Ken Miller (00:11):
Welcome to from the Crow's Nest and our special Chief Technology Officer or CTO series, powered by L three Harris. I'm your host, Ken Miller, director of Advocacy and Outreach for the Association of Old Crows. The CTO series is a deep dive into key technology areas that are driving innovation and development of cutting edge capabilities in EMSO. In this first season, we are looking at next generation EW systems to counter evolving threats with speed, agility, and resilience across the electromagnetic spectrum. In this episode, I am joined once again by Paul Delia, director of Electronic Warfare Strategy and Development at L three, and we're going to sit down and talk about the technology required to combat the threat. Now, as our listeners will know in our last episode, episode two, I had the pleasure of sitting down with Tripp Raymond at a FA to talk about the threat and the evolution of the threat, past, present, and future, and what that means for maintaining our advantage in the electromagnetic spectrum. So today in this episode, we are going to take a deeper dive into the technology that we are working on to respond to this threat. So without further delay, I would like to welcome Paul Delia to our from the Crows Nest CTO series, episode three. Paul, thanks for joining me here. I'm from the Crows Nest. It's great to have you on the show.
Paul DeLia (01:32):
Great to be with you again, Ken.
Ken Miller (01:34):
Alright, so a couple episodes, episode one, we talked briefly to kind of introduce the series and then I had the pleasure of sitting down with Trip Raymond and episode two at a FA where we really focused kind of on the threat from a war fighter's perspective and a lot of the conversations so far has gotten into the exponentially increasing complexity of the threat and what that's meant for how we respond to that on behalf of the war fighter, both in terms of protection and targeting and other aspects of the mission. So wanted to have you on the show today to talk really to deep dive into the technology to combat this evolving threat and the complexity of it. So with that, based on trip's, comments about the layered complexity of the threat that we face or that warfighters face on a day-to-day basis, what are some of the responses that you're working on that you see from an industrial perspective? What are some of the technology challenges that you're dealing with on a day-to-day basis?
Paul DeLia (02:36):
I think before we get into what we're using for technology to address some of the challenges that Trip talked about, we should talk a little bit about what sort of exists currently in broad terms and how that informs our design. So the approach to electronic warfare in general on platforms, let's say like the F 16 or the F 15 or AC one 30 U gunship or B two,
Ken Miller (03:02):
Which are more like legacy aircraft, correct. Legacy, right.
Paul DeLia (03:05):
Okay. But they'll be relevant in the fight. They have a bunch of boxes, a bunch of line replaceable units, lru on board in order to accomplish the task of either radar warning or a combination of radar warning and jamming. And they take up a lot of room on the airplane and require a lot of power and there's a whole sort of maintenance cycle associated with that. And we've been, those boxes have been developed by industry for many, many years. I started in this industry in the 1980s and we've been supplying boxes, aircraft like that for many, many years. What's been happening in the background is this technology democratization? It's a term I like to use. So in the 1970s we achieved the late 1970s, we achieved sort of an inflection point in the development of integrated circuits and integrated circuits are a big part of how we build systems for the military.
Paul DeLia (04:07):
And the driver for more and more capable ICS really was the defense industry. Around the mid 1970s pong, one of the first video games became available and the video game industry and then the consumer market started driving capabilities and integrated circuits. And that has actually greatly benefited the defense industry. Bizarre, insatiable need for better and faster video games, better and faster communications through our mobile phones enables an entire ecosystem of integrated circuits and capabilities that we can use to build smaller and highly capable systems for the war fighter. So with that framework, what we're looking at in order to counter the pacing threat that trip was described is a combination of using commercial off the shelf software-defined radios combined with very, very smart software to make them mission capable to address the pacing threat. And what I mean by that is the bespoke systems that I talked about earlier that are on F sixteens or F fifteens are purposely built to be on those airplanes that have purpose-built software for them. The nice thing about software-defined radios, which are commercially available is that it's just that a software defined radio, so I can put software on there to make it a radar warning receiver. I can put software on there to make it a radar warning receiver and a jamer integrated in one package. I can put software on there to make it a communication system. I could put software on it and make it a data link.
Ken Miller (05:50):
And that really gets to the heart of the multifunction system in general that you can come up with one system that you can program to do a number of different things versus having to take up the space and weight and power to run several different systems.
Paul DeLia (06:04):
Right, exactly. And that software-Defined Radio is a fundamental building block. Another building block is something that is again, informed by fifth Generation Wireless where you've all seen the commercials 5G and now sixth generation wireless is being defined, but in that architecture they use something called elemental beam forming. And in plain English, what that means is being able to softer control and electromagnetically, combine multiple beams together to perform a function. What that also enables is within one aperture you can have a sensing function, a electronic attack function, a data link function, a rate radar function, and in communications function all in one aperture. So you're taking up less space on the platform and providing an increasing capability. And with programs like the collaborative combat aircraft and programs like Replicator, it's going to be very interesting to have systems that don't take up a lot of space, so low swap, but are in fact highly capable and they're made highly capable by the software that will be loading on.
Ken Miller (07:23):
Now since everything's kind of shifting to software, does this allow you to then, you mentioned that you have systems specifically designed for an F 16 or an F 15 or whatever. Can you more easily develop the capability and put it on the platform that you need it to go on? Or do you still have to provide more definition for that capability to fit specifically on F 16 or an F 15? Are they more universal in capability?
Paul DeLia (07:50):
I want to say more universal in fit and application. The capability is really driven by the software that you put on there. So an example would be if I took a software to find radio that I can buy from, let's say as an example of Mercury. A Mercury computers makes a bunch of software defined radios. If I buy one of their systems and I need to use it on a B 52 as an example, the mission of a B 52 is different than the mission of an F 16. So the fundamental software to make the receiver receive information and process the radio frequency environment and turn it into ones and zeros so that the computers can figure out whether or not it is a threat that's illuminating the aircraft or a cell phone signal just to keep it generic, that processing kind of remains the same.
Paul DeLia (08:44):
The mission data for that platform is going to be different somewhat from the mission data, say on an F 16. The F 16 will be much deeper into the weapons engagement zone versus a platform that would be standoff. And then on the product attack side, the sort of waveforms that you would put out for a fighter aircraft are going to differ from the waveforms you put out from let's say a bomber or a tanker or even a CCA. They'll be different waveforms that you'll be putting out. Core software would remain the same. It would be missionized for a particular application. Then the other part of missionization would be that typically these systems have to work with other systems on board the platform. The DOD has done a good job of establishing standards, ossa, osa, mosa, these are all open system architectures and standards that our software needs to be developed to so that they could be readily integrated with very little cost onto these platforms.
Ken Miller (09:47):
You mentioned the reliance, or at least the increasing contribution of commercial off the shelf technology, and that's been a real shift in how we develop new capability in recent decades. Could you walk us through a little bit of some of the challenges or some of the opportunities even that has provided when dealing with commercial off the shelf? Because a lot of this technology was not developed originally in the commercial sector for a military use. You mentioned you can develop in a certain way to apply it for military purposes, but how has that, could you talk a little bit about some of the challenge of identifying that commercially available off the shelf technology and how easy or difficult it is to kind of use the current processes that we have in DOD to allow for that type of contribution from commercial technology?
Paul DeLia (10:39):
Sure, Ken, I think there's a couple of opportunities and a couple of challenges here to address. On the opportunity side, being able to scale quickly can be a challenge. I mean, unlike let's say the mobile phone industry that we'll make, maybe an assembly line could probably kick out 5,000 phones a day. A large order in our industry would be 5,000 units in a lifetime. So we don't have the volumes like the commercial industry does, but if we want to scale quickly and address things like replicator, which is going to have a bunch of unmanned platforms that are going to require different sort of capabilities, it's much easier for an aerospace and defense company to use the supply chain that has been established for commercial industry and just glom onto that, for lack of better words, to help us help react, react quicker. On the implementation side, there are challenges, right?
Paul DeLia (11:43):
Because yes, these are software-defined radios and they're largely adaptable to what we need to, but the amount of frequency coverage that they have may not be what we need. The amount of RF space, that's the instantaneous bandwidth that these software-defined systems can see at one time may not suit our needs as well. We address that with some clever software and putting some discrete components sometimes on the front end in order to handle the frequency, the frequency disconnect. But what we're finding is more and more companies see the opportunity in electronic warfare, and many of these companies have commercial off the shelf SDRs that do cover the frequency ranges that we're looking for and have the instantaneous bandwidth that we're looking for. And as industry, we work closely with these companies to help inform their roadmaps as to what they should be including in there. So it's sort of a symbiotic relationship that we've developed.
Ken Miller (13:01):
Now, you've mentioned a couple of times the replicator program, which is a new initiative in DOD to basically rapidly field a larger quantity of unmanned systems and try to get the best ideas to percolate up, especially to avoid the valley of death and all the obstacles of the current acquisition process. How has that initiative helped your efforts from a technology standpoint to develop some of the commercial off the shelf software-defined capabilities onto any new system or just UAVs in general?
Paul DeLia (13:39):
Yeah, so in broad terms, when you're talking about smaller unmanned platforms, whether unmanned aerial, unmanned surface, unmanned surface vehicles, and the like, when you come to programs like replicator where you're going to need a lot of mass out there, cost becomes a very big driver. So I can't really take a system that was developed for a fighter aircraft and sort of shrink it down and meet that cost target that I need to. So this approach has really helped us come up with low cost, highly capable, highly capable systems that can be a treatable and expendable should the need be. I think the other part of the equation, we talked a lot about software defined radios, is that you can have all of these software defined radios out in these unmanned platforms, but because of the way that threat is arranged in kill webs versus individual integrated air defense systems, they're arranging kill webs that are sharing information at the speed of light.
Paul DeLia (14:53):
A lot of them are AI informed. So what's the counter to that? And it's really to have these unmanned platforms with these software defined systems on there, all sort of governed in some matter or integrated in some manner through an architecture that allows information sharing in real time and the reprogramming of these assets in situ as they're exposed to the RF environment. So as an example, if systems go into wartime reserve modes that are previously unobserved, what these systems could return is an unknown signal. And so you don't know whether or not it's a friendly or a foe or a previously unseen mode of a certain type of system. So the jamming response would be generic or confused. But if you're able to ship that information back to some cloud-based architecture or even work on those with cognitive algorithms at the combat edge, you could ascertain whether or not that really is an unknown or a previously unseen mode of a threat system, reprogram the mission data file in situ at the combat edge and present the correct jamming sequence out to confuse, deny or deceive the adversary.
Ken Miller (16:19):
That seems rather revolutionary in terms of how we've conducted missions in the past over the last number of decades. What is the next step then? I mean, because there's still a lot of uncertainty as to whether or not the replicator initiative is going to actually produce what it needs to produce, and obviously from your perspective, from an industry perspective, there's still you're trying to blend in a rapidly evolving threat with rapidly evolving commercial technology into an otherwise technology development process that still lags behind. So what is the next step to basically open up doors to kind of key pace with the evolving threat that you're seeing around the world?
Paul DeLia (17:03):
So Ken, one thing that we're keenly aware of in industry is that the playing field, because of technology democratization has sort of been leveled. The technology is generally available to everyone. You can go on Amazon, for instance, and buy something called a Hacker one radio and download some code from GitHub and be up and running in five minutes looking at a DSB information off a commercial airline. Anybody can do it. So that throws a bit of a wrench in the works in this classic measure countermeasure, counter countermeasure chess match that we play in electronic warfare. That's why this connectivity through a cloud architecture becomes so important. Doing things traditionally where you receive intelligence information, turn that into threat information that you reprogram, that cycle takes quite a bit of time. There was a RAND report in the beginning of last year that it can take up to 30, maybe even 90 days to get that done, and that sort of is out of sync with the requirements.
Paul DeLia (18:16):
So where we see the next evolution really is again, taking a page from the commercial environment, right? Many of us have internet connected thermostats, internet connected doorbell cameras, internet connected water sensors in our homes, et cetera. We have this whole Internet of things architecture that presents information and takes that information, sent it right to your phone where you could interact with. So what we're looking at is a connected modern digital infrastructure for electronic warfare or EMSO assets, enabling them to react in near real time to the threat environment that they see, and take it a step further, really, the electromagnetic battle management of these assets, given the distances that we may be engaged in the limited amount of magazine depth that our platforms carry the supply chain to replenish those magazine depths, being able to balance out and come up with the recommended course of action of kinetic versus non kinetic effects to successfully prosecute a mission is going to be increasingly important for us to be successful. And the primary way to achieve that is through this connected digital infrastructure.
Ken Miller (19:43):
And I would imagine that critical to that, and particularly when it comes to time and latency, is going to be the integration of artificial intelligence, machine learning, large language models, different ways of collecting and transmitting and that data, as well as making basically decisions for the commander. Talk a little bit about the role of AI and machine learning and stuff in this effort.
Paul DeLia (20:07):
Sure. Artificial intelligence and machine learning are a technical pillar underpinning everything that we just discussed. If you're operating on a computer console in an intelligence agency to look at elint information that has been collected in order to make it into mission data, you're going to be using AI and machine learning tools to sort of sift through information and get results out to the field, to the field quicker. In the architecture that I just described, being able to operate at machine speed is going to be critically important. The adversary is operating at machine speed. You almost have sort of a machine to machine sort of jousting match going on. The implementation of AI will at a minimum present a combatant commander with a set of courses of action, a set of COAs that they could employ to prosecute a particular part of their responsibility and take into account things like a LR, the acceptable level of risk, probability of kill, probability of win, and present that to 'em at machine speed so they can make decisions much quicker and actually better informed. These architectures will have access not only to the information that is in with that particular area of responsibility, but since it would be all connected, you'd have intelligent, you'd have access to intelligence feeds as well, which you may not have, which combat commanders may not have all the time.
Ken Miller (21:54):
So one of the concepts that oftentimes comes up when we talk about EW operations is Boyd's classic ood loop, and I think it's observe, I might get this wrong, observe, orient, decide, and act, I believe, or what it stands for. So how has technology affected that process when you're talking blue force against red or even blue and ide? So in terms of closing that kill chain, could you talk a little bit about how technology has changed that process?
Paul DeLia (22:25):
Yeah, that's a great question, Ken. So boy zoo, the loop is well understood and well-known kind of how we operate, but technology has really served us well here. And what we discussed a little bit earlier with these connected architectures is really going to have a positive effect. Our purpose in electronic warfare, one of our purposes in electronic warfare is to add as much confusion to the enemies OODA loop as possible. So if you think of the U loop in a circle, observe, orient, decide, and act right, you want to add a lot of clock cycles in between that decision loop so that we can get in with our assets and execute our mission. So traditionally with sort of federated electronic warfare assets, you have F sixteens doing their thing, F fifteens doing their thing, F 20 twos doing their thing, et cetera, and they can be effective.
Paul DeLia (23:30):
But by having an AI informed course of action and electromagnetically battle managing these assets, you can be extremely effective in the area that you're responsible for and add significant amount of confusion to the enemy's OODA loop in order for you to get inside and perform your mission. The way we see that happening right now is really through the use of artificial intelligence and cloud connected architectures. So these things can happen almost transparent to the operator and at machine speed, taking into account that the adversary is using the same sort of tactics on their end to add confusion to our OODA loop, and our mission really is to close our kill chain, our long range kill chains quicker than they can,
Ken Miller (24:28):
And that really kind of gets to the heart of what we're going to be covering in the next episode where we talk about the data fighting and the role that data is going to be playing in this evolution.
Paul DeLia (24:39):
Yeah, hugely important.
Ken Miller (24:40):
Well, Paul, it's always great to talk to you. That's all the time that we have for today. We'll be back very shortly when we sit down to record episode four, but I want to thank you for taking time out of your busy schedule for joining me here on episode three of our CTO series, powered by L three Harris.
Paul DeLia (24:55):
Great to speak with you, Ken.
Ken Miller (24:57):
Thank you. That will conclude this episode of From The Crow's Nest. I want to thank my guest, Paul Delia from L three Harris for joining me in our next episode, episode four of our CTO series. I'm joined once again by Paul Delia, as well as Trip Raymond to take a look at operations from dog fighting to data fighting. As always, we enjoy hearing from our listeners, so please take a moment to review and share this CTO series. You can also contact me via email at host at from the crows desk.org to share any of your thoughts, comments, or recommendations. That's it for today. Thanks for listening.