Should 5G Coverage Targets be Optimized for Mobility?

by Assaf Aloni, Continual

What should mobile network operators’ obligations be regarding coverage, especially on highways? It has long been a source of annoyance to subscribers that the percentage figures targeted by MNOs relate only to where subscribers live, rather where they are likely to travel. Although it can be a nuisance not to have adequate coverage in your home, it could present a real safety risk if you don’t have coverage in mobility, for example in the case of an accident or breakdown when you’re out driving. And this will only become more critical when significant numbers of connected and autonomous vehicles (CAV) begin to depend on 5G for normal operation.

This mobility paradox is now starting to be recognized by the regulators. The recent competitions for spectrum for the launch of 5G services in particular countries are as likely to be decided on the promises of the operators regarding highway and land mass coverage–rather than just population coverage—as on their ability to pay huge sums for the licenses. Indeed there is a growing recognition that lower prices for licenses will leave the successful operators with more funds to invest in network densification, which is the real key to ubiquitous connectivity.

The German 5G spectrum auctions, which ended in August 2019, introduced for the first time legally-binding coverage obligations for the license winners that include road, rail and waterway routes as well as households, including a requirement for 100% coverage to be in place on motorways by 2022.¹

Other countries are taking a somewhat different approach. In the spectrum auction in Denmark that ended in April 2019, the licenses at 900MHz carried an obligation to improve coverage of less densely populated areas by sharing responsibility among the three successful bidders, who each had to take on the commitment to serve just one of three individual segments of the designated land area with minimum speeds of 30Mbps download and 3Mbps upload by April 2022. Although not specifically tied to highways, they would implicitly benefit from this approach as well.

The UK had also planned to impose geographical spectrum coverage obligations to a certain proportion of the landmass in each of its four national regions², but fell short of pinning targets against the actual roads to be covered. However, as of mid-October 2019, a new plan has been announced under which all four UK operators will collaborate to deliver the Shared Rural Network, whereby the four operators will share existing masts and infrastructure to address almost all partial “not-spots”, and provide coverage for an additional 16,000 kilometers of roads.

There is no doubt that highway coverage needs to improve if Connected Car customers are not to lose patience with the car OEMs, the perceived providers of the connectivity. But what is the best approach for regulators to balance this criterion with their aims of maintaining competition and securing maximum license revenue?

With auctions having largely replaced the old-style ‘beauty contests’ for allocating spectrum licenses in recent years, it is now being recognized again that a more controlled policy can have a positive effect when it drives the market to deliver desirable outcomes in terms of land and highway coverage, capacity and competition. Countries like China and Japan, who both allocate spectrum based on the operators’ commitment to meet coverage needs rather than charging upfront fees, have denser networks with higher levels of connectivity than that seen in Europe and the United States. In Latin America, the best geographical coverage is found in Chile, which uses a ‘command and control’ spectrum allocation policy. Traditional percentage population coverage is not a difficult sell, as it has a direct relationship to potential subscriber market, but covering highways that run through under-populated areas requires a bit more persuasion.

So can a combination of more closely controlled contests and legally binding coverage obligations for travel routes, similar to the German model, could be able to show us the way forward? Or should operators be encouraged to cooperate, as in the UK, or share responsibility, as in Denmark? The road ahead for 5G will be an interesting one.

Notes:

1. The four German operators— Deutsche Telekom, Vodafone Germany, Telefonica Deutschland, and new entrant 1&1 Drillisch—accepted coverage conditions in return for extended terms of payment for acquiring the spectrum licenses, although the auctions attracted some criticism about the high sum (a total of $7.3 billion) and the lengthy bid process. While 1&1 Drillisch has been given a relaxed obligation to allow it to become established, the specified coverage conditions for the three incumbent operators include a requirement for 100% coverage to be in place on motorways by 2022. This is in addition to a commitment to supply minimum speeds of 100Mbps to at least 98% of households in each federal state by the end of 2022—a metric that also applies to federal highways, other major roads, and railways. By the end of 2024, 5G spectrum holders in Germany will be obliged to provide speeds of 100Mbps to all other main roads, while covering smaller roads, railways, seaports, and main waterways, with data rates of at least 50Mbps.
2. Ofcom 2018 policy document on improving mobile coverage