Executive Interview: Edgar Garay, CEO, Sanghoon Lee, CTO, and James Kaney, VP of Business Development, of Falcomm
1. What's the origin story of Falcomm, and what does the name represent?
Edgar Garay (EG): Falcomm started as a way of commercializing IP that we created as Ph.D. students at Georgia Tech, more specifically our patented Dual-Drive™ power amplifier (PA), which achieves the highest energy efficiency of any PA in its class. Falcomm is a play on words between falcon and communication; the peregrine falcon is one of the fastest birds on the planet, and we feel that that quality represents our technology and our ability to execute as a start-up.
Today, we are on a mission to help connect the world using energy-efficient electronics. As Ph.D. students, we spent many years performing research in some of the best academic labs in the world and working on different ways of improving key building blocks in RF/mmWave integrated circuits, such as PAs, low noise amplifiers, beamforming ICs and on-chip antennas, just to name a few. Georgia Tech has a long-established relationship with GlobalFoundries, and their support and collaboration has allowed us to really push the technology development that will support the deployment of next-generation wireless infrastructure.
When we were finishing our doctoral studies, we realized that there were a lot of parallels between innovation in the car manufacturing industry in the early 2000s and innovation in the semiconductor world. It took a newcomer to start the electric vehicle revolution, push car innovation forward and achieve what the incumbents could not accomplish. In the same way, the semiconductor industry has become very antiquated, and most semiconductor products are now commodities with only a few of the big players monopolizing the supply chain. Therefore, the big semiconductor companies are not often too interested in technology innovation. Let’s take the PA as an example: commercial PAs were developed in 1930 at Bell Labs and, since then, similar variations of that original PA architecture have been employed in commercial systems with little innovation.
At the same time, we saw massive amounts of IP being created at top academic institutions like Georgia Tech without ever making it to a commercial product. Once we understood all these elements as a huge opportunity to bring extremely innovative semiconductor technology to the market, we got to work. The first step for us was to apply to the UC Berkeley Skydeck accelerator program, which is one of the top accelerator programs in the world. Due to the rich history of connection between UC Berkeley and the semiconductor industry, Skydeck has a chip-track, so every year they bring two semiconductor start-ups through their program, and we were lucky enough to be one of them. Besides financial support, they also provide us with all the Cadence tools for free, which is a huge advantage for an early-stage semiconductor start-up.
2. What is your ongoing relationship with Georgia Tech?
EG: Georgia Tech is one of the top academic institutions in the world, and Atlanta is becoming a hot spot for tech companies. Recently, we have seen many of the big tech companies make a move to the Atlanta area. Moreover, Georgia Tech is pushing really hard to become leaders in commercialization, bringing in people like Professor Raghupathy Sivakumar, the new vice president of commercialization and the chief commercialization officer. He is also an entrepreneur with multiple successful start-ups under his belt, so he understands what it takes to make a successful start-up.
Georgia Tech’s Office of Technology Licensing has also been very supportive, creating a great licensing program for start-ups to help them find a successful path to market. In addition, the Georgia Electronic Design Center has been essential to our mission, since they provide access to test equipment and world-class testing facilities that otherwise would have been cost prohibitive for start-ups. Moreover, we have had people from Georgia Tech’s VentureLab in our corner from day one, like Jonathan Goldman, whose many years of experience in start-ups and technology innovation guided us through some difficult challenges.
3. Your vision is developing the most energy-efficient mmWave PAs. How do you aim to do that?
EG: The first step for us was creating a technological advantage by reinventing the PA. We completely changed the way transistors are driven in PAs and created a new kind of PA that we called the Dual-Drive PA. In doing so, we were able to demonstrate our technology and the incredible advantages that our Dual-Drive PA offers in energy efficiency and power density.
Once we established our massive technology differentiation as compared to commercial products, our objectives changed from technology demonstration to hacking the supply chain. While most B2B start-ups encounter tremendous challenges to attract customers without a track record, semiconductor start-ups face exponentially more difficult challenges since there are numerous factors that customers look for in their vendors other than great technology, such as operational experience, financial stability and experienced management with domain experience. If a customer designs their product with our Dual-Drive PAs in mind, they want to make sure that we can meet their design cycle and can provide our PAs for the lifetime of their product.
After speaking to many companies, we figured out that satellite manufacturers were our ideal first customer and that they were willing to take the risk. Satellite manufacturers are constantly aiming at reducing the weight of their satellites since that decreases launch costs and operational expenses. Based on our conversations, we have seen that some of our customers can reduce their launch costs by hundreds of thousands of dollars per spacecraft by implementing our Dual-Drive PAs in their communication modules. Currently, we have found great partnerships with some satellite manufacturers and, with their help, we will achieve our first product wins.
Once we achieve this goal of putting our Dual-Drive PAs in orbit, we will tackle the IoT/wearable and cellular infrastructure market, since they will also benefit tremendously from our ultra-efficient Dual-Drive PAs. We envision that in a few years we will have a product offering that will not only include our ultra-efficient Dual-Drive PAs but also beamforming ICs, better positioning us to start conversations with tier-1 phone manufacturers.
4. What proof points have you demonstrated?
Sanghoon Lee (SL): Our first generation 30 GHz Dual-Drive PA using a 45 nm SOI process achieved 50 percent power-added efficiency (PAE) for our two-stage proof-of-concept, which is the highest efficiency ever achieved for a two-stage PA at 30 GHz on CMOS.
Today, we are in our third generation, and we are achieving drain efficiencies close to 63 percent and a PAE of 56 percent for our one-stage Dual-Drive PA and 53 percent for our two-stage dual-drive implementation, all fabricated in the 45 nm SOI process. Our newest generation 30 GHz PA achieves excellent modulation results with 34 percent average PAE at 64-QAM, -25 dB EVM and 14.1 dBm average output power, which is a higher average efficiency than a Doherty PA can offer but at half the Si area.
These results demonstrate that our technology will become the industry standard for PAs and, based on our current design efforts, we are pretty confident that we can push our technology further.
5. You said your results have been demonstrated using a 45 nm SOI process. Can your circuit design techniques be applied to other processes such as GaAs and GaN?
SL: Part of our efforts are transferring our design and knowledge expertise to compound semiconductor devices for a high power Dual-Drive PA version, particularly GaN. GaN offers exceptional power density and efficiency, making it an ideal candidate for future space applications and cellular infrastructure. Additionally, there are other high power and voltage semiconductor processes that we are currently exploring, such as GaAs and LDMOS/EDMOS.
GaN and GaAs technologies are inherently more rad-hard than their silicon counterparts, which makes them ideal choices for space applications, where ionizing radiation can severely degrade the performance and reliability of on-board electronics. The great thing about our technology is that it is process/device agnostic, allowing us to service different market segments. Currently, the bottleneck for us is manpower; we have seen a lot of customer traction, so our design team is working around the clock to satisfy the demand for our Dual-Drive PAs. This is a great problem to have for a semiconductor start-up, so we are currently hiring to bring more talented PA RF/mmWave IC designers to help us grow faster.
6. What is your approach to convert your technology into products and engage the market?
James Kaney (JK): This is a question that we ask ourselves every day, and it drives our obsession with figuring out clever ways to enter the market and bring value to our customers. Part of Falcomm’s mission is to make sure that in a few years our Dual-Drive PAs are the industry standard. Right now, we are engaging customers through non-recurring engineering to develop custom solutions for their systems. Through these engagements we are maturing our technology and creating a broader product offering that we can then push through conventional supply channels.
While these are very challenging tasks and we most likely have very tough years ahead, we have an incredibly hard-working and forward-thinking team that is working around the clock to bring Falcomm’s technology to the market. Customers interested in seeing what our technology can do for them can go to our website (www.myfalcomm.com) and request information and a custom PA quote.
7. You outlined your market roadmap, initially focusing on satellites. Talk more about the other “sockets” or markets you see as good fits for your capabilities.
JK: Though we are working hard to bring the most value to the satellite market, we are also starting to engage with customers from other market segments, such as IoT, wearables and cellular infrastructure.
Energy efficiency is something that will benefit not only battery powered devices but the entire network infrastructure, since reducing the electricity bill, shrinking our carbon footprint and increasing the performance of RF/mmWave components will facilitate the deployment of future generation networks. Commercial PAs are very power hungry, and in recent years the improvement in efficiency has plateaued, reaching the theoretical efficiency limit. This creates a bottleneck in modern wireless networks, and its impact spans from handheld devices to the cellular network infrastructure, especially with the demand for data and data rates growing exponentially.
We have seen, for example, how companies like Nokia are adding water cooling to their base station products as part of their commitment to sustainability. We believe that the answer to increase sustainability in the wireless network infrastructure should start by creating more energy-efficient electronics, which is where we are focusing our efforts and making energy efficiency and sustainability the main part of our mission.
8. How are you funded and how far will current funding take you?
EG: In the past few years there has been renewed interest from venture capital funds in semiconductor innovation and start-ups. In a way, Silicon Valley is going back to its roots and relooking at silicon as the next frontier of innovation. We are lucky enough to have found great partners and investors; we are backed by The Skydeck Fund, Helium-3 Ventures, Little Green Bamboo Capital and Asymmetry Ventures. We have very ambitious goals, but our investors have provided us with enough capital and a great network to achieve our milestones in a two-year timeframe, and we are very determined to do so.
9. What are your near-term staffing plans?
EG: Being a start-up operator and leader means we are constantly looking out for great talent. We are very fortunate to have our offices within walking distance from Georgia Tech, which has a great talent pool. Anyone that is interested in joining our RF/mmWave PA design team can go to our website and apply for the current openings that we have.
10. Tell us about your own backgrounds and what inspired you to start Falcomm.
EG: I have spent most of my career in research and academia, with a brief stint as a design engineer at a small engineering company in South Florida, and then as an RF/mmWave IC designer at Intel Labs in Oregon. I started my career in research at the University of Florida, working in the cleanroom trying to develop a clever way of microfabricating flexible electronics. After that I worked as a research associate at the U.S. Army Research Lab and then finally moved to Atlanta to finish my Ph.D. My Ph.D. research focused on zero-power RF/mmWave signal processing, multi-functional digital transmitter systems for next-generation radar and 5G wireless communication and power generation for THz IoT devices.
SL: I had started off with RF antenna research during my undergraduate career at the University of Massachusetts, Amherst, studying ultra-wideband RF antenna arrays. After multiple internships with Systems and Technology Research investigating RF transceiver software-defined radios in heavily forested environments, I became interested in back-end transceiver electronics design. Thus, I joined the Georgia Institute of Technology, where my M.S./Ph.D. research focuses on silicon-based antenna and circuit co-integration for emerging RF/mmWave applications. I also participated in an internship at the MIT Lincoln Laboratory, working on compound semiconductor electronics for full-duplexing radios. The broad variety of experiences in both antennas and circuits helped me realize the bottleneck experienced in the transmitter system due to the PA, thereby prefacing Falcomm’s mission statement to provide the most energy-efficient electronics.
JK: I have a background in business, engineering and physics from Georgia Tech. I have also worked in the oil industry in North Dakota and in sales in the roofing industry in Colorado. I met Edgar and Sanghoon through the Georgia Tech network and quickly realized the incredible value that Falcomm’s technology will bring to the market, so I quickly decided to join Falcomm. My favorite hobby is racing cars at endurance races around the Southeast!