Rail's Digital Transformation: FRMCS, ETCS and the Case for Trackside Private 5G

The Invisible Piece of the Digital Railway

Everyone who has watched a signaller's simulator knows what ETCS Level 2 looks like. A train driver no longer stares at lineside signals; instead, the cab display tells them exactly how far they may travel and at what speed, and the next movement authority updates as the train rolls. The whole system only works because the train and the radio block centre can hold a two-way data conversation that is both safety-rated and continuous. For the last twenty years that conversation has been carried over GSM-R.

GSM-R was a brilliant piece of engineering in 2000. It gave European railways a harmonised 900 MHz bearer, mission-critical voice, and enough data capacity to carry the relatively light ETCS traffic. By 2030 or thereabouts, though, it is unsupportable. The chipset supply has dried up. The base-station vendors have already given end-of-sale dates. The capacity is far below what the next generation of ETCS — and the rest of the digital railway — wants to use. And nobody wants to be the rail operator trying to explain to a regulator why the train-to-ground radio depends on 2G technology nearly thirty years after the rest of the world moved on.

FRMCS — the Future Railway Mobile Communication System — is the answer agreed internationally. The UIC has defined it. The European Union Agency for Railways is certifying it. 3GPP has baked mission-critical services into its 5G specifications specifically to support it. In the UK, Network Rail's Digital Railway programme has the transition on its roadmap, and the East Coast Digital Programme is the flagship deployment vehicle for the next generation of signalling.

FRMCS is a 5G technology. That single fact carries implications the rail industry is still working through.

Signalling, Reimagined

ETCS Level 2 already does a lot of work in the GSM-R pipe. ETCS Level 3 — moving-block signalling, where the railway stops relying on fixed lineside blocks and instead knows where every train is with enough precision to hold them a safe braking distance apart — requires far more. Every train has to report its position and speed at high frequency. The radio block centre has to reason over dozens of trains in a corridor and issue movement authorities continuously. The wireless bearer has to be low-latency, deterministic, available everywhere along the line, and robust to the loss of any one base station.

FRMCS on a 5G bearer gives all of this. Network slicing — a capability built into 5G — lets the railway carry ETCS traffic on one virtual network, safety voice on another, and everything else on a third, with guaranteed quality of service between them. Cellular handover is native, not bolted on, so a train passing between base stations never loses its movement authority. The capacity is several orders of magnitude beyond GSM-R, which means the cab display can carry richer information, the track worker's tablet can carry video, and the rolling-stock telemetry stream can sample at a useful cadence.

The hard part is not the standard. It is building the physical trackside network that the standard runs on — a chain of base stations, fibre backhaul, core network, and spectrum licensing that covers tens of thousands of route kilometres and has to hand off seamlessly from one railway's coverage to the next.

Trackside Coverage Is A Private Cellular Problem

Here is where the argument turns. The obvious way to build FRMCS coverage is the way GSM-R was built: a dedicated railway radio network, owned and operated by the infrastructure manager, using spectrum reserved for railway use. That is broadly the European plan, and Network Rail is aligned with it. The harmonised spectrum for FRMCS in the European band is being defined.

But "a dedicated railway cellular network" is, in technical terms, a private 5G network. The base stations are railway-owned. The core is railway-owned. The spectrum is railway-licensed. The tenants are railway applications. Everything in the operator's playbook — network slicing, mission-critical services, service-level assurance, security — applies directly.

Seen that way, the interesting question stops being "FRMCS or not?" and becomes "what else should share the trackside network once it exists?" A trackside cellular fabric capable of carrying ETCS and safety voice can, with almost no marginal cost, carry:

• Passenger Wi-Fi backhaul — currently the UK's greatest source of passenger complaint about rail and a stubbornly expensive problem to fix with point solutions
• On-train CCTV and body-worn camera footage — today uploaded in bulk when the train hits a depot, tomorrow streamed in real time
• Rolling-stock condition monitoring — bearings, wheels, brake discs, HVAC, door cycles, all feeding back to the maintenance depot rather than waiting for a driver fault report
• Track worker safety — wearable devices reporting location, vital signs, and proximity to line, integrated with possession management
• Station operations — CCTV, PA, passenger information displays, platform gap sensors, and retail
• Trackside infrastructure monitoring — axle counters, point machines, level crossing equipment, OLE telemetry

Every one of those workloads is currently served by a separate system, with its own radio, its own contract, and its own operating model. A unified trackside private network collapses them into one managed fabric, with slicing to guarantee that safety traffic is never crowded out by a passenger streaming video.

What The UK Programme Looks Like

Network Rail has been explicit that the trajectory is towards FRMCS and towards digital signalling more generally. The East Coast Digital Programme is rolling out ETCS Level 2 across the southern section of the East Coast Main Line between London King's Cross and the approaches to Peterborough, with commissioning stretching across the decade. The Transpennine Route Upgrade is incorporating digital signalling principles into the reworking of the Manchester–York corridor. The Elizabeth Line in central London already operates with ETCS in the core tunnels, supported by a dedicated in-tunnel cellular infrastructure carrying both operational and passenger traffic, originally delivered by Boldyn Networks — and widely cited as an early example of what an integrated trackside cellular layer looks like.

Beyond the main programmes, the UK rail industry is full of smaller, quieter projects that all end up needing the same wireless layer: depot connectivity for new rolling stock, in-station small-cell deployments to improve passenger signal, hybrid fibre-wireless pilots on regional lines, and freight-specific telemetry programmes on long haul corridors. Rail Delivery Group, the Rail Safety and Standards Board, and the Department for Transport's rail innovation funding have all backed pilots where private cellular is the common thread.

Freight And Depots Are Part Of The Same Network

There is a tendency to treat freight as the poor relation in rail-digitalisation conversations. That is a mistake. Freight wagons are exactly the asset class that benefits most from continuous condition monitoring: bearings fail with catastrophic consequences, and today's maintenance regime is largely calendar-based because the wireless layer to do anything else has not existed at scale. FRMCS-grade trackside coverage makes it possible. Adding a modest private cellular cell at a freight interchange or a wagon-maintenance depot — already an entirely standard deployment — extends the same coverage into the yard where wagons are built up, inspected, and handed over.

The same argument applies to passenger depots, where the question of how to keep a large fleet of new EMUs continuously uploading condition data has been solved on a project-by-project basis for a decade. A single private cellular network covering the depot is the obvious, boring, correct answer.

An Honest Note On Timing

Nobody should pretend the FRMCS transition is imminent in the sense that a live UK line will be using it next summer. The standards work is still finishing, the spectrum co-ordination is still in progress, and the rolling-stock changes required at scale are a decade-long undertaking. But the decisions being made now — about trackside cellular infrastructure, about depot connectivity, about station small cells, about passenger Wi-Fi architecture — are the ones that determine whether FRMCS arrives on a ready-made private cellular fabric or requires a second, parallel build.

Any trackside wireless project commissioned today, whether it calls itself FRMCS or not, should be designed to be compatible with the direction of travel. That means picking the hardware vendors, the spectrum strategy, and the operating model that make the eventual transition an upgrade rather than a rebuild.

Trackside Cellular For Regional Operators

The mainline programme will play out on its own timescale, with its own governance, across the big corridors. Meanwhile, regional and heritage operators, depot owners, and freight terminals have the opportunity to install private cellular networks at their own sites now — improving their own operations, building familiarity with the technology, and being ready for the day when the national bearer lands on their doorstep. None of those projects is a multi-year capital programme. Most of them are straightforward deployments over a working depot or a regional yard, running in weeks rather than years.

The digital railway is coming whether the sector is ready for it or not. The quiet question is not when it arrives, but whether the wireless fabric underneath is already in place when it does.

---

If you're planning a depot, a station refurbishment, a regional line, or a freight terminal and want to talk through what a trackside private cellular layer looks like in practice, get in touch. Read more on our transport and logistics sector page.