2 December 2022

In the 1990s, ambitious plans for constellations of Low Earth Orbit satellites to provide global coverage were dashed by high costs, bulky equipment and low demand. Times have changed, bringing the promise of satellite technology closer to market. Andrew Wright looks to the stars.

Low Earth Orbit (LEO) satellite constellations were a folly of the dot com boom. Satellite and wireless technology companies with the post-cold war, economy-boosting peace dividend on their minds, embarked on ambitious plans to launch constellations of satellites that would provide the world with mobile coverage. Others planned to do similar for broadband. Satellites had already caused a minor revolution in broadcasting, and now they would do the same for mobile telephony and broadband.

Blue sky thinking

Terrestrial mobile telephony, the theory went, was so full of black spots that a satellite version – which reached almost the entire planet – would be markedly superior. Similarly, good broadband was the preserve of the lucky few – satellite could bring the whole world up to speed. Limited mobile and broadband were already available from geo-stationary satellites but came with low capacity, high costs and bulky terminal equipment – adequate for an ocean-going vessel, but not for average consumers.

LEOs would bring the satellite close to the user, making the terminal less bulky, allowing higher capacity than geostationary satellites and eliminating round-trip delay.

Down to earth

However, the theory did not stack up to reality. In the 10 years the constellations took to deploy, terrestrial networks kept on improving – coverage was enhanced, and mobile telephony became mobile broadband. Fixed broadband also improved as ADSL became VDSL, cable TV networks transformed into broadband networks, and eventually fibre was rolled out.

Factor in the cost of satellites and the capacity they could provide, which was limited compared to the improved terrestrial networks, and the business case didn’t add up. Two mobile LEO constellations (Iridium and Globalstar) were launched but failed commercially. The terminals were too bulky, the capacity too limited and the service too expensive, although Iridium did find a niche for military use.

No broadband LEO constellations were launched – the business case was not good enough.

The new age of LEOs

Now LEOs are back, providing both mobile and broadband capabilities to the mass market. The mobile service, for now, is limited to SOS messages for iPhone users, via Globalstar, but will extend in future.

Whereas Iridium, with its second-generation constellation (Iridium NEXT) is focussed on its military niche, Globalstar has struck a high-profile deal with Apple in a new role as a provider of emergency (SOS) messaging services. Through this service, iPhone users will be able to send SOS messages from almost any location – the service is live in North America and will be expanded soon.

The key aspect of the Globalstar/Apple service is the messaging capability delivered to a standard mobile handset – which includes Globalstar’s spectrum band, but no bulky antenna to boost the signal gain towards the satellite. The price for using a standard smartphone as the terminal is that the capacity of the system is low, as will be the data rates – hence the limitation to SOS messages.

It seems an appropriate use for Globalstar’s second generation satellites. The SOS messaging system nicely matches Globalstar’s capability – high value, low volume traffic – and is a differentiator for Apple. It has the added benefit of getting Globalstar’s spectrum into handsets, opening up future possibilities.

In the stars

Globalstar’s new role is valuable and innovative, but a new generation of broadband LEOs could make a more meaningful impact. The first generation stayed purely in the mobile space, but new LEOs are aimed at broadband – and the leading broadband LEO constellation, by far, is SpaceX’s Starlink.

Starlink is highly regarded by its users – it provides very usable (>100Mbps) broadband connections to rural customers and certain mobile users (e.g. cruise ships) from an advanced, dish-like terminal. It’s not exactly mobile, but it is transportable, with its RV service available for USD135 per month in the US.

Where Starlink differs from earlier LEOs is that it uses greater amounts of spectrum in the Ka and Ku bands, has far more satellites in a lower orbit, and uses advanced antennas. There are currently ~3000 satellites in orbit with the final constellation to consist of ~12 000, being launched ~50 at a time.

Other broadband LEO constellations, such as OneWeb (an emerging constellation of 648 satellites) and Amazon’s Kuiper (a planned constellation of ~3000 satellites), are behind Starlink in launching their offerings, and it is difficult to see how they can catch up sufficiently to secure commercial success.

But can Starlink succeed where earlier LEOs failed, at a time when 5G and fibre deployment are increasing user expectations and mobile broadband is reducing the areas where Starlink is required?

There is no doubt it has an enthusiastic user base in rural America – for those without access to high-speed terrestrial infrastructure, Starlink is a game changer. However, it still suffers from limited capacity. It was recently denied funding under the FCC’s Rural Digital Opportunities Fund, and the limited number of connections it can provide in rural America appear to be part of the reason. An inherent limitation is that the capacity is spread across the globe – you can’t sell too much of it to wealthy rural Americans.

Looking to the future

Elon Musk, SpaceX’s CEO, has described Starlink as financially risky. He places a great deal of emphasis on its second generation constellation, which will add spectrum in the V band for broadband, and in the 1900MHz band for mobile. It will also add a mobile component that is more capable than Apple’s Globalstar SOS messaging system. SpaceX announced, with its launch partner T-Mobile USA, connectivity for messaging apps, with the ambition to potentially add voice and other low-speed data.

Starlink must therefore deliver a significant step-up in capability and capacity, including:

• higher capacity for broadband terminals – to keep up with broadband usage while driving prices to a level that allows capacity to be sold worldwide as successfully as it is in North America
• a mobile connection that can deliver services with a reasonable capacity to handled terminals with conventional antennas (i.e. no bulky specialist satellite antenna).

To achieve this, the second generation of Starlink is expected to have the following characteristics:

• 30 000 satellites
• a separate set of antennas and radios for mobile – including a large unfolding antenna
• a launch rate of up to 400 satellites per launch – using SpaceX’s Starship rocket.

Will new LEOs be more successful than the original?

Arguably, Starlink is already demonstrating more success than Iridium and Globalstar. However, long-term success may require improvements in the second generation constellation. Such improvements are not a given and require further support and development, including: regulatory approvals; delivering the most powerful rocket ever launched – which must also be re-usable; reducing the launch cost to a dramatically lower level; and achieving a good launch cadence at ~400 satellites per launch. While huge challenges remain, all of the above might be within reach – but only time will tell if Starlink is a success.