A bold plan to blanket the world with high-speed satellite internet may not be as crazy as it sounds and could be a license to print money, according to a leading internet networks expert.
Low latency routing in space, in which a swarm of satellites in low Earth orbit beam down wireless super-fast internet, sounds like more effort and expense than it's worth. But the idea has been heavily explored by Silicon Valley in recent years – and one man in particular wants to make it a reality. And by now you know him well.
He is controversial billionaire Elon Musk, whose private rocket company, SpaceX, wants to build a constellation of communication satellites as part of a project dubbed Starlink.
The company was granted approval on Friday by the US Federal Communications Committee (FCC) to send an extra 7518 satellites into space as part of the ambitious plan, on top of the 4400 already approved.
The main components of such a project have been done before but certainly not to the extent that SpaceX would need for Starlink to be successful.
Professor Mark Handley from the Department of Computer Science at University College London is an expert in network topology and recently set out to create a simulation of how Starlink might work.
"The devil is in the detail, and SpaceX seems to be pushing the bounds of what has been done before on several fronts simultaneously," he told news.com.au. But he thinks the project is feasible.
Prof Handley has pored over the company's FCC submissions to get a rough idea of what SpaceX hopes to do. Most interestingly, the company will likely rely on lasers instead of radio waves to shoot messages between satellites as it has not requested any radio spectrum for satellite-to-satellite communication.
"Mostly we inferred that from the omission of any radio frequencies for inter-satellite communication, and the discussion of certain optical communications components that may survive re-entry," Prof Handley said. "This has subsequently been confirmed in additional FCC communications, but we still do not know precisely how they plan to use laser links to connect the satellites together."
The video below shows what it might look like. Prof Handley said he used some educated guesswork and basic physics to "fill in the gaps" of what might be possible for SpaceX.
ELON'S BIG WEAPON
SpaceX has some of the most advanced rocket technology in the world and its pioneering reusable rockets will prove crucial to the Starlink plan. The rocket system allows the boosters, which are usually discarded after a single use, to land safely back on land and be reused for other launches.
"Without that, it's hard to see how this would have been viable," Prof Handley said. "It's important to understand that they're not just building this once. The satellites only have a lifespan of five to seven years, so they're looking at needing to launch an average of two satellites per day, on an ongoing basis.
"They can likely get around 25 to 30 satellites on one rocket though, and considerably more if their next generation BFR rocket works out, so that's not as crazy as it might at first seem."
It's clear that SpaceX, which works closely with NASA, has the backing of the US government.
"I'm excited to see what these services might promise and what these proposed constellations have to offer," FCC chairman Ajit Pai said on Friday after SpaceX was granted approval to launch more satellites, provided it goes ahead with its plan.
LICENSE TO PRINT MONEY
Aside from providing internet to just about every corner of the globe, a network like this provides one major benefit – it has the potential to significantly lower latency for long distance communications. That's because free-space lasers communicate at the speed of light in a vacuum, which is faster than the speed of light through glass like what is used in fibre optical cables on the ground.
And according to Prof Handley, there lies the potential genius.
He believes something like Starlink could be hugely appealing to high-frequency traders at big banks who may be willing to fork out for a speed advantage when it comes to algorithm-based trading on the stock market and currency interchanges.
It might sound like a foreign concept, but being able to shave milliseconds off your latency can translate into big bucks for these firms, which look for an edge to be able to respond to the market quicker than others.
Michael Lewis's 2014 book Flash Boys Chronicled the rise of high-frequency trading and starts out describing a $ US300 million project by Spread Networks – the construction of a 1331km cable that cuts straight through mountains and rivers from Chicago to New Jersey – with the sole goal of reducing the transmission time for data from 17 to 13 milliseconds.
Theoretically, SpaceX could charge high premiums to access its super-fast Starlink network.
"I think it's the low latency benefit that will be what makes the most money, and its use by the finance industry will probably pay a lot of the bills," Prof Handley said.
"I think the societal benefits from connecting remote places will be huge, and they'll contribute revenue, but if it was only about connecting remote places, I do not think Starlink could possibly pay for itself."
Prof Handley was in the US this week presenting his research paper and work on the Starlink simulation at a conference in Seattle.
"Perhaps surprisingly, not that many networking researchers know about the Starlink plans," he said.
"This is not just going to be the existing internet placed in space – the rapidly changing nature of the satellite paths poses all sorts of interesting networking research questions, and will no doubt keep us networking researchers busy for many years."
Ultimately, he thinks such a network is inevitable. But whether SpaceX is able to pull it off in the coming years remains to be seen.
Like everything else Musk has done lately, there'll be no shortage of people watching.