Academic Collaborations Strengthen, Hasten GF’s Path to 6G Leadership
by Gary Dagastine
In an earlier era, the way forward for semiconductor companies was to spend fortunes and untold hours of labor on finding ways to scale semiconductor devices to smaller sizes, because scaling produced dramatically greater performance that opened up many new applications.
But the advances the industry has made in recent decades have brought about many technology platforms that are already quite powerful, and new features and capabilities can be added to them cost-effectively to adapt them to new requirements. As the world’s leading specialty semiconductor manufacturer with an ever-expanding offering of differentiated solutions to evolving applications, GlobalFoundries (GF) exemplifies this approach to technology development.
Nowhere is this more evident than in the company’s push to be a leader in 6G wireless communications technology. GF offers a number of platforms and solutions that have not only been proven in the most demanding communications applications, but whose full potential remains untapped. They include GF’s 22FDX™ platform and 22FDX+ solutions, and GF’s families of RF SOI and SiGe (silicon germanium) solutions.
They represent a compelling pathway to 6G, the next generation of wireless communications technologies, anticipated to make its debut toward the end of this decade.
In our recent blog on GF’s push for leadership in 6G technology, we described the company’s University Partnership Program. Through it, GF provides access to technology to more than 35 university teams, who work collaboratively with GF’s R&D staff in various areas such as 6G. They share their research results that support the addition of new features and capabilities to GF’s platforms, publish research results at technical and academic conferences, discover new application possibilities, and introduce students to these technologies, who will then have familiarity with them throughout their careers.
We profiled one of our research partners, Gabriel Rebeiz, Ph.D., Distinguished Professor at the University of California San Diego. A pioneer in integrated phased arrays for communications and defense systems, he directs a broad set of research projects ranging from wideband systems in 45RFSOI to 140GHz phased arrays.
In this blog, we’ll highlight three more high-profile university partners, who will tell us about their research, how they use GF’s technologies, and what working with GF has meant for them and their students. Their diverse experiences and research interests will illustrate GF’s commitment to 6G and show how the company’s strategies and technologies are helping to make advances in 6G possible, better, faster and more cost-effectively than otherwise:
- Einstein Prof. Friedel Gerfers, Ph.D., is the Chair of Mixed-Signal Circuit Design at Technische Universität Berlin (TU Berlin), where he has been awarded an Einstein Fellowship to pursue his work in the area of mixed-signal circuit design for 5G/6G communication systems. The Einstein Foundation Berlin, founded by the regional government to sponsor cutting-edge science and research, funds his cost- and staff-intensive high-frequency laboratory, one of only two laboratories in Germany with the capability to fully test and characterize electrical and optical systems up to D-Band (up to 170GHz).
- Prof. Aarno Parssinen, Ph.D., is in the Center for Wireless Communications (CWC) at the University of Oulu in Finland. Prof. Parssinen is a major figure in the university’s 6G Flagship initiative and is a key collaborator with Finland-based Nokia, one of the world’s leading telecommunications and networking companies.
- Prof. Hua Wang, Ph.D., is the Director of the Center of Circuits and Systems (CCS) at the Georgia Institute of Technology. His team has been collaborating closely with many semiconductor companies. Prof. Wang is the recipient of several highly prestigious academic awards, including the DARPA Director’s Fellowship. He is recognized for his contributions to wideband energy-efficient RF/mmWave circuits, novel transceiver array architectures, and antenna-electronics co-designs, which have had a major impact on the direction of the industry’s R&D activities.
Turning Ideas Into Silicon
Prof. Gerfers at TU Berlin directs currently 15 Ph.D. students on research into 5G/6G architectures and solutions, mainly using the 22FDX platform, and also pursues research into high-speed communications technologies such as automotive Ethernet and optical communications. One current project is research into the world’s first monolithically integrated 6G transceiver, the goal being to determine what other technologies can be integrated with 22FDX technology to achieve 10-15GHz bandwidths all the way from the antenna to the digital bit.
He has a background in corporate research at Philips, Intel, Inphi and Apple, and in two startups, Alvand Technologies and Aquantia. He joined TU Berlin in 2015 and began to work with GF in the same year. The successful collaborations with GF have grown in number and intensity since then, he said.
Prof. Gerfers said that energy efficiency and robust performance in high-frequency applications are critical unmet needs for future 6G wireless communications systems. “Not only in Germany but also globally, the increasing sense is that microelectronics is becoming a bottleneck in many socially and technologically relevant areas. That’s why the program GlobalFoundries has set up is so important. It is a key enabler for continued progress in microelectronics, in that it gives us access to a cutting-edge technology we need to turn our ideas and innovations into silicon,” he said.
“Also, the exceptional performance and features of 22FDX technology addresses a wide application space. We use it to build high-bandwidth transceivers, and there are few planar transistors to achieve the target power efficiency while maintain the stringent noise and phase noise budget. We believe 22FDX technology has the potential for use well beyond 250GHz, and is therewith optimal for our technologies and applications we want to address. In addition, German carmakers, for example, are using FDX-based ICs not only because they are power-efficient but also because the technology is robust enough to handle the stringent automotive requirements.”
Prof. Gerfers also noted that compared with FinFETs, FDX-based circuits are simpler and easier for students to design and lay out, and that without the University Partnership Program, “We’d be stuck using planar transistors at 28nm or even older nodes, and it would be practically impossible to prove the power efficiency of our circuit solutions.”
Bridging The Gap Between Academia And Industry
Finland boasts a long and distinguished history in the development of mobile communications. It’s where GSM (the 2G standard) was developed and first deployed in 1991; where the industry powerhouse Nokia is based; and where the 6G Flagship initiative, one of the world’s first and largest 6G research programs, was launched in 2018.
Prof. Parssinen at the University of Oulu said academic engagement with industry is necessary for continued progress. “In Finland, we in academia helped drive this rising industry right from the beginning. Our role has always been to bridge the gap between classical academic studies and corporate product development activities, and we have been at the very heart of things with 2G, 3G, then 4G and we’re helping the world implement 5G,” he said.
“Now, we have built a substantial program here at Oulu around 6G technology. It’s true that 5G is only beginning to be launched and there are many challenges that go with it which we are working on. However, the time to look forward is now because however long it takes to do fundamental studies, it takes even longer to bring that knowledge to the point where industry can use it effectively,” he said. “The university’s task is to look forward, to try to do things that we don’t yet know how to do. That is the very purpose of science.”
Prof. Parssinen’s expertise is in cellular circuits and transceivers up to 5G and beyond. In graduate school he led the team which produced the first 3G circuit on a single die, and later he was one of the contributors to the Bluetooth LE (low energy) standard. He was with the Nokia Research Center for 10 years and was on the company’s CEO Technology Council, and also served in key technology development roles at Renesas and Broadcom.
At Oulu he leads a team of about 20, split between Ph.D. students and other professors who seek to understand future application requirements and develop circuits and systems to address them. Their research encompasses phased arrays, antenna design, efficient beamforming and other relevant topics. The team is also developing a sophisticated laboratory to measure the over-the-air performance of radio systems.
“We work with GlobalFoundries’ 22FDX and 45RFSOI technologies, and the differences between the two are intriguing. The integration capability of 22FDX technology is superior and we are still learning about its capabilities, which means that for our students who are doing IC design, the opportunity to work with it is a unique advantage,” he said. “We have used a lot of 45RFSOI technology over the years, and because we have more experience with it, we use it for our larger chips.”
“The access to silicon and the freedom to explore innovative ideas are major advantages of our relationship with GlobalFoundries, and it is wonderful to be able to share our work through the program with other professors who are also on the leading edge,” he said. “We’re competing with them, in a sense, but our work also benefits from their knowledge and the interactions which take place.”
A Dream System for Future Wireless
The Center of Circuits and Systems (CCS) at Georgia Tech is at the core of the university’s efforts to develop high-frequency electronics for communications, radar and healthcare applications, said its Director and GF partner Prof. Hua Wang, Ph.D. The Georgia Tech CCS center hosts eight core faculty members, more than 90 Ph.D. students, and 12 post-doctoral scholars.
Prof. Wang has been with Georgia Tech since 2012, and prior to academia worked at Intel and Skyworks Solutions, where he led the development of novel solutions for mm-Wave circuits and systems as well as low-cost cellular front-end modules (FEMs).
Among the key focuses of the GT CCS center are the exploration of RF/mmWave/THz integrated circuits and systems for beyond-5G and 6G communication and sensing. Other wireless related research topics include antenna/electronics co-design; power amplifiers; and artificial intelligence (AI)-assisted adaptive RF/mmWave circuits and MIMO systems. The CCS center also has a wide research portfolio in high-performance computing, cryogenic electronics, bioelectronics and biosensors, physical-layer security, and AI-based design automation.
“We are focusing extensively on innovating wireless circuit/system designs; in particular, pushing the output power, bandwidth, and the reconfiguration of RF/mm-Wave FEM electronics,” he said. “All of these are really important for 5G and beyond because the higher the frequency, the higher the signal path loss, and to overcome it, each circuit element just needs to be more powerful. Also, because more of these mm-Wave systems are going to arrays, heat management becomes difficult, and energy efficiency has never been so critical.”
“For future wireless communication, a major priority for us is to find optimum ways to transmit and receive information at a massive data-rate with high linearity, given the increasing use of complex spectrally efficient modulations,” he said.
Prof. Wang said there is a growing need for antenna/circuit co-design as well, because at higher frequencies wavelengths are shorter and they’ve now reached the point where they are at the same dimensions as planar electronic circuits. “This opens up intriguing possibilities for power combining, filtering, noise cancellation, and even STAR communications directly at the antenna domain, but everything must be considered holistically,” he said. “For example, now we really have the opportunities to re-architecture the wireless frontend systems and consider how distributed electronics and radiation structures can together modulate, transmit, and receive complex electromagnetic signals. And we can explore their use for communication, imaging, and sensing. But many innovations here must rely on co-designs with different domain knowledge at different levels of abstraction. Co-design is important at the package level, too – to put the antenna on the package, or on the chip, that is the question.”
Prof. Wang said GF’s technology platforms offer many advantages in this regard. “45RFSOI platform has high resistivity in the substrate, and you can use it to make highly efficient mmWave frontend circuits and antennas. For 5G, we’ve been using it and 22FDX platform because they are perfect for high mmWave applications. Furthermore, we see prospects for SiGe devices based on various studies which indicate that their Fmax – a measure of transistor speed – can be pushed to 700GHz and above, and the technology lends itself to high yields and efficient manufacturing.”
Prof. Wang cautions that speed is only one critical requirement for future high-frequency wireless systems. The ability to handle signal complexity in unknown or dynamic environment with low latency is just as important.
“My dream system would integrate advanced SiGe devices with high-performance CMOS technologies to help me achieve the required configuration for the next-generation wireless electronics. Fortunately, both of these are available to me and my students, thanks to GlobalFoundries.”