By Shelly Stalnaker

I recently had the chance to attend an on-demand webinar introducing the new GLOBALFOUNDRIES DRC+ hotspot solution that incorporates machine learning in the Calibre toolsuite. Presented by Sriram Madhavan (GLOBALFOUNDRIES) and Michael White (Mentor, a Siemens Business), the webinar walks through the new DRC+ DFM solution, and explains how the addition of machine learning expands and improves both the capabilities and results.

Click here to read the full blog post, on the site of our ecosystem partner Mentor.

Through its University Partnership Program, GF benefits from the expertise of academic researchers while providing them with access to the technology to demonstrate innovative designs

by Gary Dagastine

It may seem premature to talk about 6G wireless communications just as 5G technology is beginning to be deployed in earnest, but the R&D community is already hard at work investigating the technologies needed to make 6G a practical and commercial reality later this decade.

GLOBALFOUNDRIES (GF) is taking steps to establish a leadership position in 6G by collaborating with top researchers at leading universities to leverage the unparalleled benefits of its FD-SOI, RF-SOI and SiGe platforms, which already have been proven in 5G and other wireless applications to deliver industry-leading performance and cost-effectiveness.

Wireless connectivity is a major focus for GF, along with artificial intelligence (AI), edge-to-cloud computing and automotive solutions, because the company’s strategy is to be a leading supplier of the differentiated, feature-rich technologies needed to help shape the digital transformation of our world.

6G is the sixth generation of wireless communications technology. It will be significantly faster than 5G, capable of transmitting huge amounts of data at speeds that may reach or exceed 100 gigabytes per second (Gb/s) with little or no effective latency, or lag time.

Gammel said there are countless other ways 6G technology will help change the world. “We’ve gotten familiar with the concept of telemedicine during the Covid-19 pandemic, and 6G could raise it to new heights. In robotic surgery, for instance, lifelike 3D images not only could guide a surgeon through an especially challenging operation, but the surgeon using the robot could be located thousands of miles away from the patient,” he said. “6G also will facilitate the use of AI everywhere in a network from the edge, to the customer, to the core, for greater efficiencies, speed and reduced costs.”

6G Technical Challenges & Opportunities

Technical standards for 6G systems haven’t been developed yet, but the initial 6G frequency range is likely to be from about 50 GHz to 200 GHz, with initial applications expected at the low end, near the top end of the 5G range. This is a blazingly fast and uncongested part of the RF spectrum, but these millimeter-wave (mmWave) frequencies have certain characteristics that make the technical development and economics of 6G systems challenging.

One major technical challenge is the need for power-efficient LNA amplifiers (a key component of wireless systems) to amplify low-power 6G signals without significantly degrading the signal-to-noise ratio, which is key to error-free performance.

Another is the need for accurate mmWave device simulations and models, along with hardware-validated process design kits, or PDKs, for successful, cost-efficient semiconductor design and production. This is a huge unmet need at high frequencies.

Perhaps the biggest issue, though, is that mmWaves suffer high propagation losses because they are absorbed by water vapor and oxygen molecules in the atmosphere, so finding ways to increase the over-the-air output power of transceivers is critical. Another propagation challenge is that mmWaves are easily blocked by walls, trees and other objects.

These propagation challenges mean that 6G networks will require many base stations and small cell sites located in fairly close proximity to one another to relay signals. Given the large number of semiconductors that will be needed for these dense networks, economic considerations will be critical.

Gammel said all of these challenges play to GF’s strengths, which include:

• 22FDX™ and 22FDX+ FD SOI solutions, which combine RF, analog, embedded memory, and advanced logic in one chip, with dynamic voltage scaling and unmatched design flexibility for peak performance and energy-efficiency. Customers use FDX for such tasks as integrating front-end module (FEM) elements like data converters, LNAs, power amplifiers (PAs) and switches with the transceiver.
• GF’s family of RF SOI solutions, used in integrated FEMs and beamformers in 5G base stations and smartphones.
• GFs family of silicon germanium (SiGe) BiCMOS solutions for Wi-Fi and mmWave FEM’s

“6G gives us a vision for solutions based on merged technologies in which GF already has undisputed leadership,” Gammel said. “What’s great about these proven, cost-effective solutions is that they are nowhere near the limit of their capabilities. Their performance can be extended in step with the wireless industry as it evolves, moving past the upper reaches of the 5G spectrum and extending into the 6G frequency range. Customers will not have to turn to new technologies and exotic materials to get the performance they need; they will be able to get it from well-understood, production-ready and cost-effective silicon-based technology.” 

Partnerships with Leading 6G Researchers

GF is actively promoting 6G circuit and system research via the company’s University Partnership Program, through which GF provides access to technology to select university teams who collaborate with GF’s R&D team and share their research results.

The program is large and impactful. “Worldwide, we work with over 35 universities in various areas of technology, including 6G,” said Bika Carter, Sr. Manager and Deputy Director of External R&D Management for GF. “Peer-reviewed published papers are a key measure of the quality of our academic partners, and their publishing output is large and growing. In 2019 and 2020, there have been over 200 publications from our professors across our technologies. We have active university programs in 22FDX, 45RFSOI, silicon germanium (SiGe) and silicon photonics technologies.”

Ned Cahoon, Senior Director in GF’s Mobile and Wireless Infrastructure CTO office, works closely with many of the professors affiliated with GF’s University Partnership Program who are working on 6G technology. “We are technology leaders in mmWave, so we look for professors and academic programs that are tackling what we see as key 6G circuit and system issues that can be addressed by GF’s differentiated technology, such as FE, or front-end, circuits, at frequencies above 100GHz,” he said. “The professors we work with are renowned in their field, with great track records, and we work with them as partners. They share their research with us, and we help further their work by giving them access to our silicon on multi-project wafers.”

No Doubts About Silicon

One of GF’s academic partners is Gabriel Rebeiz, Ph. D., Distinguished Professor and the Wireless Communications Industry Endowed Chair at the University of California San Diego. Professor Rebeiz is a Member of the National Academy of Engineering and an IEEE Fellow. He is a pioneer of integrated phased arrays for communications and defense systems, and was the first to introduce MEMS and micromachining to the RF/microwave field.

At UC-San Diego, his group has led the development of complex RFICs for phased-array applications. His phased-array work is now used by most companies developing complex communication and radar systems, and he has graduated some 100 Ph.D. students and post-doctoral fellows.

At present his students are working on a broad set of research projects ranging from wideband systems in 45RFSOI to 140GHz phased arrays. “Before 6G comes along, there’ll be 24 GHz, 28 GHz, 39 GHz and 46 GHz chips used in 5G communications, so we’re working on a lot of wideband chips using the same processes and techniques that we will extend to 6G devices,” he said. “These are high-risk, high-payoff types of projects, and we push the limits of technology. GLOBALFOUNDRIES has been a great partner as we do this. We cover a large technical area and they support our innovative work.”

Prof. Rebeiz said he is convinced silicon is the solution for the higher reaches of the 5G band and for 6G applications up to about 220 GHz. For example, using GF’s technologies in a forward path transmit module (FPTM), his team recently achieved 12dBm of output power at 140GHz with 11-12 percent efficiency. “That’s a phenomenally good figure for point-to-point communications, given that the current figure-of-merit is 6dBm, but also, we’re doing it at 140GHz!” he said.

“As the number of elements needed in dense 6G networks increases so much, the power needed per element must necessarily decrease if these systems are to be practical, and instead of the 20dBm per element we now have at 28 GHz, we might need only 3-6dBm,” Rebeiz said. “So, without any doubt power-efficient silicon technologies like 22FDX will be dominant above 100GHz for any array application.”

Many challenges remain, of course – Prof. Rebeiz said packaging and testing are critical needs:  “What are we going to do to make test affordable in the future? Nobody’s going to test up to 140GHz because it is so difficult and expensive, but we can’t make progress without it” – but this work represents a great opportunity for his students.

“My students and I are hardware people,” he said. “We like building stuff. What my students are doing is vitally important for the world, and as such, this is truly a golden age for them.”

by Lakshmi Jain

The innovation that lies ahead on the road to a trillion connected devices is nothing short of astonishing, as we push compute and artificial intelligence (AI) and machine learning (ML) capabilities from the cloud to the edge. To get there, engineers are rethinking how to push the boundaries of design around device size, reliability, and efficiency. 

Nowhere is this transformation more vital than in edge computing, where most edge devices are small, ultra-low power and cost-constrained …

Click here to read the full blog post, on the site of our ecosystem partner Arm.