Table of Contents
Panelists say defense networks must assume day-one attack and lean on civilian infrastructure to stay up
The definition of resilience in defense communications is shifting, and so is the way to deliver it. At the Defense Communications Forum, a panel of experts from the Department of War, Cohere Technologies, and Orange’s defense and security business laid out a picture of networks that have to assume they’ll be attacked on day one, integrate civilian and military infrastructure, and somehow keep pace with commercial innovation that lives mostly inside the 3GPP standards process.
The conversation, moderated by RCR Wireless News editor-in-chief Sean Kinney, featured Regina Tyrrell of the Office of the Undersecretary for Research and Engineering, Anton Monk, head of technology strategy at Cohere Technologies, and Olivier Martel, CTO and operations chief of Orange’s defense and security business unit.
Shift from redundancy to cognitive resiliency
The panel opened on the question of how resilient defense networks actually are today, and the answer was essentially that the goalposts have moved. Tyrrell argued that the very definition of resilience is changing.
“I think to your point that you started with resiliency, is that definition has changed…it used to be redundancy, and I think it’s now more towards a cognitive, smarter way of thinking about our networks and our connectivity,” she said. Pace planning, gradual degradation of comms environments, and edge node processing are now central to how planners think about staying connected under pressure.
Monk picked up the thread by framing the threat environment in stark terms. “We’ve seen it in the recent war, we’ve seen it in others over the past year or two, that the expectation needs to be of a day one network attack,” he said. That assumption, in his view, drives everything else — heterogeneous networks, distributed comms and sensing, and integration between non-terrestrial and terrestrial ecosystems become essential rather than nice-to-have. He also flagged the long-running work on alternate waveforms, with the open question being how to integrate secure waveforms into a stack that actually scales.
Adding complexity to a network does introduce new points of failure. But the panel’s consensus was that this is a tradeoff worth taking on, because the alternative — a small number of monolithic systems — fails worse and faster in a contested environment.
Securing operational sovereignty through interoperability
Martel grounded the discussion in recent history. In Europe, and particularly in the lessons coming out of Ukraine, roughly 90% of communications were running over civilian infrastructure. That reality has reshaped how Orange is thinking about its defense and security business, which launched last year to bridge the gap between civil and defense infrastructures and leverage telecom assets for homeland security.
For Martel, sovereignty isn’t really about a national flag on a piece of equipment. It’s about freedom of movement — the ability to combine different technologies to answer different use cases. And that requires interoperability. “I think interoperability is a good answer to sovereignty,” he said.
Tyrrell echoed that point through the lens of the Department of War’s Open Compute and Universal Deployment (Akudu) effort, run out of the Future G office. Standardizing on commercially-driven technologies and contributing to a shared codebase, she argued, actually reduces single points of failure rather than introducing them, because problems get found and fixed by a community of practice in real time.
“You are actually eliminating single points of failure, because in the case of a Kudu, you’re relying on the community of practice to find, fix, and solve problems in real time in the software that the Linux Foundation is hosting,” Tyrrell said. The added benefit, she noted, is breaking vendor lock and giving the department faster access to innovation without waiting on the standards bodies themselves.
Overcoming hurdles in the 3GPP standardization process
That last point led into one of the more pointed stretches of the panel — the difficulty of actually influencing 3GPP. Monk, whose company has spent years pushing the OTFS waveform inside the standards process, was candid about the structural challenges.
Small companies aren’t alone in struggling with 3GPP influence, he said. Academia, the US government, and even some large companies face the same problem, for reasons that range from geopolitics and IP protection to the sheer headcount disparity. Large commercial vendors routinely send 100 people to working group meetings, with several hundred more back home doing research and IP development. The US government, by comparison, sends a handful.
The response, Monk argued, isn’t to try to overhaul 3GPP. It’s to build extensible frameworks alongside it. Akudu is one piece of that. So are domestic efforts like the Next G Alliance, which is trying to align North American companies around 6G strategy, the National Spectrum Consortium, which is driving innovation in areas like ISAC through targeted funding, and an upcoming NSF Alliance workshop focused on US wireless leadership. None of these replace the standard. But they create room for the kind of competition-driven innovation that the US has historically done well, and they let specific national security requirements get addressed without waiting on consensus from a global body.
Tyrrell, whose office sponsors the NSC’s other transaction authority, said the relationship works both ways. Industry partners can pull the department aside and say, in her words, “no, no, don’t go in that direction, you need to be thinking about this.” That’s the kind of informed-buyer dynamic she wants more of.
Advancements in integrated sensing and communications
The conversation closed on integrated sensing and communications, with counter-UAS as the headline defense use case. Martel offered a tangible example, like Orange’s Drone Guardian service, announced in March, which is now running in parts of France, with plans to expand across Europe. The service today relies primarily on RF-ID sensing, but Orange has deliberately built it on an open architecture so it can interface with a range of sensor technologies, and the company is already testing what 5G-based sensing might look like alongside an ecosystem of startups.
Monk’s preview of the ISAC discussion that followed the panel sharpened the business problem. Commercial operators have to make money, and dedicating part of the network to sensing eats into capacity. “The operators have to make money,” he said. “Sensing overhead is a really critical piece…if it’s on a continuous basis, is going to cut into capacity and therefore margin.”
He also flagged a standards risk. 3GPP groups are already pushing to lock down OFDM as the baseline waveform for ISAC, the same way they did for comms. That may be fine for initial commercial sensing use cases like health, wellness, and home security, but it’s a problem for national security applications where the priority is detecting drones as early and as accurately as possible. The likely path forward, Monk suggested, is a phased one, with US government investment seeding counter-UAS use cases and other commercial use cases evolving over the next five years or so.
Tyrrell’s closing point tied the spectrum and ISAC questions back to the broader theme of getting things into the field. The Department of War, she acknowledged, has historically not had much appetite for shipping incomplete solutions. That has to change.
“I know I’ve never seen in the department an appetite for getting an 80% solution out there and possibly failing to some point. And I think we just need to all have that mindset that we need the commercial sector to pull us along and get us out there, and you know, help us learn from the failures — or maybe not so much of a failure, but the not-complete solution, and how we keep iterating on improvement,” she said.
That, more than any specific technology, was the panel’s actual conclusion. Resilient, multi-technology defense networks aren’t going to be built by waiting for the standards to settle or for a perfect product to arrive. They’ll be built by integrating heterogeneous networks, embracing the new points of failure that come with that complexity, and letting commercial competition drag the public sector along faster than it would otherwise move.
