A millimetre-wave system trialled on Caltrain's corridor sustained around 2 Gbps through heavy rain — 20 times the published requirement of 100 Mbps to the train — and the outcome is reshaping how transit agencies should write rail connectivity contracts that run for a quarter of a century.

That is the core argument advanced by Anthony Murray, CEO of Blu Wireless, in an analysis of specification practice across the rail connectivity market. His starting point is structural: contracts span 25 years, yet the performance figure at their centre reflects the traffic volumes of the year they are signed.

Why do bandwidth floors become ceilings?

The pattern repeats across the industry, Murray writes. An agency publishes a bandwidth requirement based on current needs; the winning bidder meets it; the service launches. A few years in, that initial requirement has hardened into a ceiling. The agency then chooses between a degrading service and a second procurement midway through the contract — with trackside construction, track outages and premium night labour on a live railroad.

Regulator modelling shows the scale of the gap ahead. In 2018, Ofcom concluded that within 10 years a busy UK mainline train could require around 1 Gbps, and a crowded commuter train two to three times that. Deutsche Bahn projects a median demand of 1.15 Gbps and an upper limit of 3.6 Gbps per 1,000 passengers by 2028.

Does passenger demand tell the whole story?

No, according to Murray. Safety and security systems draw on the same network. Agencies increasingly stream onboard video to the ground in real time, and with AI-enabled analytics running onboard, each 4K camera may need roughly 20 Mbps of continuous upload capacity. Ten cameras place 200 Mbps of demand on the uplink before a single passenger connects. A specification written solely around passenger downloads risks starving safety-critical traffic.

Murray sets out four tests agencies should apply before signing:

• Treat the published minimum as a starting line. The margin a bidder demonstrates above the floor identifies who can serve the railroad for decades, not just today.
• Test on live track, in real weather. Paper evaluations group bidders together; live trials in the actual corridor separate them before a 25-year commitment is made.
• Stress-test the economics. Site works dominate trackside capital costs, far outweighing the radios, so tower spacing drives the budget. Per-gigabyte pricing converts rising demand into a rising invoice; owned infrastructure on unlicensed spectrum behaves more like a fixed asset.
• Write technology refreshes into the contract. Radio technology will move through at least one generation over the term; renewal should be a planned line item, not a future negotiation.

What did Caltrain's procurement prove?

Caltrain's tender, run alongside electrification of its San Francisco–San Jose corridor, set the 100 Mbps floor and attracted widespread expressions of interest. The agency invited two finalists onto live track in spring 2022. Over a 0.7-mile section in Redwood City, the winning millimetre-wave system sustained around 2 Gbps in heavy rain — 10 times the competing trackside Wi-Fi bid and roughly 20 times the requirement.

The network now operates across 65 masts. Route testing reported by the project team shows average speeds above 500 Mbps with peaks approaching 1.5 Gbps, while observed passenger demand tops out near 300 to 380 Mbps — a margin by design. The radios ride on masts erected for electrification and connect over fiber already in the ground, containing both construction cost and disruption.

UK field data points the same way. A 2026 measurement study commissioned by Ofcom found onboard Wi-Fi on the one British route with trackside high-bandwidth backhaul met the regulator's good-performance threshold 83.2% of the time; across the rest of that journey the figure fell below 1%, matching a 1% national average for onboard Wi-Fi.

Every specification embeds a forecast, Murray concludes, whether its authors intend one or not. Agencies that accept it pay in instalments — first through a service that falls behind expectations each year, then through a construction programme to catch up. The agencies that will look farsighted a decade from now are the ones demanding demonstrated headroom before the ink dries.