The challenges satellite Internet must address for take off

Telecoms.com periodically invites expert third parties to share their views on the industry’s most pressing issues. In this piece Marco Hogewoning, Manager of Public Policy & Internet Governance at the RIPE NCC, looks at what it will take for satellite connectivity to become commercially successful.

The satellite Internet industry has taken off at great speed in the last year, as companies like SpaceX, OneWeb and Amazon each work to build satellite-based networks consisting of thousands of small individual satellites. SpaceX has launched more than 1,000 of its Starlink high-speed Internet satellites to date and now has permission from the FCC to launch 12,000 satellites in the first phase of its deployment. While SpaceX has led the way, we are seeing OneWeb and Amazon follow at breakneck speed, as each of them strives to connect the unconnected parts of the world.

The impact of this will be that no matter where businesses, governments, schools, or individuals are in the world, they are able to access high-speed Internet. But as satellite-based services take off, concerns around regulation, technical limitations and economics are mounting amongst the Internet community. There are roadblocks ahead that the industry must tackle if satellite Internet is to get off the ground.

Who governs satellite Internet?

As satellite systems develop, questions are being raised around who exactly should govern an Internet in international space. While improved global access to information and communication is certainly positive, concerns around managing content regulation, hate speech and illegal or harmful materials are coming to the fore.

Understanding who is best placed to govern these systems is complicated because low-Earth orbit (LEO) satellites might make use of ground-based infrastructure in different countries and customers can be anywhere around the world. This leaves uncertainty over which country’s laws apply where and which regulatory bodies can – or should have a mandate to – enforce them. But it is important to remember: while satellite systems are above the earth, they should not exist above the law.

As well as discussions around what happens on the Internet links these systems provide, we can also expect more debate on the rules and responsibilities around frequency allocation, orbital debris mitigation and de-orbit procedures. Launching a satellite is one thing, but repairing one is another. As the latter is not so easy, it’s vital for operators to figure out what happens when they break down or are no longer in use.

Moreover, with the emergence of these truly global networks, we might also have to revisit certain Internet policies, for instance how Regional Internet Registries should assign resources. This presents a contrast to many public policy discussions which we see increasingly focussing on national and regional interests, with governments seeking more autonomy over Internet infrastructure and resources.

What technical limitations face the industry?

From a technical standpoint, a space-based Internet still has a number of challenges that need to be addressed. Very densely packed constellations of LEO systems are a cause for concern on account of the space debris that those systems leave behind, not only from launch vehicles but abandoned, broken or end-of-life platforms as well. The latest figures related to space debris, provided by the European Space Agency, show some 28,210 debris objects are already circling our planet. Similarly, with constellations running into the thousands of systems, a collision is becoming ever so likely, not to mention that some of these orbits are not all that far above where the International Space Station (ISS), which is packed with people, makes its rounds.

Back on the ground, we also need to look at spectrum allocations and the possible interference of these systems with our ground-based infrastructure. Many systems are designed to operate in the Ka band which poses challenges with things like “rain fade”, which operators will want to avoid. If patches of bad weather can disrupt connectivity, satellite-based connectivity might not be the solution people hope it will be in some regions. Using alternative frequency bands might be better in that respect, but the deployment of 5G and other ground-based systems are competing on what ultimately is a limited resource meaning trade-offs have to be made.

Outside of transmitting the packets, routing traffic around a network of thousands of satellites is another challenge. LEO systems are not stationary and circle the globe at breakneck speeds. From a users point of view that means that you might be talking to a new satellite every few minutes, as one disappears behind the horizon, another one comes up on the other side to take over the system. Whilst this is not all that different from how cellular networks deal with a moving person, this time the antenna is what’s moving. That’s if the user is staying still, which is often not the case. As the user starts moving around, things get even more complicated.

Another complexity in routing is that due to their low orbit, not every individual satellite is in reach for the ground-based antenna that provides a connection to the rest of the Internet. This means that data packets need to find their way across, being transmitted from one satellite to the next until it’s in range of a ground station. Doing that efficiently and keeping track of the different paths is a challenge that engineers are currently busy finding solutions for, adapting routing protocols to cater for these rapidly changing networks.

When will satellite systems become profitable?

Launching satellites into space still remains a hugely expensive investment and as Starlink’s founder, Elon Musk, once put it: “the biggest challenge remains not going bankrupt”. While the market continues to evolve, the cost of deploying a satellite system ranges from $5 billion to $10 billion. But unfortunately for operators, it doesn’t stop there. Satellite companies will need to manage different upkeep costs to maintain a satellite system year-on-year. Satellites also have a limited lifespan and of course, like any system, they sometimes break before that time. Together with other anomalies and the potential for launches to go wrong, companies probably have to replace satellites before their expected life span, which alone could total $1 billion to $2 billion each year. Other costs include the materials and operating costs for the ground segment too, meaning a substantial upfront investment is needed if companies want to make headway.

Unsurprisingly, a few companies have encountered financial issues as they sought to develop large LEO constellations. We saw this happen with OneWeb, which filed for bankruptcy after reportedly running out of money and failing to secure additional funding. If it wasn’t for the UK government taking a significant step, the company probably would not have survived. Other companies, like LeoSat, were not fortunate enough to find financial backing and were forced to cease operations as a result.

Especially in the last few years, we have seen this space rapidly mature and evolve. As the global pandemic showed us how important connectivity is for our daily lives and our safety, many eyes have turned to the skies, hoping for these satellite networks to fill in the blanks. While I am certain that these will deliver, we still have quite the way to go and many challenges to overcome. Technical advances and innovations mean that the sky’s, quite literally, no longer the limit, but we need to manage expectations and understand the limitations of these systems. Satellite systems can and will fill some of the gaps, but don’t expect them to replace ground-based systems and cable-based connectivity anytime soon.

Marco Hogewoning is Manager of Public Policy and Internet Governance at the RIPE NCC. As part of the External Relations department, he helps lead the RIPE NCC’s engagement with membership, the RIPE community, government, law enforcement and other Internet stakeholders.