This article was originally published in Capacity on 13 March 2023.
High altitude platforms (HAPs) and satellites have been promising to revolutionise mobile connectivity for at least 30 years. HAPs such as drones, aerostats and balloons have been suggested for spot coverage, emergency use and more recently projects such as Google Loon tried balloons to deliver connectivity in Africa. Starting with Iridium, satellite operators have tried to connect to handsets since the late 1990s. Yet I doubt that many cell phone users have ever made a call, or connected, via anything other than a terrestrial base station.
HAPs providers have come and gone, and satellite operators such as Iridium had to go through bankruptcy before a commercially viable system could emerge. Yet a raft of new providers is trying all over again. New HAPs providers include Airbus with a lightweight drone solution and Altaeros with a tethered aerostat. There are many others using large drones, small planes, balloons and more. New satellite providers or services include the iPhone emergency messaging solution, recent announcements from Iridium and Qualcomm and new entrants including AST and Lynk. Why is there such a resurgence of activity? And will it finally lead to a world where connectivity from the skies is routine for mobile users?
The technical aspects of the connectivity are relatively straightforward. For HAPs the link budget is similar to terrestrial use given that the platforms are typically only a few km away – similar to cell towers. For satellite the link budget is more challenging, but low earth orbit (LEO) satellites and sufficient power and bandwidth can deliver at least messaging and perhaps voice and browsing services to standard handsets.
Previous failures have broadly been due to economics. The cost of keeping enough HAPs platforms airborne has been too high, often because they frequently need to be brought down for recharging, replacement of helium, change of pilot or other maintenance. The cost of LEO satellite networks is high, and hard to justify for a mobile messaging service.
For coverage from the skies to work this time round costs need to be lower, the willingness to pay higher, or ideally a combination of both.
Costs are getting lower
Costs are reducing. HAPs providers are innovating to deliver lower-cost drones and aerostats that can stay airborne for months at a time. New antenna systems are enabling these platforms to deliver larger numbers of spot beams with greater power levels, reducing the number of platforms needed. Satellites have become much less expensive as launch costs have plummeted due to SpaceX’s innovation, and satellite manufacturing costs have reduced as unit numbers have grown and economies of scale start to arise. Quite what these cost reductions add up to is mostly commercially confidential, but it does appear that the economics are now much more favourable.
The willingness to pay, at least indirectly, also appears to have grown. While consumers may not, in general, be willing to pay more per month on their cellular subscription, Apple has shown that they might pay more for their handset which can fund the cost of satellite emergency calls. Operators have costs associated with extending their coverage, perhaps as a result of coverage obligations, and airborne solutions might be a lower way to deliver against those objectives.
There remains much doubt about the economics. It seems unlikely that this will be an area with large revenue compared to, e.g., mobile operators, but it may be sufficient to pay for the costs of delivery. Apple, T-Mobile, SpaceX, Iridium and Qualcomm clearly believe this to be the case.
Spectrum and regulation remain a barrier
As with all mobile solutions, spectrum is paramount. Ideally, the HAPs or satellite system should use spectrum and standards already present in the handset. Otherwise, special handsets will be needed. Apple partially get around this by making both the handset and providing the service, but even there only the iPhone14 can currently use the service. The Iridium/Qualcomm solution is not, at the time of writing, available commercially because modified handsets are not yet on the market and early interest has not included the largest handset manufacturers such as Samsung.
Often HAPs is not allowed because it is, effectively, a cell tower with height well beyond the maximum permitted in spectrum licenses. There may also be other restrictions from aeronautical regulators. Satellite transmissions in cellular bands are generally not allowed because current spectrum management separates terrestrial bands from satellite bands.
This is changing. The Federal Communications Commission (FCC) has put forward its supplementary coverage from space (SCS) proposal to allow satellite operators to use mobile spectrum if the mobile network operator (MNO) that owns the spectrum agrees. If approved, this would also seem to open the way for HAPs. Regulators in other countries will take note, and may, over the coming years, adopt a similar approach.
Even with regulatory change, finding a way to use the same spectrum both terrestrially and from the skies is challenging. Carefully designed spot-beams can help tailor HAPs and satellite coverage around cellular usage, but such beams are expensive, and hard to form, especially from a satellite that is moving rapidly across the sky. Border areas between countries are equally challenging. But if there is cooperation from MNOs and from regulators on both sides of the borders then there are potential solutions.
Increased societal pressure for a solution
Cellular users, in general, may not value HAPs/satellite solutions because they rarely have any need for them. But politicians and others are exercised by issues such as rural hardship, digital divide and the large number of unconnected people in developing countries. As a result, pressure for these solutions may come more from governments than from operators. Governments might partly-fund solutions in order to achieve near-ubiquitous cellular coverage and philanthropists or aid agencies might also help financially.
For example, in the UK, the government has spent £500 million partly funding the Shared Rural Network, a set of about 500 base stations in areas with poor coverage that can be used by all MNOs. It is conceivable that they might spend another £500 million to deliver complete coverage via HAPs. The US has similar funding programmes for rural areas.
To be clear, HAPs and satellite systems will not provide as good a service as a well-designed terrestrial system. HAPs can come close with 5G-like services, satellite solutions to handsets may never extend beyond messaging and basic browsing. Neither are likely to be as good inside buildings or cars because their signal comes vertically and has to pass through roofs. But this is still a lot better than no service at all.
Will it be different this time?
There have been so many false dawns for both HAPs and satellite-to-handset connectivity that it would be easy to be cynical. And there is no step change in cost or functionality that has suddenly overcome past problems, rather there has been a gradual evolution. It is not certain that it will be different this time.
But conversely, there is now a satellite service to handsets – albeit a very limited one. There are compelling drivers to deliver ubiquitous coverage. Economics are better and regulators are moving in the right direction.
Perhaps most importantly, governments are intervening to try to deliver rural connectivity. It is apparent that terrestrial coverage is not the answer, with rollouts slowing or stalled in developed countries. The steadily improving economics of HAPs and satellite and governmental pressure and incentives may well be sufficient.
It is far from a certainty, but on balance, this time might be different. The world should get behind this initiative.