Constellation architect
By Debra Werner|April 2023
David Voss, director of the U.S. Space Force's Spectrum Warfare Center of Excellence
To prevent strategic and tactical surprise, U.S. intelligence analysts and military officers have long relied on a rich variety of collections, including optical and infrared images, radar detections and electronic signals gathered by satellites. Historically, those collections came almost exclusively from large, government-owned satellites, but that is changing due in part to analysis and advocacy led by David Voss, an electrical engineer turned military planner at the Air Force Research Laboratory facility at Kirtland Air Force Base in New Mexico. Voss and his colleagues conducted analysis, peer review and war games showing that a higher-functioning collection architecture would include small satellites, some perhaps owned by a government agency, but others not. So in 2015, they began research into an idea they called the Heterogenous Space Architecture, later renamed the Hybrid Space Architecture. The idea is now reflected in a portfolio of projects and has spread beyond intelligence. I reached Voss via video conference to discuss the Hybrid Space Architecture.
Q: Where did the idea for the Hybrid Space Architecture originate?
A: About eight years ago, a number of folks realized we were moving out of an era of large monolithic systems in GEO. We were seeing the smallsat technology evolution. Commercial capabilities were coming online rapidly. Our allies were investing very heavily. We were trying to understand within an enterprise, not necessarily a single program, the right mix. I was leading the smallsat portfolio at AFRL. We were developing about 20 satellites. Our objective was to understand information latency in distributed satellite systems. That means obtaining information as rapidly as possible. When we look at applications for intelligence, surveillance and reconnaissance or space weather, oftentimes we’re moving from a single sensor doing the job to multiple sensors coming together to achieve the information product that we want.
It’s important to understand how we got to GEO. Historically, space was extremely expensive and the domain of large nation-states. We went to GEO because it was the most efficient, affordable way to provide as much global service as possible. Space has become a contested domain, and we had this technology evolution happening over the last 20 years with smallsat technologies. A lot of times, people would compare very large, many-thousand-kilogram satellites to cubesats, which is comparing apples and oranges. What we were trying to go after was, in any mission area — whether it’s communications, intelligence, surveillance and reconnaissance or space weather — what is that right mix, and how do these systems work together collaboratively? We originally were calling it the Heterogeneous Space Architecture. But it was a mouthful to say. “Hybrid” became the word a large number of us across the community started using.
Voss explained later that this Hybrid Space Architecture is a portfolio of projects. For example, a Hybrid Space Architecture project from the Defense Innovation Unit, the Pentagon’s Silicon Valley arm, is focused on applying the concept to military communications. — DW
The idea was to stop thinking about trying to cram all the requirements for a mission into one platform, which tends to drive schedule challenges and requirements complexity, and to embrace the hybridity in a way that has the highest performance, lowest cost and highest resiliency.
Q: How was the concept studied?
A: There was an explicit research investigation at AFRL. As is common with AFRL, we work across the community. We have strong partnerships at NASA and strong partnerships with industry. We were trying to be true to the data. When would you use proliferated LEO, a constellation covering a geographic area with a large number of LEO satellites, and when would you want a large monolithic satellite in GEO?
It’s been presented in conferences. It’s been peer reviewed a lot. It was trying to encourage people to stop thinking about it as an “us versus them” mentality. How do we embrace all the capabilities that the nation can leverage? It’s also not only the technical hybridity. It’s not just the orbit hybridity. There are very legitimate differences between how we might procure military systems in terms of a military conflict, and how we would want to embrace commercial systems when we’re talking about global information transport. When are there unique military requirements? When would we embrace commercial capabilities from a DoD or a NASA standpoint? There’s lots of dimensions to this hybridity that we were trying to understand across not only the U.S. government, but also in partnership with commercial and our allies. I ran a nine-nation hybrid architecture investigation with our allies to work the same kind of questions as they are investing heavily in space. There have been quite a few analytic efforts here at the Space Warfighting Analysis Center. We also have attempted, in partnership with our allies, to bring some of that Hybrid Space Architecture to joint exercises to understand that utility in a war game scenario.
He’s referring to war games including RIMPAC, the 26-nation maritime-focused Rim of the Pacific Exercise held every other year from Honolulu. During RIMPAC 2018, military, civil, allied and commercial satellites helped participants assess airbase activity, and identify and track vessels. The military exercise showed “that a cooperative heterogeneous space architecture does have advantages and value, and that micro-satellites and nanosatellites contribute significant capability,” according to the paper, “Demonstration of a Heterogeneous Satellite Architecture During RIMPAC 2018.” — DW
Q: What are the opportunities associated with the Hybrid Space Architecture?
A: You can have far better information products if you’re able to appropriately ingest all the information sources available to you. Your weather models, your ISR products, your space situational awareness can become better. There’s a lot of capability in LEO. There’s a lot of capability in GEO. There’s a lot of capability in MEO. Smallsats are good for distributed spatial and temporal variation investigations, but they don’t have the size, weight and power for some of the exquisite instruments that are important for our weather models or other applications. If we have a NOAA satellite with a high-performing sensor that allows it to make exquisite measurements but it’s so expensive that we only have a few of them, what if we combined a bunch of sensors that are less performant, but we have far more of them because they’re a lot more affordable? That would give us much better spatial-temporal resolution.
Q: What are the challenges?
A: Remember, many of our legacy missions are delivering critical services to warfighters every day that can’t be disrupted. If you’re in a program office delivering, let’s say, the GPS capability to the globe, it’s very hard to take a cubesat precision navigation and timing source and fold that into your baseline. Without a framework that allows us to think about performance variability and trust variability, it turns into a problem of, “Hey, that’s great. But I can’t use it.” How do you trust the data? How do you ingest data that might not be as exquisite but gives us much better spatial-temporal resolution? We have to not only understand the sensors, we have to understand the dissemination of data from those sensors. We have to understand how we’re processing that data. And then we have to fuse that data in a way that gives credit to the performant sensors versus the less performant sensors in a model that we refer to as a variable-trust architecture. We have to do that now in very short periods of time. And then once we have that information product, we have to disseminate it to the users. You can think of your map application on your phone. At the end of the day, all we really want to know is “How do I go from point A to point B most efficiently?” The mapping algorithm fuses together many data sources. There’s a long way to go still for all of us to figure out what that recipe looks like for the Hybrid Space Architecture across different dimensions for different communities. If you’re in the intelligence community, the recipe for embracing different capabilities might be different than if you’re on the civil space side of the house.
Q: You mentioned a variable-trust architecture. What is that?
A: We oftentimes think of binary trust: You either trust data or you don’t. But if you have a variable-trust perspective, you know which data sources might be interesting. You’re going to take in the data sources but not act on them. If you have an extremely high-trust source, you can go straight to acting on that data.
Q: Could this architecture lead to government agencies deciding not to procure certain kinds of satellites because industry or allied partners were already doing that?
A: Yes, when you know how you want to partner with allied and commercial, it allows you to focus the military unique systems for the conflict, and the benign environment can be for commercial. For instance, we leverage commercial fiber transport as an information distribution mechanism today but would want military satcom for a fight with a peer.
Q: Are you surprised at how quickly the idea of the Hybrid Space Architecture has caught on and been embraced? I’ve heard NOAA officials talk about its potential to improve weather models, and National Reconnaissance Office leaders discuss its potential to augment traditional intelligence sources.
A: I have been very surprised at the fact that the Hybrid Space Architecture framework has been agreed upon so rapidly across the community in many different agencies. There’s a sense of urgency that is helping to drive that speed of transition as we look at some of the pacing threats out there. I am surprised, but many folks were already starting to move in that direction. Giving a name to the framework has allowed some of the coalescing to happen much faster than it might have happened. It is exciting to see us rapidly moving in this direction.
