New LEO constellations

After the failure of the low-Earth orbit satellite constellation projects in the 1990s and 2000s, the mood today has decidedly lifted. Both technologies and competition have evolved, and the number of projects has been climbing steadily since 2015. Despite this there is still no shortage of obstacles. Setting aside all of the media hype, who will truly be able to rise to this technological and financial challenge, and what impact will it have on the players involved? Are LEO constellations really the new frontier that everyone is saying?

Is LEO the best orbit to bring high-speed access to the planet?

At less than 2,000 km above the Earth, a low-Earth orbit satellite travels round the globe several times a day. The coverage that a satellite at this altitude can provide is of course much less than what a geostationary satellite located at 37,000 km is capable of, but the greater proximity also means lower latency, which is crucial for certain applications. More importantly, when launched in constellations, these satellites make it possible to increase available capacity and decrease bandwidth costs, which is vital to enabling satellite to hold its own in new markets.

At the end of 2017, there were some 15 LEO projects in the works, each involving anywhere from around a hundred to several thousand satellites. The SpaceX constellation is the one that has received the most media coverage, although it is not the most advanced.

There are still a number of obstacles to overcome, not only regulatory but also financial, logistical and technological. Here, the OneWeb and Telesat projects are the most developed, and could become a reality by 2019 – 2020.

A slew of challenges…

The theory is simple: launch a large number of small satellites that are cheaper to buy and send into orbit. With this denser network, spectrum can be reused and capacity multiplied. The reality is more complex, however. From a regulatory standpoint, orbit authorisations and frequency licences must be obtained, which often requires long coordination periods. From a logistical and manufacturing perspective, the satellites’ very design and production need to be rethought, to move from manual to chain assembly, while also finding the necessary launch capabilities. Some technologies also still need to prove themselves and mature. So, again, there is no shortage of challenges to ensure these initiatives are successful.

… and probably not enough room for everyone

Only two or three of these initiatives are likely to come to fruition. The actual demand is relatively small, and excess capacity would undermine their business models, while terrestrial networks continue to expand, and 5G is in the starting blocks. There are, however, certain areas where existing approaches complement each other: LEO and GEO on the one hand, and satellite and terrestrial networks on the other. Movement is already afoot in the satellite industry to take this new environment into account. What remains is collaboration with terrestrial operators, where a great deal still needs to be done.

 
A number of LEO constellation projects of widely varying capacity
Main LEO satellite constellations
Project Estimated cost Backers No. of satellites Total estimated capacity (Gbps)
LeoSat 3.5 billion USD JSAT, TAS (as a satellite provider) 100 1 000
OneWeb (World VU) 3.5 billion USD for the initial 640 satellites (incl. Gateways) Softbank: 1 billion USD + 700 million USD from previous backers (Intelsat, Virgin Galactic, Qualcomm, Hughes Network Systems, Airbus Defence and Space, Bharti Enterprises, The Coca Cola Company, Grupo Salinas) 2 862 20 000
SpaceX A potential total cost of 10-15 billion USD according to observers 1 billion USD investment from Google and Fidelity 11 518 103 662
Boeing Not known Unknown support. Reportedly a big US Internet company is ready to support the project 2 956 26 604
Telesat Not known Telesat mainly 117 936
Source: IDATE DigiWorld
 
Truly but not fully complementary orbits
Competition and complementarity between LEO and GEO companies
Source: IDATE DigiWorld