The hidden infrastructure behind intelligence

Connectivity is at the heart of nearly every major technology trend today. From mobile networks and AI infrastructure to satellite communications and IoT devices, RF enables the movement of information across the modern economy. It plays a critical role in every market we serve, making it one of the industry's most important and active areas of innovation. Every AI query, satellite connection and streamed video ultimately depends on RF technologies moving information. RF rarely gets noticed when it's working. Most people don't think about the technologies managing frequency, power and signal integrity until a call drops, a connection fails or data stops moving. That invisibility is RF doing its job.

To meet these demands, RF designers are being asked to do the impossible: higher bandwidth, more output power and greater efficiency, all in the same die area they had before. As AI pushes more computing onto devices, RF systems must provide greater performance using a smaller share of available power budget, while supporting more frequency bands, antennas and simultaneous wireless connections. What once required tradeoffs increasingly requires everything at once.

At GF, our strength has never been a single technology or application. Over the past two decades, we’ve built leadership across RF technologies and the system-level expertise required to bring them from innovation to high-volume manufacturing. RF-SOI, SiGe and RF GaN allow designers to optimize different parts of the connectivity chain, while FD-SOI, FinFET and silicon photonics help move data through increasingly complex communications and computing systems.

"We can serve nearly every component of the radio chain."

As the boundaries between terrestrial and satellite networks continue to blur, new architectures are emerging that promise to extend connectivity to places traditional infrastructure cannot reach. Concepts like direct-to-cell communications, alongside advances in 5G evolution, future 6G networks and next-generation satellite systems, will place even greater demands on the RF technologies that enable them, while creating new opportunities to rethink how wireless systems are designed and deployed.

In this issue, we explore how our RFGaN1 and SLATE™ advanced packaging technology are turning that leadership into scalable platforms customers can deploy today. We also go inside a GF Labs breakthrough that reimagines RF switching itself, built on GF's commercial RF-SOI platform and published earlier this year in Nature Electronics.

— Shankaran Janardhanan , SVP of Photonics & RF

Doubling down on GaN: A Q&A with Dan Denninghoff

RF GaN has spent years as a specialty technology. Earlier this year, GF moved it into volume production, expanding GF's ability to serve connectivity applications that demand higher power and efficiency while complementing the broader RF portfolio customers already rely on. Dan Denninghoff , Senior Director of RF GaN Product Line, breaks down what RFGaN1 changes for designers building the next generation of comms and defense systems.

RF GaN has been around for a while. Why is moving it into production at scale a big deal?

Dan: Up to now, GaN has lived as a high-performance, niche technology that a designer would reach for on a project with a big budget. RFGaN1 is our first RF GaN-on-silicon technology qualified for volume production. It combines GaN’s high output power, efficiency and frequency performance advantages with the cost structure and volume of 200mm silicon manufacturing, enabling customers to build a real supply chain around it instead of treating every design as a one-off.

We are already seeing this shift play out. At IMS 2026, Falcomm launched GaNdalph.ai, an AI design tool built with first-class support for RFGaN1. Together, these tools speed up power amplifier design and deployment. The pairing works because the underlying process is stable and manufacturable at volume. Maturing this technology is becoming the foundation that other companies build tools and roadmaps on top of.

Who is this platform built for and what performance does it deliver?

Dan: RF GaN supports numerous emerging sectors that require efficiency and high-power wireless infrastructure. Power amplifiers and high-power RF switches alter the way satellite communications and cellular infrastructure, for example, sustain themselves. RFGaN1 runs 12 to 28V with depletion-mode devices, which yields sustained high output power and efficiency across 1 to 15 GHz applications.

“GlobalFoundries’ RFGaN1 platform provides a powerful foundation for the future of adaptive RF systems, delivering the performance and efficiency required for advanced front-ends in the critical defense, 5G/6G and satellite communications markets we serve.” - Victoria Pereira, CEO, Otava RF

What is the biggest barrier to GaN adoption?

Dan: Access and volume manufacturability are obstacles many customers come to us with. At GF, we’ve worked hard to address these pain points directly. We built early-access programs and GlobalShuttle™ MPW runs, so designers can prototype and validate cost-effectively before committing to full production. This is backed by our existing RF design infrastructure and GlobalSolutions™ partner ecosystem. Our 200mm fab in Burlington, Vermont is an Accredited Trusted Foundry, meeting the U.S. government’s requirements for manufacturing chips used in aerospace and defense systems. For many of the applications we’re talking about, it’s the reason a customer can use the platform at all. Combining GF’s high-volume 200mm manufacturing excellence with the power of GaN makes RFGaN1 the most cost-effective RF GaN in the world. This is an attractive feature for all our customers.

Read Dan’s full blog on what it takes to scale RF power and efficiency:

Smaller, smarter RF front-ends: Introducing SLATE on 9SW

GaN adds power. SLATE removes area. Together, they cover both ends of what next-gen RF systems need.

Every smartphone is fighting for physical space inside its own case. More bands, more 5G, more antennas, more sensors — they all compete for room. For years, this meant sacrificing features, battery life and performance.

This past June, GF qualified SLATE, a wafer-to-wafer bonding technology on its 9SW RF-SOI platform that challenges this notion. Instead of laying components flat and overtaking board space, 9SW SLATE allows designers to stack them vertically, bonding two wafers together and folding large transistors across both. The result is up to 45% less die area, with no loss in RF performance, giving designers more room to manage the growing complexity of modern wireless systems.

Antenna tuning is a good example that shows why space reduction matters. Every time you move your phone, your hand shifts or you walk into a building, the antenna’s performance changes. Tuners compensate for these micro-shifts in real time, switching on and off dozens of times a second so your call doesn’t drop and data doesn’t stall. It’s the underlying tech that no one notices when it’s working. Those switches must handle high voltage while staying small, and SLATE on 9SW means they can do that job in a fraction of the space they used to need.

Recommended by LinkedIn

Satellite terminals, IoT sensors, wearables and future AI infrastructure all face a version of the same challenge: delivering more capability in less space. By enabling new forms of integration across technologies, SLATE on 9SW helps designers rethink how systems are assembled rather than simply shrinking individual components. This marks an important advancement in RF front-end design, giving designers a new path to increase functionality and integration when traditional scaling approaches begin to reach their limits. 3D integration stops being a specialty project for the most space-constrained edge cases and can now turn into a routine design choice.

To learn more, join Cadence ’s upcoming webinar on October 27 where technical experts from GF and Cadence will discuss how SLATE™ advanced packaging technology on 9SW enables 3D RF integration and supports next-generation RF front-end design. Registration details coming soon.

The research-to-production pipeline: A new path for RF innovation

For decades, the semiconductor playbook was simple: shrink the transistor, get more performance.

That approach transformed computing. But in RF, engineers have been running into a different problem. As switches get smaller, parasitic effects and contact resistance increasingly offset those benefits.

So, Mohammad Nikoo , now an assistant professor at Nanyang Technological University Singapore, and Hua Wang , professor of electrical engineering at ETH Zurich, tried something unexpected. The work was carried out while Nikoo was a postdoctoral researcher in Wang's IDEAS Group at ETH Zurich. They built a switch around a fundamentally different mechanism, using displacement fields and quantum tunneling currents to carry a signal.

Built on GlobalFoundries' commercial RF-SOI platform through GF's University Partnership Program, the device demonstrated:

Sub-30 picosecond switching speeds 

More than 10x higher power handling than conventional transistor-based switches on the same platform 

Significant improvements in linearity 

Lower parasitics and higher switching frequencies overall 

Equally important, the team demonstrated the technology using a "zero-change" approach on an existing commercial process, showing a path toward manufacturability and integration with real-world systems.

The results, published this year in Nature Electronics, point toward potential applications in reconfigurable terahertz systems, next-generation wireless networks, satellite communications, advanced sensing, and the high-speed connectivity that increasingly underpins AI infrastructure.