Radar, not standards, may be the faster path to ISAC drone detection
In sum – what we know:
- Wide-area gaps – Weiss said the DoD wants ISAC to track RF-dark targets across broad domestic areas for critical infrastructure protection, not to replace existing radars.
- Heroic self-interference – An ISAC tower must cancel roughly 140 dB of its own transmitted signal to see small drones returning -10 dBsm echoes.
- Systems before standards – Anderson argued working deployments should come first, with 3GPP codifying what works while operators need a sensing-as-a-service market to fund it.
Integrated sensing and communications, or ISAC, has become one of the more talked-about capabilities on the road to 6G. The premise is straightforward. Cellular networks already blanket wide areas with RF energy, so why not use that energy for sensing — essentially radar — in addition to carrying data? A panel moderated by Anton Monk, SVP of Technology Strategy at Cohere Technologies, brought together Martin Weiss from the DoD’s Future G office under OUSD R&E, Gus Anderson of Lockheed Martin, and Daniel Bliss, a professor of electrical engineering at Arizona State University and CEO of Dash Tech Integrated Circuits, to dig into whether the concept can actually serve defense needs.
The central tension is pretty clear. 3GPP is building out an ISAC roadmap as part of its broader 6G effort, but the defense community’s requirements — particularly around drone detection — don’t map neatly onto a standards process designed to satisfy global commercial interests. Bliss framed ISAC as a special case of what he calls “RF convergence,” the idea that if you’re radiating RF energy, you should use it for multiple purposes simultaneously. That framing matters, because it suggests the defense path and the 3GPP path may not converge as smoothly as some expect.
The question, then, isn’t whether ISAC has potential. It clearly does. The question is whether cost, commercial incentives, and standards timing will align quickly enough to deliver something the DoD can actually use — or whether working systems will need to come first and standards will follow.
Why defense is looking at ISAC alongside radar
The DoD already operates some of the most capable radar systems in the world. So why bother with ISAC at all? The answer, according to Weiss, comes down to wide-area surveillance of objects that don’t emit RF signals — so-called RF-dark targets — across broad geographic areas. Domestically, that means critical infrastructure protection and homeland security missions. He pointed to Operation Spider Web in Ukraine as a wake-up call. “I think that woke up a lot of people to realize that we need to have better approaches to detecting potential hostile threats from the inside,” Weiss said. At the tactical edge, any additional intelligence source that feeds into the broader operational picture improves situational awareness for warfighters.
Weiss was careful to note that ISAC is meant to supplement radars, not replace them. The alternative — mounting low-cost radars on cell towers — runs into spectrum problems and other logistical challenges. Whatever ISAC delivers, he argued, its total cost of ownership has to beat building out an equivalent suite of standalone radar hardware.
Anderson, speaking from the perspective of a company that builds large, exquisite long-range radar systems, drew a useful contrast. Traditional radar covers enormous distances with high precision using megawatts of tightly focused energy. ISAC, by comparison, leverages many low-cost, low-capability sensing points using energy that’s already being radiated. “The real problem is how do we leverage that which already exists to do something new,” Anderson said. The value proposition isn’t in matching radar performance at any single point, but in converging detections across large geographic areas to build a composite situational awareness picture. That’s a fundamentally different design philosophy, and it requires radar engineers and communications engineers to meet in the middle — something that, by all accounts, hasn’t fully happened yet.
Drone detection is the headline use case, but Weiss argued it probably isn’t sufficient on its own to justify commercial deployment. He pointed to biometric sensing — detecting heart rate and respiration rate — and hyper-local weather detection as dual-use applications that could strengthen the commercial case. The Future G office generally leans toward leveraging commercial systems wherever possible, because the economics of bespoke DoD-specific hardware are, as Weiss put it, “pretty punishing.”
Technical hurdles for drone detection
Even with the broad strategic rationale in place, the engineering challenges for drone detection through cellular infrastructure are serious. Small UAVs have a radar cross-section on the order of -10 dBsm — roughly 0.1 square meters — which makes them extremely difficult targets. The goal is to detect these objects at around five kilometers, which is roughly one to two cell sites out from a given tower.
Range resolution, interestingly, isn’t the main problem. Bliss noted that the bandwidths available in cellular systems provide adequate range resolution for this kind of detection. The harder problem is cross-range angle estimation, particularly in elevation, which Anderson identified as one of the biggest obstacles to generating accurate 3D tracks. In urban and suburban environments where ISAC would be most useful — you wouldn’t bother with it in open terrain when a traditional long-range radar works fine — clutter from buildings can overwhelm the target return. Bliss described clutter mitigation as having a dramatic effect on system performance.
Self-interference is arguably the most daunting technical barrier. Because an ISAC-equipped cell tower would be transmitting continuously, the system needs to mitigate self-interference by roughly 140 dB to detect the weak returns from small targets. Bliss called that level of mitigation “heroic.” The implication is that practical ISAC designs will likely need pseudo-monostatic or multistatic architectures rather than true monostatic configurations, where the transmitter and receiver are co-located. Cellular receivers were engineered for carefully controlled receive power with limited dynamic range — a very different problem from trying to pick a faint radar return out of a sea of self-generated noise.
Anderson offered a useful way to think about accuracy trade-offs. One type of sensing mechanism might deliver roughly 10 m/s accuracy, while another could achieve around 1 m/s or one meter of positioning accuracy. Sub-one-meter precision is only necessary for targeting solutions. For a use case like police tracking a drone through a suburban environment, the accuracy bar is lower. “It doesn’t have to be A to the exclusion of B,” Anderson said. “You can have a portfolio of capabilities where there’s a trade space.”
Standards, cost, and commercial incentives
So would accelerating the 3GPP roadmap by five years deliver what defense needs? Anderson was blunt. “The design requirement said build a horse, and so they built a committee, and out came a camel,” he said. His view is that waiting for a standard to emerge and then designing around it is backwards. Working systems should come first, and 3GPP should follow — codifying what works to improve competition and enable new vendors to participate.
3GPP is making progress on the hardware front, accepting advanced MIMO configurations such as 30 2T/30 2R and 60 4T/60 4R antenna arrays. But those configurations imply hardware upgrades to cell towers, which contradicts what many operators had hoped for from 6G — software-only improvements. With a typical intersite distance of roughly three to five kilometers, Monk raised a practical point. If hardware upgrades are required on every second or third tower rather than every tower, the capital deployment cost picture changes dramatically. Anderson countered that existing link budgets already support some level of detection with infrastructure as it stands, and a portfolio approach across different sensing mechanisms could avoid an all-or-nothing upgrade scenario.
Weiss described his office’s stance as minimalist on standards. “Standardize what you must, but leave it at that,” he said. The real pressure point for standardization comes in multi-vendor environments — an Ericsson base station needing to coordinate sensing with a Nokia installation, for example. Whether that coordination happens through 3GPP or through partner agreements worked out between vendors is, in Weiss’s view, an open question. His office favors software-based approaches, like the Okudu program, precisely because software-defined systems can adapt to changing environments faster than hardware-locked standards cycles allow.
The commercial incentive question looms over all of this. Mobile operators exist to monetize spectrum. Anderson argued there’s an opportunity to create a “sensing as a service” market that would give operators reason to invest without waiting for a formal standard. If operators can sell a new capability from infrastructure they’ve already deployed, the economics start to work. Bliss, though, offered a cautionary note. “If we honestly look at ourselves as technologists, we did oversell millimeter wave in terms of its usefulness fairly broadly,” he said. Commercial players may be skeptical that ISAC economics will hold up any better. The tension between national security urgency and commercial return on investment isn’t going to resolve itself easily — and figuring out how to satisfy both is, as Bliss put it, a complicated problem.
Table of Contents
