It’s been just over a year since the government announced its ambitious target to have free Wi-Fi on trains by 2017. While the intention is obviously a good one (who doesn’t want connectivity on trains?) there are still significant barriers in place that are hindering the country’s changes of getting anywhere close to that target.
I was recently at an event called Going Underground a couple of weeks ago discussing the ins and outs of connectivity on trains. What’s clear from that event is that there are technical challenges with on-board Wi-Fi that won’t go away. Wi-Fi’s access technology “Carrier Sense Multiple Access – Collision Avoidance” (CSMA-CA) is not designed for high-density environments, such as busy commuter trains in rush hour with high capacity demand caused by a large number of concurrent users. In other words, when everyone on a train is trying to use on-board Wi-Fi at the same time to stream live sport or the latest Game of Thrones episode, we drive Wi-Fi into its limitations.
The technical limitation in such a high usage scenario lies in the way the Wi-Fi access points and devices interact with each other. To avoid data collisions, devices “sense” the Wi-Fi channel — listening to see if another device is transmitting data. Once a device sees that the channel is busy, it backs off to avoid collision of data, and a counter starts to count down before the device checks again to see if the air interface (the channel) is available. So, when too many users try to transmit data, devices start to go through a downward spiral of repeatedly backing off and trying again, thereby reducing the AP efficiency by 50% or even much more depending on the number of users trying to access it — resulting in less capacity per access point for more concurrent users.
Peak hour trains on busy commuter routes in particular take a triple whammy when it comes to on-board Wi-Fi:
- The sheer number of people trying to access a single access point overloads the system
- The sheer amount of bodies in one train can attenuate the signal between the access point and devices, rendering it poor (meaning low efficiency) to useless
- Peak trains tend to be full of commuters whose data needs tend to be far greater than non-commuters, which, again, overloads the system
Small cells, in particular femtocells, might be a better solution than Wi-Fi because they are more efficient when handling a high number of concurrent users and high traffic — but that still doesn’t solve the the backhaul challenge. Performance is always limited to whatever the backhaul can achieve, which is typically 4G. Hence, if there’s no mobile coverage, the whole on-board connectivity system — whether it’s Wi-Fi or femtocell based — is useless. Connectivity systems could use satellite backhaul for rural locations, but that in itself is a very expensive option. Alternatively, connectivity systems can use on-board repeaters, which don’t need backhaul and bring the signal outside the train to the users inside. On-board repeaters, though, still rely on reasonable outside coverage.
However, none of these technical challenges are insurmountable, with the exception of the inherent Wi-Fi technology challenges.
We do believe that the main barrier to enhancing on-board connectivity is the business model. We also believe that the requirement for trains should be on-board connectivity and capacity, independent of specific technology (such as Wi-Fi). At the moment, mobile network operators don’t have a revenue incentive to cover railway tracks or install on-board equipment because in a world of fixed and all-you-can-eat data packages, the average revenue per user (ARPU) doesn’t increase with incremental coverage and capacity on trains.
Therefore, the business case is the biggest bottleneck at the moment to improving on-board connectivity. If the government truly wants to provide Wi-Fi on 90% of journeys by 2018, it will have to manufacture a business case through regulation in order to kick things along.
Our own research a few years ago found that a clear business case could exist if the industry looks beyond Wi-Fi to mobile connectivity as a whole. We also found that on-board equipment deployment is cheaper than improving outdoor coverage to such a level that users inside the train could be served from outside. A business case would therefore have to clearly list the benefits to multiple parties, including advertisers (amongst many others), who would be able to clearly see the opportunity for ad revenue based on the length of passenger journeys, and rail companies, who could use the connectivity to improve day-to-day operations to become more efficient.