via geometry. Instead of following 4G and 5G’s connectivity model—strong signals near towers and weak signals far away—future 6G networks will let homes connect to multiple towers simultaneously, using a technology called distributed MIMO (multiple-input, multiple-output).

Where 5G’s version of MIMO (a.k.a. massive MIMO) concentrates user communication with dozens of antennas at a single tower, distributed MIMO uses antennas across multiple base stations and coordinates them to deliver signals to your home from multiple directions simultaneously.

The practical result: Because no single tower is responsible for any given connection, the “edge” of a cell network—that outer boundary where signal strength falls off and service degrades—no longer represents a hard limit on who gets well served. A home that would once have been too distant from a tower, or blocked by terrain, could now be within reach of several base stations working together.

6G may eventually adopt distributed MIMO technology for mobile users, when synchronization challenges and other signal engineering hurdles are solved and deployed for real-world cellular networks. The jury, as of 2026, is still out on whether the full distributed MIMO problem will be solved once the 6G standards start to be set in place, within three years.

As demand for FWA grows, carriers will also deploy increasingly capable millimeter-wave infrastructure for fixed customers first—the stationary CPE use case that millimeter wave best suits. The dense millimeter-wave antenna infrastructure that FWA requires is the same infrastructure that future mobile applications will eventually inherit. AR glasses, AI-powered wearables, and other bandwidth-hungry applications originally promised for 5G are not canceled—they are waiting for the infrastructure to arrive.

The pathway to FWA is being prepared at lower frequencies, too. There is growing interest today in the largely unoccupied FR3 band, which spans roughly 7 to 24 gigahertz, situated between crowded low/mid-bands and the much higher millimeter-wave frequencies.

Recent field trials by Nokia have demonstrated FR3’s viability for both cellular and FWA applications. FR3 is emerging as one of the more promising near-term frontiers for extending FWA coverage beyond its current footprint.

None of this was the plan. No carrier executive in 2020 stood on a stage and announced that 5G’s defining achievement would be delivering living room broadband to rural homes and suburban subdivisions underserved by cable.

FWA became 5G’s killer app because the engineering economics made it happen. Surplus wireless capacity met unmet consumer broadband demand, with the physics of a stationary receiver doing the rest.

That is not a criticism of the engineers or the carriers. It is simply how technology sometimes advances—sideways, through gaps nobody was trying to fill.

But FWA’s model of prioritizing unconnected users may in the end prove to be telecom’s on-ramp to everything else. Fix the digital divide first. Tomorrow’s sci-fi future appears set to follow close behind.

A man raises his phone as police move into a crowd. The video is shaky, loud, immediate. Within minutes, it is online. Within hours, it is everywhere. This is how accountability works now. Something happens, someone records it, and that footage can show what really happened, sometimes contradicting official accounts. It can empower citizens and create consequences for officials.

But the footage’s life cycle does not end there.

In recent months, civil liberties groups have warned that adding facial recognition to consumer smart glasses could turn everyday recording into something more troubling: real-time facial identification. It reflects a broader shift already underway, where images and videos captured for one purpose can later be searched, matched, and used for another.

An ouroboros is an ancient Egyptian symbol, a snake or dragon eating its own tail. As I began to see patterns in my broader research on surveillance corporatism and governance lag, I began using the term “surveillance ouroboros” to describe this recursive pattern of observations intended to hold power accountable becoming new input for the same surveillance infrastructure.

Facial recognition changes accountability

During the George Floyd protests in 2020, people filmed police in real time. Phones were pointed at officers, not at each other. The goal was simple: to show what the state was doing. That footage spread quickly and became part of a much larger pool of public data.

At the same time, reporting from outlets including The New York Times and BuzzFeed News showed that law enforcement agencies were using facial-recognition tools, including systems built by Clearview AI. Those systems were built from billions of images scraped from across the internet, including publicly available photos and videos.

The basic approach is now routine: People record the state, or anything else (as in the January 6 attack on the U.S. Capitol), and the state compiles that footage and data into a searchable environment, which may later be used to identify some of the same people who made the footage.

Facial-recognition systems used by law enforcement are increasingly outpacing the legal safeguards.

A 2023 Government Accountability Office review found that federal law enforcement agencies continued to expand their use of facial-recognition systems for criminal investigations despite ongoing concerns around training, privacy protections, civil-liberties safeguards, and oversight. Earlier GAO findings showed that agencies had conducted roughly 60,000 facial-recognition searches before formal training requirements were put in place for personnel using the systems.

The American Civil Liberties Union and other groups have warned that these tools could be used to identify people from images shared online, including protest-related footage. Concerns about facial recognition led some U.S. states and cities, including San Francisco and Boston, to restrict or ban government use of the technology, while federal agencies have continued to face scrutiny over how such systems are tested, deployed, and audited. A 2024 analysis published in Internet Policy Review warned that facial-recognition systems used by law enforcement are increasingly outpacing the legal safeguards meant to govern them, creating growing tensions around data protection, oversight, and proportional use.

The spy network that built itself

Surveillance used to require infrastructure. Cameras had to be installed, and data had to be collected deliberately. That is no longer the case. People carry cameras everywhere. They record constantly and upload in real time. Events are documented from multiple angles without planning or coordination. The cumulative result is a continuous stream of usable data: faces, locations, timestamps, and interactions. The Internet of Things (IoT) also waits all around us, gathering information and releasing it when people least expect it, as Andrew Guthrie Ferguson describes in a recent excerpt of his book Your Data Will Be Used Against You.

RELATED: “Sensorveillance” Turns Ordinary Life Into Evidence

Similar dynamics are emerging globally. A recent analysis in the International Journal of Law and Information Technology examined how facial-recognition systems in China and Japan are expanding faster than the legal frameworks governing them. Reporting by The Guardian described the limited legal protections around the rapid deployment of AI-assisted surveillance infrastructure across parts of Africa.

There used to be a clear distinction between surveillance and accountability. Surveillance meant the powerful watching the people; authorities tended not to share their imagery except under duress or a court order and usually after a long delay. Accountability meant the people watching the powerful and often publishing imagery immediately to head off or counteract official mischief. That distinction no longer holds. The same footage can serve both roles. A recording meant to expose misconduct can later be used to identify someone else entirely.

Surveillance ouroboros is not a future risk. It is already here.

This dynamic persists because people still need to record. In many places, it is one of the only tools available when formal accountability breaks down. When oversight institutions weaken or fail, public documentation becomes a substitute. In that environment, people turn to visibility. But that visibility comes with a cost. The more people that document, the more data that exists. The more data that exists, the easier it is to search, match, and store. Every video feeds the ouroboros. People are not feeding the system because they trust it. They are feeding it because the alternative is silence.

Most of the people in these videos are not the focus. They are in the background, passing by or standing nearby. But that distinction does not matter once the footage enters a system. Today’s facial recognition can identify even a face that passed through the corner of a frame. Someone who did nothing can still become part of a dataset without ever knowing it. As recognition systems improve, older footage becomes more useful—and invasive.

No single decision created this outcome. It emerged gradually through more cameras, better recognition, larger datasets, and easier integration. Each step made sense on its own. Together, they changed what recording means.

Public recording is still necessary. Without it, many forms of abuse would remain hidden. But recording is no longer just exposure. It is also contribution. If you published imagery or video last year, you may already have contributed to a system you have never seen but the ouroboros has.

Surveillance ouroboros is not a future risk. It is already here. Every time someone presses publish, they are doing two things at once. They are exposing power, and they are helping build the system that the powerful will later use to track the less powerful.

The decades-old “satellite-IoT” race, in which Internet of Things devices connect to satellites in orbit, has reached an inflection point in 2026. Satellite-IoT is growing in response to a rising need for connected devices in applications including livestock tracking, wildfire detection, global supply-chain monitoring, pipeline and grid sensing, and other industries where network connections can be sparse or expensive.

Many anticipated satellite-IoT use cases will require two-way signaling, carrier roaming, or continuous availability—all things that are common features for cellular networks. That’s why the telecom standards body that created 4G and 5G has, as part of the development process for 6G, already settled on a satellite-IoT protocol that will make use of the upcoming generation of cellular tech, to be available sometime around 2030.

But 6G won’t work (or work well) for some IoT devices, because they won’t use SIM cards, or they’ll have tiny sensors that need years of battery life, or they’ll operate on networks that need to stay strictly private.

That means that a window currently exists for alternative technologies to stake their claim in the growing space. That’s exactly what two aerospace companies, Kinéis and Hubble Networks, have done in recent months, in their efforts to lock up enough hardware, customers, and infrastructure to stave off the oncoming challenger.

LoRa Satellite IoT Deployment

At the IEEE International Conference on Communications in Glasgow last week, the Toulouse, France–based satellite operator Kinéis presented the results of its early satellite-IoT commercial deployment—25 microsatellites in low Earth orbit, opened to customers in June 2025. A team of company representatives and university researchers made the case that LoRa—a radio protocol developed for long-range, low-power IoT—fits a world in which millions of cheap sensors are all communicating with orbiting network hubs.

“LoRa is explicitly designed to address terrestrial and satellite IoT use cases without requiring additional firmware,” said Vincent Deslandes, strategic product manager at Kinéis. He added that the version of LoRa that the company uses, one that continuously hops across carrier frequencies to reduce signal collisions, will span a coverage area “thousands of kilometers of diameter” per satellite, while serving thousands of users simultaneously.

Deslandes said that while 6G’s satellite-IoT market “will expand with a more exponential curve,” the niche will remain in place for low-power alternatives to 6G and its precursor standards today.

“There is room for both [6G] and LoRa-based satellite-IoT, as they answer to different needs,” Deslandes said.

Researchers at ground stations in France and Switzerland in 2024 tested a new, ultraefficient protocol for satellite-based Internet of Things (IoT) signals. Satellites overhead successfully received the test transmissions. Kinéis

In the International Conference on Communications paper, a team of researchers from French and Swiss universities, along with two Kinéis collaborators, described the results of a 2024 experiment involving test signals transmitted using comparatively lower frequencies, in a band long associated with wildlife tracking and ocean buoys.

The group experimented with an ultraefficient protocol that transmits only data packets and error correction bits. (Most signals contain preambles and reference sequences. For an energy-constrained chip like a tiny IoT device, that means wasting battery on unnecessary overhead.) The researchers found that a low Earth orbiting satellite overhead could pick up 100-milliwatt pings sent from the ground. Signals using the experimental protocol, the researchers found, could be deciphered even when ambient noise outpowered the signal—by a factor of nearly 27 to 1.

Hubble Network Bluetooth Satellites

While LoRa squeezes every efficiency from purpose-built radio hardware, the Seattle-based startup Hubble Network is promoting a different approach to satellite-IoT. Using a low-energy standard developed by the Bluetooth consortium, Hubble Network’s seven satellites can pick up signals from any device with a Bluetooth chip.

“The receive sensitivity that we have with our satellites is roughly the same receive sensitivity as devices have with the GPS network,” says Hubble Network CEO Alex Haro. “So anywhere a GPS device could get a lock, it could send a Hubble packet and be received by our satellite.”

In April, the company announced a new partnership with the Irvine, Calif.–based sensor startup InPlay. The companies are developing technology around a range of supply-chain and logistics applications.

“The use cases that we’re focused on are what IoT was meant to be,” Haro says, “which is, how do we connect everything and eventually get everything an IP address? How do we do things like track every package ever shipped, so that we can help solve the porch pirate problem and the massive insurance liabilities in shipping? Or how do we keep track of every pallet to make moving of materials and food much more efficient?”

Some Bluetooth chips, whose pings Hubble Network’s satellites can receive from orbit, are approaching the price of RFID tags—which require cumbersome RFID reader devices as well. Whereas Bluetooth chips using the Hubble Network, Haro says, talk directly to satellites overhead.

“If you could have a global network that works with a 10-cent Bluetooth chip, you can really start to solve these interesting problems, like how do we keep track of every package that UPS has shipped?” Haro says.

Haro notes that, unlike anticipated 6G satellite-IoT standards, Hubble Network is uplink only. “And our users find a lot of value in that,” he says.

According to Petar Popovski, professor of connectivity at Aalborg University in Denmark, Bluetooth seems “quite limited as a standalone satellite-IoT contender.” Without a viable downlink, he says, a Bluetooth device on the ground can whisper messages to low-Earth orbit. But two-way, standalone communications require both a receiver and a transmitter.

Popovski adds that while LoRa wide-area networks were designed to send and receive long-distance signals, LoRa assumes “predominantly uplink-dominated traffic, which is well-suited to sensor reading but creates structural limitations for bidirectional applications.”

Both LoRa and Bluetooth, in other words, will ultimately struggle to compete with 6G for applications that demand steady, two-way connections between IoT devices below and satellites above. Such higher-bandwidth, bidirectional satellite-IoT connections are expected to underpin future tech innovations in smart cities, autonomous vehicles, industrial automation, digital twins, and remote surgery.

NB-IoT NTN and 6G Satellite IoT

Strictly speaking, the anticipated 6G satellite-IoT protocol is currently called Narrowband Internet-of-Things Non-Terrestrial Networks (or NB-IoT NTN). While a version of NB-IoT NTN was ratified in 2024, it will become far more useful for more applications when it is rolled into broader 6G standards in the coming years. And for all the name’s unwieldiness, the standard itself remains the frontrunner for many anticipated satellite-IoT markets in the coming decade.

Which still leaves plenty of room for competition, as Kinéis’ and Hubble Networks’ recent product developments and announcements reveal. Meanwhile, Kinéis and Hubble Networks are themselves far from the only companies working today to connect remote devices to the sky.

Legacy satellite-IoT operators Iridium (located in McLean, Va.) and Globalstar (located in Covington, La.) have served industrial customers for years using proprietary protocols. Myriota, a startup based in Adelaide, Australia, recently developed a satellite-IoT payload designed for lunar deployment that pushes the technology’s frontier deep into cislunar space. And the Mountain View, Calif.–based Skylo is already offering commercial NB-IoT NTN service via geostationary satellites. The same standard that’s now on the pathway to 6G, that is, is also a player in the current satellite-IoT race.

And it’s expected to be a tight and competitive race for at least a few more years. According to industry analyst Johan Fagerberg at the Gothenburg, Sweden–based Berg Insight, satellite-IoT built on cellular standards “will probably win the narrative after 2030, especially once 6G specifications are ready.”

He adds, however, that unlicensed technologies like LoRa and low-energy Bluetooth “will not disappear and will be relevant for ultralow-cost, low-payload, long-battery-life devices where users value simple modules and private/hybrid networks.”

Which is why, Popovski says, the clock is now ticking. “The window for building durable lock-in is real but likely time-limited,” he says.

Kinéis, Hubble Network, and every other contender in satellite-IoT are running hard against each other today. But the real finish line ahead is 2030, when the rules of the whole contest change.

This article appears in the August 2026 print issue as “Satellite IoT Faces a 6G Deadline.”

Direct-to-cell technology uses LEO satellites as spaceborne cell towers. It delivers LTE services to existing smartphones without hardware changes, bridging global coverage gaps.

What Attendees will Learn

1. How DTC works as a spaceborne cell tower — LEO satellites carry LTE eNodeB payloads in regenerative mode. How they serve unmodified phones using quasi-earth-fixed multi-beam antennas. How the satellite compensates for Doppler shift and time delay on thenetwork side.
2. Why Doppler shift and round-trip time are critical challenges — A LEO satellite’s high velocity causes carrier frequency offsets in OFDMA systems. Pre-compensation at a reference point helps, but cell-edge users still face residual Doppler.
3. How spectrum sharing and regulation shape DTC deployment — DTC has no dedicated spectrum allocation. It relies on spectrum sharing between terrestrial and satellite operators or re-farmed MSS bands. How national regulations like the FCC SCS framework govern access.
4. Where DTC fits in the evolution toward 5G NTN and 6G — DTC is an interim technology offering fast time-to-market satellite services. It bridges the gap until 3GPP NR-NTN matures. How NR-NTN will bring purpose-built NTN features and international spectrum frameworks.

A practical introduction to phase noise concepts, explaining how oscillator instability affects RF systems and how phase noise is measured, analyzed, and reported.

What Attendees will Learn

1. What phase noise is and why it matters — Learn how real-world oscillators differ from ideal ones, why short-term frequency instability arises, and why phase variations typically have a much greater impact than amplitude variations on system performance.
2. How phase noise degrades system performance — Understand the most common effects of excessive phase noise: spectral regrowth, reciprocal mixing, and constellation rotation in digital communications.
3. How phase noise is measured and reported — Explore the spectrum analyzer method and the cross-correlation technique, understand single sideband (SSB) phase noise plots and spot noise tables.
4. What advanced phase noise measurements look like in practice — Discover additional measurement types including integrated phase noise, additive (residual) phase noise, pulsed signal phase noise, and amplitude noise.

The IEEE Communications Society (ComSoc)’s Research Collaboration Pitch Session initiative is proving to be a catalyst for meaningful engagement between academic researchers and industry innovators. Launched last year, the program connects promising researchers with industry leaders who can offer them funding, mentorship, and connections to bring interesting ideas closer to real-world deployment.

Rather than relying on chance encounters at conferences, the pitch sessions create a focused environment. Five academic presenters share their work with five industry representatives, known as “innovation scouts”: senior leaders primarily chosen from ComSoc’s Corporate Program partner companies such as Ericsson, Intel, Keysight, and Nokia. The curated format ensures that each idea receives dedicated attention from professionals who are seeking new concepts aligned with their organization’s priorities.

The initiative was launched in November at the IEEE Middle East Conference on Communications and Networking (MECOM) in Cairo and appeared in December at the IEEE Global Communications Conference (GLOBECOM) in Taipei, Taiwan.

AI-driven communication network

One of the most compelling outcomes came from the inaugural session in Cairo. Angela Waithaka, a student member and biomedical engineering student at Kenyatta University, in Nairobi, Kenya, presented her “AI-Driven Predictive Communication Networks for Enhanced Performance in Resource-Constrained Environments” paper. You can view her presentation along with others on IEEE.tv.

Waithaka’s research tackles a critical challenge: Next-generation communication systems increasingly rely on artificial intelligence and machine learning, yet most existing architectures consume abundant computational and energy resources, which are not always present in developing regions.

Waithaka proposed lightweight, adaptive AI/machine learning models capable of delivering predictive, reliable communication performance even under tight resource constraints.

Her vision resonated with Ruiqi “Richie” Liu, a master researcher at ZTE in China. ZTE is a global leader in integrated information and communication technology solutions. Liu says he recognized the relevance Waithaka’s proposal had to his company’s work with the International Telecommunication Union. He invited her to establish an ITU account so she could participate in the organization’s meetings discussing global telecommunications standardization projects—which would elevate her work to an international stage.

Simplifying data center protocols

The momentum continued at GLOBECOM. Among the presenters was Nirmala Shenoy, a professor at the Rochester Institute of Technology, in New York. Shenoy, an IEEE member, spoke on the topic of simplifying data center network protocols. She highlighted the growing complexity of the critical networks, which underpin cloud services, enterprise IT, and emerging AI workloads.

Shenoy’s focus on reducing protocol complexity while maintaining scalability, resilience, and low latency caught the attention of an innovation scout from Nokia, who heads its eXtended Reality Lab in Madrid. He found the key person at Nokia for Shenoy to connect with to discuss her research, and it led her to record a video for the company detailing her approach and its potential applications.

A model for accelerating innovation

The early success stories demonstrate the power of intentional, structured engagement. By bringing researchers and industry leaders together in a format designed for discovery, ComSoc is helping accelerate innovation and expand opportunities for collaboration. The pitch sessions are not merely conference events; they are becoming a bridge between academic creativity and industry implementation.

This year sessions will be held during the IEEE International Conference on Communications in Glasgow from 24 to 28 May, and more are scheduled during the IEEE International Mediterranean Conference on Communications and Networking in Sardinia from 6 to 9 July, and at GLOBECOM in Macau from 7 to 11 December.

As the program continues to grow, it could become a signature ComSoc initiative, one that strengthens the research ecosystem, supports emerging talent, and ensures that promising ideas find pathways to real-world impact.

A comprehensive review of how spectrum congestion, dynamic sharing, and cognitive radio systems are reshaping RF coexistence testing for military and commercial applications.

What Attendees will Learn

1. Why spectrum congestion threatens wireless reliability — Explore how over 30 billion connected devices, more than 4,000 allocation changes worldwide, and the expansion from 11 to over 80 cellular bands are intensifying contention for finite RF spectrum resources.
   
2. How real-world coexistence failures affect safety-critical systems — Understand the interference risks between 5G C band transmitters and aircraft radar altimeters, and between terrestrial L band networks and GPS receivers that were not designed for adjacent high-power signals.
3. Why tiered spectrum sharing frameworks are essential — Examine how CBRS uses a cloud-based Spectrum Access System (SAS) and environmental sensing to dynamically protect incumbent Navy radar while enabling commercial cellular services across three priority tiers.
4. What coexistence test architectures look like in practice — Learn how controlled environment testing with anechoic chambers, over-the-air signal generation, and standards such as ANSI C63.27 enable repeatable evaluation of RF device performance under real-world interference conditions.

For 10 days last month, Artemis II astronauts sent live HD video back to Earth using experimental laser links that cost a fraction of what ground stations might normally run. The connection ran through a telescope on a hillside near Canberra, Australia, which complemented two traditional receiving telescopes in the United States.

Compared with radio communications, infrared laser light packs up to a thousand times as much data, because of its higher frequency. Laser communications tech provides broadband connectivity from deep space, where previous missions had to make do with tenuous radio links of a few megabits per second. The Artemis II mission, among its many other accomplishments, served as a proof of concept for reliable and inexpensive space communications.

And NASA’s latest moonshot is hardly alone in that technological distinction. SpaceX, for example, uses laser links to haul large amounts of data between satellites of the Starlink constellation. Other companies have plans to build laser-based space relay networks around Earth that could replace undersea high-throughput fiber-optic cables in the future. But laser comms have an Achilles’ heel when trying to deliver data to Earth, said former NASA astronaut Josh Cassada. Once clouds intervene, data links break. Radio waves, despite their considerably lower bandwidth, can pierce through clouds with no difficulty.

Cassada, who retired from his astronaut career in 2024, is a cofounder and head of R&D at Quantum Opus, a Michigan-based startup developing ultrasensitive photon detectors, which helped NASA secure Artemis II’s laser links against inclement weather.

“If you’ve got clouds, the 1,550-nanometer wavelength that we’re using will scatter and never make it to the telescope,” Cassada said. “You need geographic diversity to immediately switch to another site that has good weather.”

During the 10-day journey to the moon and back, Artemis II’s Optical Communications System (O2O) laser terminal mounted on the Orion spaceship transmitted 450 gigabytes of data. Two ground stations in the U.S. (in Las Cruces, N.M. and on Table Mountain in Southern California) and one on Mount Stromlo near Canberra, Australia, served as Artemis’s terrestrial downlink sites. The Australian site tested a low-cost ground-terminal system developed by Quantum Opus, Los Angeles–based telescope manufacturer Observable Space, and Australian National University, in Canberra.

“In prior systems, you were looking at ground stations that cost tens of millions of dollars,” said Connor Poole, chief technology officer at Observable Space. “Our systems are in the single-digit millions.”

That fact alone could resolve the reliability problem. It means a network of ground stations worldwide can be built affordably, ensuring clear skies over an available ground terminal somewhere on the planet.

The Australian telescope helped reduce “a large portion of the ‘blind spot’ created by only using U.S. ground stations,” according to a press release from Observable Space. It noted that a network of 15 to 20 ground stations could provide round-the-clock connectivity for future missions to the moon and Mars.

During the Artemis II tests, the Australian ground station performed as well as the two American telescopes, transmitting 260 megabits of data per second, enough to stream 4K video and run multiple conference calls in parallel.

“The [Mount Stromlo] site was officially a demonstration to see if this would work,” said Cassada. “But about two or three days into the mission, NASA had transitioned it to [routine] operations. It was a mid-mission upgrade, which was really exciting to see.”

How Single Photons Carry Streaming Data From the Moon

The ground station relies on a 0.7-meter telescope from Observable Space fitted with a fast-steering mirror that focuses the incoming stream of photons onto a deformable mirror, controlled by actuators, which counteracts distortions to the signal from Earth’s atmosphere.

“You’re getting multiple orders of magnitude higher bandwidth at almost an order of magnitude lower size, weight, and power on the spacecraft with this technology,” said Poole.

At the heart of the system sits a single photon detector developed by Quantum Opus that intercepts the faint signal that traveled hundreds of thousands of kilometers on its way from the moon.

Quantum Opus CEO Aaron Miller describes the cryogenic detector as “the world’s most sensitive.” Cooled to near absolute-zero temperatures, the superconducting nanowire single-photon detector catches over nine in 10 arriving photons. When a particle of light strikes the wire, the energy released makes the material briefly transition from a superconducting to a normal state, creating a sudden voltage pulse, Miller said.

“By exploiting this discrete electronic trigger, [the detector] can register the smallest possible increment of energy with nearly perfect efficiency,” he said. “The active area of these sensors is usually only about 50 micrometers across or less—less than half the width of a human hair.”

In fact, because signals from the Quantum Opus detector were arriving too brightly, researchers had to reduce the device’s sensitivity.

Artemis II Tested How Future Missions Will Communicate

Artemis II’s data-transmission rate via O2O was around 5,000 times as great as the rate that Apollo-era missions of the late 1960s and early 1970s used.

That said, Poole added, the technology could still be scaled up to support communications at several gigabits per second.

Miller said the technology could also one day pave the way for quantum communication from space—for example, for quantum cryptography–secured communications. Laser communications networks would also provide a stepping-stone for constant high-bandwidth connectivity between the Earth, moon, and Mars, he added.

Mars ranges from about 55 million kilometers at opposition to over 400 million kilometers at its most distant. Even at opposition, a signal from Mars arrives far fainter than one from the moon—and such a signal already has roughly 1/10,000 the strength of a comparable one from low Earth orbit.

NASA

“It’s a good stepping-stone for much farther things,” said Miller. “If you want to go to Mars, you really have to push the detectors, the telescopes, all the technology harder, to get this sort of data rate from farther away. But it’s a proof of principle to even go to weaker signals.”

During the mission, NASA still relied on radio signals received via the agency’s Near Space Network and the Deep Space Network of radio ground stations. The O2O laser system transmitted less critical scientific information and enabled the crew to live stream their adventure in real time and in high definition.

“During Artemis II, the O2O system downlinked high-definition images from the lunar flyby shortly after they were captured,” said Jan Wittry, news chief at NASA Glenn Research Center, in Cleveland. “Because laser communications handled the high-bandwidth imagery, the mission’s radio-frequency links could remain focused on critical spacecraft telemetry and command data. Without O2O, those images would have either arrived with significant delay or at reduced quality.”

During the Artemis 2 mission, communication was lost only for about 40 minutes, when the Orion spaceship passed behind the moon. The station in Australia helped to minimize other communication blackouts caused either by clouds or by Earth’s rotation.