A Defense Communications Forum panel, moderated by retired Colonel Michael Black of AFCEA International, asked how to turn edge AI and private 5G from slides into something a forward-deployed unit can actually use.
MITRE chief technologist Venkatesh Ramaswamy argued the architecture must be built for fast-changing conditions, while other panelists focused on spectrum coordination, maritime deployment constraints, and assured command-and-control.
Reference architecture: mission, orchestration/security, and edge
Ramaswamy described a defense-grade reference architecture as broadly similar to a commercial 5G network but with important differences. He placed a mission layer at the top that defines intent and priority, including who gets what, followed by an orchestration and security layer that translates centralized control into distributed execution.
He described the edge as the heart of the system where data lives, where compute happens, where data sovereignty is preserved, and where latency is kept in check, with connectivity treated as the 5G transport.
Deployability over planfulness: dynamic networks and anti-overengineering
Ramaswamy said nothing is planned and conditions are rapidly changing, so the network must be able to change quickly and dynamically, with the architecture assuming conditions will shift faster than any pre-planned configuration can keep up with. He also argued the industry often builds for what is possible rather than what is necessary, likening it to designing a Swiss army knife when a knife will do.
Ramaswamy tied over-engineering to complexity, saying complexity breaks, and proposed a litmus test: whether the network is quickly deployable, can be operated by a soldier without much training, and produces a measurable improvement to the mission. If all three answers are yes, he said the network probably is not over-engineered.
Spectrum coordination mechanics: tokens, 1494 friction, and CBRS-style sharing
Salvador D’Itri, chairman of the National Spectrum Consortium, framed spectrum superiority as a coordinated maneuver across air, land, and sea, and said 5G has to operate at the same speed of coordination to have a seat at the table. He argued the gap is not really about regulators but about the military’s own coordination machinery, pointing to the 1494 spectrum certification process as showing its age against the pace of modern radio hardware.
D’Itri said he wants a model where a 5G system gets issued a token that authorizes it to operate across DOD-controlled territory, the UK, CBRS in the U.S., or elsewhere, but he said that token-based authorization model is not yet available. He also suggested NATO Band 79 as a candidate sandbox for allied coordination in the sub-6 range, describing it as high enough in the sub-6 range to be useful and common across NATO, while having real range limitations in difficult terrain.
On sharing models, the panel said CBRS has already proven mil-on-commercial sharing at scale, and work at Marine Corps Logistics Base Albany is demonstrating mil-on-mil 5G sharing. The panel also said the Advanced Spectrum Coexistence effort in OUSD R&E, led by Josh Weaver, is pushing into bands like 3.1 to 3.45 GHz where incumbents are faster and less predictable than CBRS-style sharing can handle.
Maritime and assured C2: hybrid coverage and edge decisions at the tactical edge
Florin Constantinoiu of the Romanian Navy Headquarters offered a coalition and maritime perspective and said he was speaking from his own experience. He described maritime environments as lacking dense tower infrastructure, with signals scattering off waves and salinity corroding equipment as platforms move through international waters.
The panel said coverage beyond 50 to 70 kilometers from shore generally requires a hybrid model combining satellites and floating base stations. It cited NATO’s REPMUS exercises in Portugal as a useful working example, with Constantinoiu’s Centre of Excellence participating for the last five years; starting with REPMUS 2023, a mobile private 5G network was used to test uncrewed maritime systems and stretch their operational range from roughly 10 kilometers in 2023 to almost 30 kilometers in 2025.
Constantinoiu said the ultimate issue is about people, trust within teams, and understanding that they need each other, and he said he hopes mixed results from bridging two providers’ 5G bubbles will be sorted in the next iteration. The panel also described interoperability as a recurring issue because most uncrewed systems use commercial-off-the-shelf components with proprietary data formats, while saying NATO’s CMRE developed protocols tested in REPMUS and industry is starting to align with NATO requirements, with formal standards still being written.
Lieutenant Colonel Benjamin Pimentel of the U.S. Marine Corps spoke on Assured C2 and Project Dynamis, arguing that pairing a high-throughput network with multi-access edge compute enables a forward unit to ingest sensor and logistics data, run an AI/ML decision aid, and act without reaching back to cloud resources. He said the high-speed network facilitates speed of operational tempo at the edge so decisions can be made faster, and he described using a high-speed last-tactical-mile network to reach back for expertise when over-the-horizon connectivity is good.
Pimentel cited a Future G project supporting an austere airfield construction effort where an edge 5G bubble helped operators coordinate using mobile devices and map-based apps instead of hauling additional radios. He argued for leveraging existing RAN sharing agreements with host nation mobile operators when possible to support faster deployments, saying that using host nation RAN sharing agreements keeps operations within an established regulatory and licensing framework and sidesteps the slower 1494 process.
On autonomy, the panel said mission autonomy and PACE plans intersect in assured command and control in contested, mobile, or disconnected conditions. It said Pimentel walked through an autonomy stack including platform autonomy, collaborative autonomy across platforms and domains, and mission autonomy; that mission autonomy allows an operator to hand a platform an objective and let it work; and that platform-level PACE plans can include a single UAV or unmanned surface vessel carrying MANET radios, cellular, and LEO SATCOM, with these approaches combining to provide continuity of command and control across a wider band of battlefield conditions.