SAN FRANCISCO — After focusing on optical communications for more than 20 years, Hedron CEO Baris Erkmen, has a clear view of the challenges and opportunities offered by the technology.
Since earning a PhD in electrical engineering from the Massachusetts Institute of Technology with a thesis on optical communications, Erkmen has designed, developed, prototyped and deployed optical technology for space, airborne and terrestrial applications. His career has taken him from NASA’s Jet Propulsion Laboratory to Google’s Project Loon, Facebook Connectivity Lab and Project Taara, a wireless optical extension of the fiber optic backbone from X, the research and development lab for Google parent company Alphabet.
In March 2022, Erkmen joined California-based Hedron, a startup building a space-data-relay network for optical and radio frequency data.
What’s your role at Hedron?
CEO and CTO. We’re a fairly early-stage company so having joint titles is okay. Our mission is to build a modernized communication infrastructure for Earth-observation satellites. The way that we communicate with our Earth observation satellites is quite frankly archaic. We’ve got beautiful instruments up there. We’ve got consumers down here that can do an incredible amount of processing on cloud-compute platforms. The pipe between the two of them is stuck in 1960s. That’s what we’re trying to change.
What brought you to Hedron?
The thing about Project Taara that excited me was that it wasn’t an engineering-demonstration project. We needed a product, something that you could get off the factory floor and deploy in the terrestrial telecommunications market. Telco’s are very price sensitive and need good performance metrics. I went through the humbling experience of taking an engineering prototype and turning it into a product that works well.
Around that time, people started talking about optical intersatellite links. I thought that since I learned so much from this experience of turning a prototype to a product, wouldn’t it be amazing if I could apply those same principles to a space environment. Serendipitously around that time, Hedron and my paths crossed.
What’s unique about Hedron?
We will support automatic networking in space. We need to support multiple access [points] for multiple satellites being able to talk to the ground dish. Scheduling should be happening in the background. Satellite operators pass their packets to a network and the network takes care of it. The data shows up on the other end.
Terrestrially, that’s what we take for granted. That terrestrial technology needs to be deployed in a space environment and it’s been slow to get adopted. That’s where we see an opportunity. Optical plays a strategic role in that but it’s not the only thing we’re focused on.
Aside from the technology piece, our focus in on creating a network that is modular and scalable. Nowhere in the world is the communication grid a uniform grid of interconnected towers, where the whole thing goes down if one of them gets knocked over. It’s seamless and all works together. You can have 5G in a mall. You step outside, go 100 meters and you have 3G service. That kind of architecture needs to be rolled out into space.
What’s next for Hedron?
We’re advancing on three fronts. We have a satellite that’s in integration phase. It’s a technology demonstrator. We’re hoping in the next year to put that up and to validate that we can perform the fundamental technological pieces of ingesting data in orbit and relaying data down to ground.
Commercially, we are focused on the first cluster. Think of it like rolling out cellular to a city. We have a neighborhood of Earth-observation satellites. If we were able to deploy some connectivity in orbit, it would improve their connectivity by more than double and it would cut down their latency by more than a factor of five. We’re focused on understanding how that solution can help customers and figuring out if there is a meaningful relationship to be set up there, so we can spend the next two years building that and deploying that.
Last but not least, we are focused on optical communications. It’s a strategic piece of our infrastructure. We think there’s opportunity for disruption in the market. We’re focused on an internal design.
It can be challenging to transmit optical signals through Earth’s atmosphere. How can that be solved?
What you do in many networks where there are links that are not reliable, is go for spatial diversity. Then, the question becomes the economics of that solution. Can you afford to have 10 optical ground stations for one link, for example? That’s where the industry has work to do. I’m part of the industry so this is self-reflection more than me wagging my finger.
What do you recommend?
Ground stations need to be a lot more economical, robust and portable, not fixed assets like telescopes that can’t move and cost a lot of money to put up. The barrier here is the economics of ground stations and the feasibility of placing them.
I’ve done some studies myself and others have done studies showing you need five to nine times more ground stations than the number of links you want to support. You also need diversity of sites, you need them to be separated more than 1,000 kilometers. That’s solvable.
Point-to-point optical signals are harder to jam or intercept than RF signals. With spatial diversity in your ground network, data moves through terrestrial networks. Does that compromise security?
I don’t think it’s fair to say the network is not secure anymore because you have fiber in the network. If you can make security for one segment of the network a non-issue, that allows you to focus your resources on making the other side more secure. Put all your resources into making the fiber part secure.
If you want true security, one of the things that’s interesting is the fact that at optical frequencies, we’re able to leverage the quantum nature of light. While we’re in the early days of figuring out how to harness that, we’ve demonstrated that we can do it. Therefore, optical has operational advantages like being narrow beams and point-to-point. In the next 10 years, it also has physics-based advantages that might make it actually quite a bit more secure to use.
Do you need to install optical communications stations on mountains or in the desert?
The reason to do that is because the probability of cloud is lower there. But no. Seattle, is not great because it’s cloudy and rainy most of the year, but certainly you don’t need to go to the middle of the Mojave Desert to set up a ground station. There are plenty of dry places that have good weather. And again, if you have diversity, you’re not relying on a particular site being always available. It becomes far more manageable.
Has the war in Ukraine changed your view in any way?
Commercial companies played a significant role in exposing what’s going on. But if you notice, everything is still archival. That’s because of the latency. It takes tens of minutes, sometimes hours, to get that data. Whatever was transpiring has already transpired. What you see is the wreckage left behind.
Wouldn’t it be amazing if you could get information with a few minutes of delay and respond? That’s the piece that I think is missing. The current conflict helps, but it hasn’t necessarily unlocked everyone’s imagination because what’s possible is not yet obvious.