Episode 16: Superior Signals Intelligence

Lisa Peña (LP): Coming up, superior signals. We're discussing an antenna technology recognized as a top 100 innovation, plus a Breakthrough in detecting chemicals with some help from machine learning.

And Ask Us Anything – we're tackling quantum entanglement. What exactly is it? Find out on this episode of Technology Today.

[MUSIC PLAYING]

We live with technology, science, engineering, and the results of innovative research every day. Now, let's understand it better. You're listening to the Technology Today Podcast, presented by Southwest Research Institute.

Hello, and welcome to Technology Today. I'm Lisa Peña. Breakthroughs and Ask Us Anything ahead.

But first, our guest is SwRI, electrical engineer Brandon Nance from our SIGINT solutions department, here to talk about the AF-369 Direction Finding Antenna. This technology recently won a prestigious and highly competitive R&D 100 Award, making it one of the top 100 innovations of 2019. Congratulations to your team and thank you for being here, Brandon.

Brandon Nance (BN): Thank you so much. It's good to be here.

LP: So let's start with a broad question first. What does the term SIGINT refer to?

BN: SIGINT is short for signals intelligence. And signals intelligence is just the practice of passively, that means without transmitting anything back or actively communicating, just passively picking up electronic signals that are being transmitted across the air already.

LP: OK. So I used the short name of the antenna, but it actually has a much longer name. So the name is the AF-369 VHF/UHF DF antenna. So what do all those terms mean?

BN: Right, right. It's acronym soup. So let's start at the end. Everybody knows what an antenna is. DF is short for direction finding, which is the practice of using an antenna array or some kind of sensor array to estimate the direction of arrival. So signals on the air, it's being broadcast. The direction finding antenna is there, and it's able to estimate what direction of arrival it came from.

And that's practical, because in SIGINT, you might want to know not just what somebody's saying when they transmit something. You might want to know where they are. And so with the direction finding antenna, or more than one direction finding antenna, you can actually figure that out. So let's see, that's DF. VHF/UHF, those are different bands. It just means different frequency bands or different wavelengths. VHF is short for very high frequency. UHF is short for ultra high frequency. And put together, those cover 30 MHz up to 3 GHz. So we're just saying that the antenna covers those entire bands.
 

Left to right: Lisa Peña and Brandon Nance

So then we're on to the model number. AF-369 is just the model number of the antenna. And so A and F just are Southwest Research Institute's model nomenclature. A is for antenna. F is for fixed site, as opposed to a mobile or a shipboard application or an airborne application. It's a fixed site, which means it's on a mast somewhere and holding still. The numbers 369 are between the range of values that we've internally decided describe stuff that covers VHF/UHF.

And so we had to pick an unused number. And kind of an inside joke, we couldn't figure out what the top end frequency coverage was going to be. Was it going to be 3 GHz, 6 GHz, or 9 GHz? We were going to try to push it as far as we could get. We ended up with three. And we actually are working on a frequency extension design right now to take it all the way up to nine.

LP: All right. Random number chosen there. That's neat.

BN: Also if anybody's a fan of Lil Jon & The East Side Boyz, shoutout to Get Low.

LP: That's right [LAUGHS] I like it. I like it. I'm liking this antenna even more now. Awesome. So this is definitely a cool antenna. We think about antennas helping us watch TV, for instance. But what are some of the practical uses for antennas in general?

BN: I mean, antennas make it possible for us to communicate electronically without wires, obviously. So all of the communications signals for emergency use, police dispatch, cell phones for God's sake, Wi-Fi, all of the stuff we depend on and just kind of take for granted these days, all made possible by antennas.
 

The AF-369 VHF/UHF DF Antenna, recognized by R&D Magazine as a top 100 innovation of 2019, utilizes a single-sleeved electric dipole that expands bandwidth by more than 80 percent compared to conventional electric dipole antennas.

LP: So let's talk about this antenna in particular. And it is bolstering signal intelligence. How does it do that?

BN: So it not only provides enough sensitivity to pick up signals in the first place so that signals across that huge span of bandwidth can be monitored so that further processing and information can be pulled out later, like we said, it provides direction finding capability and estimates direction of arrival of signal, so not just knowing what they're saying but maybe where they are.

LP: So how is the AF-369 antenna used in the real world, some real world applications?

BN: So what our customers do with these specifically is up to them. And we don't get in the middle of it, to be sure. But we envisioned, when we created this thing, that there might be a range of scenarios. Well, so it's conceivable you might want to listen to signals coming from out over the water to protect the port of entry or something like that, for situational awareness, law enforcement activities. You might want to know what people are saying out there in case they're trying to do something against the law or something that might harm the United States. If they transmit anything over the air that might give that away, these antennas can be used to help pick that up and give early warning to people for force protection and law enforcement reasons and those kinds of things.

LP: And this isn't just local law enforcement. This can be used on a national security level?

BN: Absolutely. Absolutely.

LP: And let's talk about the design of this technology. What makes it stand out? How is it different from previous designs?

BN: So antennas that cover this frequency range, this is not the first one to do that. We have a long history of providing antennas that cover VHF/UHF frequency bands. A lot of them are shipboard designs. We do have some fixed site designs kind of like this one, at least in terms of the applications, kind of like this one in the past. They have all had to use every antenna that covers more than about an octave of bandwidth, meaning the upper frequency limit divided by the lower frequency limit is more than a factor of two. They all have to use multiple antenna arrays.

So within the total assembly, there are multiple bands. You kinda have to break it up. No one set of antennas can do it all. So you have a different design covering different portions of the band of interest. And this one uses a new technique to solve an old problem, which is how can we cover all that stuff with as few antennas as possible to keep things cheap, to keep things simple, in terms of the overall system complexity.

So it uses something we came up with on this project, sleeved dipoles to help dipole antenna elements. Actually, it's the first kind of antenna ever invented, a dipole antenna. This puts a little sleeve around part of it to mask off some of the current, and I'm sorry that's getting a little bit technical. But the point is, the impact is, it keeps the antenna pattern focused toward horizon, where we need to be picking up signals.

Normal dipoles have an upper frequency limit for a given length where that no longer happens. And the sleeve corrects that and keeps things focused at horizon. So in the end, we can cover about an 80% bandwidth – 80% more bandwidth rather than a conventional dipole. And that lets us completely eliminate one of what used to be three or four bands to cover this broad frequency range. So we can do it with two bands instead of three.

LP: Why is bandwidth important?

BN: There's so much spectrum out there. And it's all chock full of signals. And there are more and more signals on the air every day, and the spectrum is getting more and more crowded. And so what is happening is everything used to be all below 30 MHz. And then communications increased and increased, and now it's up into VHF and UHF. And we've just all the time, in order to get more bandwidth in terms of communication's bandwidth, more information transmitted, things are going higher and higher in frequency. And so in order to effectively monitor all that spectrum for national interest, national security, just keeping everyone safe, you have to listen to it all. And so we have to cover those bands.

LP: And going back to arrays, you said having less does save money. But what are the other advantages to having two instead of three?

BN: Just from an overall system design, it's more elegant. So in addition to this antenna being up on a mast somewhere, there's a lot of electronics and computing equipment that goes downstairs that has to connect to it and talk to it. And fewer antennas make for fewer connection points. That means all the electronics downstairs can be even simpler and less complex, but more importantly, less costly.

LP: How did it all begin?

BN: So interestingly enough, we have these strategic plans that we put together for upper management and forces us to think reactively to what is this customer asking for that they want done next week or next year. It forces us to think a little bit down the road. We'd identified the fact that our SCOUT^™ product line, that's a survey monitoring system that our customers use for presumably these kinds of applications, it's been doing really well.

We have a VHF/UHF version of it that's been covering these frequencies and was starting to garner interest. But a lot of people were asking for direction finding. What can you do for a direction finding antenna? The system itself, the stuff that goes downstairs, like I mentioned, fully capable of it. But the only antennas we had to offer were either, let's just say they were over designed for what our customers with these terrestrial fixed sites needed. And so it was either a naval shipboard application and shock and vibe vibration standpoint. And the thing was heavily over designed and very costly compared to what these folks needed and compared to the cost of the rest the system.

Had another terrestrial design that was made to just be transportable and go back together really quickly. And again, that's a lot of design, a lot of cost. And so it didn't make sense to try to, here's a system for x number of dollars, and then here's an antenna that goes with it for two or three x number of dollars. We were never going to break into that market. So we got together and realized we had this need and sat down and thought, what can we do? Got together with Mr. Downing, our Executive Vice President, and pitched an idea for this internal research project to him. And lucky for us, he funded it.

LP: It worked out. So was there a moment where you really saw it coming together and really saw this technology take flight the way you envisioned?

BN: There's probably a series of those. But one that stands out is when - so we didn't have this sleeved dipole idea laying around forever. We kind of lucked into it in a way. And I could tell the story about that in a minute. But once we built that and verified, sure enough, this really does provide an 80% or more increase in bandwidth and comparing the measured results and looking to see, how sensitive is this thing going to be with this new dipole compared to what competitors are offering, we saw an order of magnitude improvement. And we realized this was pretty special.

LP: So want to tell us that story?

BN: Yeah. So I love collaboration. I love working with other people, because none of us, it seems like, ever come up with the best ideas on our own or in a vacuum. And so case in point, this problem of covering so much bandwidth and trying to do it with fewer antennas, again, is an old problem. And every time we start a new design project, everyone's asking for all the reasons we've mentioned, how can you do this with fewer antennas, how can you do this with fewer antennas?

So we had a new engineer, Thomas Christian, started in my section. And he had been around for maybe a year or so. And I was telling him about this problem, what can we do to make dipoles have higher bandwidth? Because we know we need to use dipoles at the low end, because we can't really make anything large enough. Well, there's a lot of reasons. But we always use dipoles for the low band. So it's really just a matter of playing games, trying to come up with more bandwidth.

And so he says, well, we could play around with the feed. He'd seen some people at another place he worked feed the dipoles in more than one place, which is unusual. Normally you just feed it right in the center, one place. And he says, well, I've seen people do this in multiple places. And I went away thinking, that's fascinating, I want to try this. So I did a quick electromagnetic model, did a CAD program we got, simulates everything. Fed it in two places. I ran an automatic optimization process on it, repositioned the feeds in different places and look for the optimal solution.

And sure enough, I saw a bandwidth increase. But it was like 10% maybe. Which, I'll take it, but that wasn't the full story. So I thought, well, sure, we'll do this, but I don't want to feed it in two places. That's complex. That's twice as many cables. I don't even know how to do that. So I thought, I'll feed it once in the middle, and I'll just have this sleeve around it that effectively, where the sleeve opens, each one of those will act like a feed point.

And I did that, and I was astonished to see it wasn't 80% yet, but it was 50% increase in bandwidth. And I was like, what is this? This is different. This is not the way he envisioned, when he described this to me, this working. And obviously, I didn't expect it either. And so just pulling at that thread and investigating, why is this working? And we realized we were really onto something.

LP: That was the big payoff moment.

BN: Yeah. That was a big payoff moment. And serendipity. I wouldn't have come up with that on my own, and he wouldn't have either. So it was working together and learning things. That's what makes the job fun.

LP: All right. So let's move on to the award, the R&D 100 Award. It's a huge honor, an annual award presented by R&D Magazine since 1963. It's become a prestigious recognition of innovative technology, really representing the very best of the year. SwRI has won 45 of these awards over the years. So how did you feel joining this group and winning the R&D award for 2019?

BN: I think the best word to describe it is surreal. I didn't know a whole lot about it. I know that our organization takes these very seriously, and they're a big deal. We've got a huge wall in our admin building showcasing all the awards. And I've walked by it a whole bunch of times and thought, wow, that's really neat, these other people in these other divisions are doing really great things, and that's really awesome.

I never dreamed that we would get a shot at it. Because most of what we do is maybe funded by a customer, and the customer doesn't necessarily want all the being spilled, or we want to hold onto this or that for various reasons. But given that we developed this fully on RI gave us a real chance at being a little more open about what's going on with this antenna. And we can market it a little more openly.

And so when my director suggested I write an application, I was shocked. I was like, we don't do this. No, this will never win. But took it seriously and had a lot of good help from our friendly Communications Department staff. And before I knew it, I was in San Francisco wearing a tuxedo and seeing a whole bunch of other prestigious companies and organizations, national labs that you hear about, sometimes run into on a regular basis. And it was just fascinating.

LP: And now you're seeing your own plaque with your own name and team up there.

BN: Yeah. I love it.

LP: Pretty nice.

BN: I love it.

LP: And we just want to clarify, you mentioned this was part of an IR program, which is our internal research funding program. So this technology was supported by SwRI internal research funds, correct?

BN: That's absolutely right.

LP: All right.

BN: And you mentioned the word team there. And that's absolutely critical to point out. Because while I got to go, wear the tuxedo, and my director got to wear the tuxedo, by the way, some folks were invited and didn't want to wear a tuxedo. But the point is we have a huge team, and this wouldn't have been possible if it were just me working on anything. I just told the story where I couldn't have come up with that idea on my own. We've got a huge team, all of which working together was really what made all this possible. And so just accepting the award on their behalf and, yea, it's an honor.

LP: Teamwork makes the dream work.

BN: Absolutely.

LP: Awesome recognition of that. So what do you enjoy about antenna technology? So much of it is behind the scenes. But what's fascinating about it to you?

BN: That's a great question. I fell in love with antenna design while I was in grad school. I'm not one of these, you hear these stories of engineers. How did you get started in engineering? Well, I liked to take things apart and figured out how they work at an early age. And I always knew that I want, no, that wasn't me. I kind of backed my way into things.

Electrical engineering, I liked the challenge. I liked a lot of aspects about it. By the time I got to college, I was really starting to be fascinated about what makes things work and so forth. But I didn't really have an end goal of where I wanted to be with my career. But I got a scholarship offered to me that was telecommunications related for grad school and went around and interviewed different professors in what I might work on. And one possibility was antennas. And the way the guy described it to me, my thesis advisor, was it sounded like the perfect mix of art and science. And I've got an artistic background, a little bit of music, a little bit of actual artwork, nothing professional by any means. But anyway, I just loved that.

LP: How is antenna work artistic?

BN: I think it's some of the 3D CAD aspects is what I'm thinking of. And also in the sense that it's art or science. People talk about I've got it down to a science, or it's an art the way you do this. A lot of people call it black magic, radio frequency stuff. And they tell us that they call it black magic. And there's a little bit of truth to that. Because there are a few people out there that really understand it all completely. A lot of us have to do a weird mix of having some theory and fundamentals in our background and just try the rest out in a model or an experiment.

LP: All right. So you told us one story already. But do you have any other favorite moments designing this technology or working with antennas?

BN: I really enjoyed just working with Don Mahoney from our Applied Physics Division. He's a former, well, my division is now called Defense and Intelligence Solutions. It used to be called a few other names. And he worked in our division back when it was named some other things. Anyway, so he had designed a lot of our antennas that had come before, especially direction finding antennas. And so he's just kind of the expert on campus when it comes to those things.

And getting together and collaborating with him, my group coming up with some of the electrical and radio frequency designs, from an electrical standpoint, not necessarily a mechanical one, what does the antenna have to look like and to do what it's supposed to do? And then getting with him to figure out, how can we meet those constraints and still build this thing and make it simple and cost effective? It was fascinating watching him work. He came up with some really amazing manufacturing techniques that I hadn't seen before. And I really enjoyed that. The final product, it's very elegant. It's simple to manufacture. It's easy to put together. It's easy to test. Just getting to see all that happen, watching him work, was another highlight.

LP: So your very own antenna mentor.

BN: Absolutely.

LP: All right. Well, I don't think antennas get enough attention. So thank you for shedding light on this area of technology. They really are key for some advanced and important applications. So it's interesting to learn about them and get your insight today. So thanks for joining us, Brandon.

BN: Absolutely. Thank you so much.

So nice to have you here. So SwRI picked up two R&D 100 Awards in 2019. Next month we'll highlight our second award-winning technology, the Superhydrophobic LotusFlo^™ coding process. And now Breakthroughs, personal stories of discovery told by the people who lived them. Today, a scientist drowning in data turns to artificial intelligence.

[MUSIC PLAYING]

Breakthroughs

And now Breakthroughs, personal stories of discovery told by the people who live them. Today, a scientist drowning in data turns to artificial intelligence.

Kristin Favela (KF): Hi, I'm Kristin Favela, principal scientist at Southwest Research Institute.

Michael Hartnett (MH): And I'm Michael Hartnett. And I'm a computer scientist at Southwest Research Institute.

KF: So I'm a mass spectrometrist by training. So I know that's a very technical sounding term. But basically, a mass spectrometer is just a big fancy instrument that identifies chemicals. And over the past 10 years, the instruments have gotten much, much better. The mass spectrometers have gotten much more able to identify many, many, many chemicals in all kinds of different samples. So we've looked at soils, we've looked at consumer care products such as lotions and lipsticks, building materials, all kinds of products that you touch, breathe in just every day.

We're looking at all this data, because we have very little understanding of what is in all of these things, these products that we come into contact with every day. And we're getting a wealth of information from these instruments, but almost too much. So I spent many, many days just sorting through thousands and thousands and thousands of chemicals, reviewing every signal coming off of the instrument manually. And it got to be too much. I would get headaches and pains in my neck from sitting at the computer and looking at all this data.

What we really needed was an artificial intelligence solution that would allow us to look at the data directly and still keep the expert chemist to make the difficult decisions but use a computer to do the back breaking, tedious work for us. So we have a fabulous machine learning division here at Southwest Research Institute. And that's how I met Michael and his team.

MH: So it was really exciting to get involved with Kristin and her team over in the chemistry world. And so basically, machine learning is best for exactly the type of application that Kristin is talking about, where you have a wealth of data but you have to sift through it in a manual process. So we leveraged machine learning, in this instance, to produce a signal quality score over those hundreds or thousands of signals in a sample that Kristin had to deal with manually before. And we're looking at similar information to what Kristin was already doing in her head. And we're basically trying to simulate an analytical chemist using artificial intelligence.

So of course, we can't do all of her job, because she's trained, and she's very intelligent, and it's hard to do that with machine learning. But we are able to do the more tedious, mundane work. So we've produced a system now that can perform that manual review process with a pretty good accuracy. We're not identifying compounds. But what we are doing is filtering out the bad signals, so things that aren't actually in there, they're just artifacts of the instrument itself.

KF: It feels great, because this is not just a problem that we're having. This is a problem that every group that relies on mass spectrometry to do their research is having. Everybody that does, as we call it, non-targeted analysis or trying to find all the chemicals in a particular sample, everybody is having this problem. So this is going to really push a lot of different scientific fields forward, not just environmental chemistry.

MH: I agree with Kristin that a lot of different fields could use this technology. And it feels really good to have that large of an impact. I think it will promote health in not just environmental but also food and agriculture forensics. It will improve safety overall for the population. It's a cool feeling to contribute to that.

[MUSIC PLAYING]

And the machine learning tool they created together is now known as a Floodlight^™. Thanks for sharing your breakthrough story, Kristin and Michael.

Ask Us Anything

And finally today, Ask Us Anything, Andrew A. asked on Instagram, how does quantum entanglement work?

Our expert SwRI scientist Dr. Jerome Helffrich has the answer. Thanks for joining us, Jerome. So what is quantum entanglement? And how does it work?

Dr. Jerome Helffrich: Quantum entanglement implies that there are two or more particles that have some kind of shared properties in a quantum state so that a measurement or an observation that you do on one of them reveals features about the other members of the entangled set. Well, it works by preparing the particles carefully. They have to be isolated from the outside world so that no other collisions with other particles can disturb their entanglement.

This usually involves cooling them to very low temperatures. They can be put into a quantum computer to make computations that are difficult for ordinary computers to make. Or we can make experiments on them, which indicate that you can get action at a distance. If you measure the properties of a particle here next to you, it actually affects the properties of a particle miles away.

There are some advanced computing facilities that are trying to use this. IBM has several quantum computers. Google reportedly has one or two. There are other maybe 10 quantum computers around the world now that have done primitive computations. But the field is really just in its infancy. There was a famous paper published by Albert Einstein in 1935 that indicated that quantum mechanics allows for this so-called action at a distance, where a measurement on one particle then entangled pair would allow you to infer the properties of another one that you had never seen before, that could be light years away.

And this seems to violate our notions of causality and that things that we can measure here must take some finite time to effect other objects. This idea of entanglement implies that you can do massively parallel computations so that you can perform computations that would be impossible for an ordinary computer. And it also allows for the possibility of detecting disturbance of publicly distributed cryptography keys.

In other words, if somebody makes an observation on your key, and without you knowing it, you can make a test of it, and the person who is supposed to receive the key would know that it had been tampered with. And so it allows for more secure delivery of cryptography codes. It's fun to read about. You can read the original paper published in 1935, it's very readable, by Einstein, Podolsky, and Rosen. And there are numerous recent papers or articles on it. One was in The New York Times not long ago.

Want to know more? Dr. Helffrich recommends the book Quantum by Manjit Kumar. To submit a question, use #askSwRI, comment on one of our Ask Us Anything posts, or visit podcast.swri.org and scroll to the bottom. Your question may be featured on an upcoming podcast episode.

And that wraps up this episode of Technology Today.

Subscribe to the Technology Today Podcast to hear in-depth conversations with the people changing our world and beyond through science, engineering, research and technology.

Connect with Southwest Research Institute on Facebook, Instagram, Twitter, and LinkedIn. Check out the Technology Today Magazine at technologytoday.swri.org. And now is a great time to become an SwRI problem solver. Visit our career page at swri.jobs.

Ian McKinney and Bryan Ortiz are the podcast audio engineers and editors. I am producer and host, Lisa Peña.

Thanks for listening.

[MUSIC PLAYING]