How Qualcomm inventor Dr. Kiran Mukkavilli helped hash out the fundamental radio waveform of 4G and 5G

We take the high speeds of modern cell connections for granted these days – but behind the speeds are countless innovations and endless hours of research and development. Dr. Kiran Mukkavilli, senior director of engineering at Qualcomm Technologies, Inc., knows first-hand the effort involved. He came up with some of the fundamental ideas in the orthogonal frequency-division multiplexing (OFDM) radio waveform at the heart of 4G and 5G – one of the most important areas of modern wireless technology – to enable the high-speed wireless data we all love and rely on today.   

At Qualcomm for nearly two decades, Mukkavilli worked on various aspects of 3G, 4G, 5G, and MediaFLO – Qualcomm’s first mobile broadcast technology – as well as terrestrial positioning technologies. He leads all positioning efforts across Qualcomm Technologies and is currently working on 5G Advanced while building the foundations of 6G.

We recently caught up with Dr. Mukkavilli to discuss his work on OFDM, MediaFLO, and what excites him most about 6G.

The following interview has been edited for clarity and length.

We hear the term “OFDM” in the cellular world often - can you explain what exactly OFDM is?

OFDM stands for “orthogonal frequency-division multiplexing.” The basic idea is that, through clever physics and precise timing, you can overlap many different sub-carriers or radio waves in the same communication link at the same time – separating them through frequencies that don’t interfere with one another. This was a revolutionary idea that allowed for a massive expansion of capacity in cell networks from the 3G days. OFDM became the basis of high-speed communications including 4G and 5G networks, with cell towers and phone modems using OFDM-modulated radio waves to communicate with one another.

Can you share details about your early work on OFDM and how it's impacted cellular communications today?

This project was at the beginning of my career at Qualcomm when the entire world was looking at the transition from 3G to 4G. There was an early project to investigate OFDM and how it could be used for different applications. We were looking into how we could have mobile broadcasts on cell phones, where you could broadcast something to every user in a network. We knew OFDM was the way to go, but there were a lot of questions to answer and a lot of problems to solve.

In this project, we developed a systematic methodology and provided solutions to several fundamental challenges of using OFDM in a mobile wireless system. For example, I started with designing pilot signals for figuring out the wireless link characteristics, including the number of paths the signal takes and the strength of each of the paths, which is referred to as channel estimation. A good, up-to-date channel estimate is essential to supporting high data rates on wireless channels, and pilot signals are used to estimate that wireless channel, which varies with time.

I came up with a pilot design based on periodic placement and staggering of pilot symbols both in frequency and over time, which became a fundamental structure in OFDM-based cellular systems in 4G and 5G. We then further expanded this pilot design in 5G to dynamically adapt the pilot pattern depending on a user’s environment and mobility conditions – for example, whether a user is walking slowly on the sidewalk or riding in a high-speed vehicle like a car or a train, etc. This wasn’t an easy task but, as always happens at Qualcomm, we worked as a team and we figured it out.

One of your early projects at Qualcomm was MediaFLO. Can you tell us about the project and what skills you learned that have been helpful later in your career?

I couldn’t have had a better first project to start my career. MediaFLO was a mobile broadcast technology fully designed and deployed in the field by Qualcomm. Every six months or so, I was putting on a different hat as the project went through different stages: design, prototyping, testing, speaking to vendors, compliance. It was also the first time I got involved in terrestrial positioning.

The basic idea was that we could also use the OFDM broadcast signals as a natural way to enhance GPS-based positioning. An immediate application of improved positioning would be for an Uber driver to reach you in downtown, where GPS would have difficulty due to tall buildings. There were many challenges to overcome to make this work – interference, timing misalignment, multipath signal degradation, etc. Eventually we found a solution that considered all these constraints, and it became a precursor for positioning in 4G and 5G. This project really taught me all the different facets of launching a major project like this and how to adapt to the various challenges presented. 

With 5G Release 18 and looking ahead to 6G, what use case(s) are you most excited about?

With 5G Release 18, this is the phase where we start to put some of the building blocks for 6G together, including AI/ML for wireless, sub-band full-duplex and passive IoT, so that’s exciting. We’re also looking into intelligent surfaces, where we’re able to modify the environment around us and make it conducive to what we need. We’ll be able to say, “the transmitter is here, and the user/receiver is there — what should the environment look like so the link between the two is the best it can be?” That’s going to be a fascinating design paradigm for 6G.

What advice would you give students or other up-and-coming inventors?

One of my professors used to tell me about the importance of spending time and effort early on in a project so that you fully understand the problem and know exactly what you’re solving. This is even more important in our present-day work environment; it's so easy to dilute our focus with the many problems we encounter each day that need to be solved, but pick one problem at a time, let it eat at you, and give it your all to solve it.