UK rail connectivity: Netflix for all?

The trade association techUK describes itself as a grouping that brings together people, companies and organisations “to realise the positive outcomes of what digital technology can achieve”.

That was certainly the aim of a recent techUK workshop called Future Spectrum Demand for Rail, which it hosted in late March. But were there any positive outcomes?

Certainly, as my presentation for Real Wireless explained, there are a number of possibilities for track coverage and passenger connectivity, mainly involving standard cellular, Wi-Fi and/or mmWave technologies.

Of course mmWave and Wi-Fi could offer dedicated high-speed, high-volume data, with mmWave having the capability to deliver in excess of 1 Gbps. But they are often proprietary and require massive investment in new trackside infrastructure and onboard components such as antennas and gateways.

A dedicated solution involving mmWave or Wi-Fi could make sense in some cases – but who is going to pay for it? Whilst we all want to achieve a great customer experience with good passenger connectivity, rail operators may not see a business case, end users may not be impressed at paying extra to access data on a train and taxpayers may not want to help government foot the bill.

A passenger train is essentially the equivalent of the population of a small village moving at high speed and through often challenging terrain from a propagation perspective. The point I felt needed to be made at this workshop was that we should perhaps step back from the expectation of at-home speeds and bandwidth when considering connectivity under such challenging conditions.

In this context, regular cellular connectivity from the mobile network operators (MNOs) begins to make more sense, with an existing nationwide infrastructure as a starting point. Continuous connectivity could be achieved by adding sites to address not-spots instead of building a completely new network. Whilst there is the potential of passive train coverage from the outside (where the signal would have to overcome the train’s attenuation), many trains are already equipped with onboard 4G/LTE gateways. These gateways contain the MNOs SIM cards, hence there is a cost of data that has to be considered too. Furthermore, train passengers would have to share the networks coverage and capacity with other nearby users, which can be a challenge, especially in rural areas in rush hour. But capacity upgrades, carrier aggregation and optimised coverage (targeting the railway track) could address this challenge to a large extent.

Whilst no solution is perfect, it still begs the question, why build a new network if you already have a mobile network that gives you coverage to most of the national railway – and which can do a lot more with enhancements and optimisation?

Which brings us to what sort of demand there is and whether it should be met. Passenger media streaming, if available, is expected to dominate usage and could generate in excess of 80% of all traffic. But should Netflix, Amazon, YouTube and BBC iPlayer usage be deemed critical?

There’s an argument that voice, video calling and basic data should be seen as the minimum requirement on trains and prioritised. Lack of continuous coverage and the ability to maintain a voice call, after all, is still one of the biggest complaints by rail passengers, not the lack of Netflix.

If that is seen as a way forward, then the UK mobile operators initiative JOTS (Joint Operator Technical Specification) Rail, to which I am a contributor, could be an answer.

JOTS Rail aims to take advantage of existing, countrywide cellular networks and fill coverage gaps based on standardised JOTS Rail (joint MNO) infrastructure. Site sharing is at the heart of this approach where all MNOs can connect to a shared infrastructure that supports all their technologies and frequency bands, including GSM-R.

But MNO coverage improvement could soon have a boost. A real revolution is taking place on the operational rail side: rail operators will soon have their own version of 5G: FRMCS. It will offer digital rail operations based on 10 MHz of harmonised spectrum at 1900 – 1910 MHz (harmonise in Europe, unfortunately not available in the UK for now). Studies from European countries forecast that the rail operators trackside infrastructure has to be doubled, for spectrum range and redundancy reasons. This is an opportunity for infrastructure sharing between rail and mobile operators. Could the vast amount of trackside infrastructure involved create sufficient synergies and be a passenger connectivity opportunity?

Real Wireless is already working with neutral hosts and rail operators in Europe, looking at exactly this kind of situation: how these extra sites for rail operators delivering FRMCS coverage could create synergies with the mobile operators and result in better experience for rail passengers. There is, however, one tricky issue for the UK: 1900 MHZ spectrum usage. Unlike much of Europe, this band may not become available.

Whichever approach the UK chooses, I certainly believe there needs to be much more dialogue – dialogue between operators, regulators and the UK rail industry, e.g.  about freeing up or sharing the 1900 – 1910 MHZ spectrum band (if possible), about shared infrastructure, about the potential benefits of FRMCS to both operational and passenger connectivity, about mmWave vs cellular vs Wi-Fi and, importantly, about how much the various approaches cost – and who pays?

Such a dialogue should also, I believe, be driven by appropriate real-world needs. You need vast amounts of data to offer Netflix; you don’t need vast amounts of data to offer business-focused applications, including occasional video conferencing. So why is streamed entertainment necessary for trains? Why not focus on real business needs rather than Hollywood films?