Wonder how SpaceX’s StarLink can achieve record low latency?

Obviously, financial applications come to mind instantly …

With its (planned) network of up to 46,000 small satellites (eventually; 12,000 to start out with) in low orbit, and interconnectivity by laser-based optical communications in the vaccum of space, this infrastructure can beat any terrestrial communications in the sense of latency, and even speed.

For a quick explanation, watch the video below.

I have worked in this field as early as 1998 with Loral’s CyberStar, which had planned to do data transmissions the traditional way (no statellite-to-satellite interconnectivity), and from geo-synchronous satellites, which are much further out, thus have a much, much slower latency.

In the early 2000’s I worked with two-way satellite systems, avoiding the terrestrial uplink requirements, with an emphasis on CyberWarfare Defense.

This new low-orbit setup on such a massive scale (Starlink or OneWeb) promises to beat all the previous (and partially still operational systems, like Hughes EchoStar, ViaSat or Sky Muster) geo-synchronous approaches.

The videos below explain StarLink‘s setup pretty well.

How Does Starlink Work Anyway?

No matter what you think of Elon Musk, it’s hard to deny that he takes the dictum “There’s no such thing as bad publicity” to heart. From hurling sports cars into orbit to solar-powered roof destroyers, there’s little that Mr. Musk can’t turn into a net positive for at least one of his many ventures, not to mention his image.

Elon may have gotten in over his head, though. His plan to use his SpaceX rockets to fill the sky with thousands of satellites dedicated to providing cheap Internet access ran afoul of the astronomy community, which has decried the impact of the Starlink satellites on observations, both in the optical wavelengths and further down the spectrum in the radio bands. And that’s with only a tiny fraction of the planned constellation deployed; once fully built-out, they fear Starlink will ruin Earth-based observation forever.

What exactly the final Starlink constellation will look like and what impact it would have on observations depend greatly on the degree to which it can withstand regulatory efforts and market forces. Assuming it does survive and gets built out into a system that more or less resembles the current plan, what exactly will Starlink do? And more importantly, how will it accomplish its stated goals?

Smallsats and Pizza Boxes; Lasers and Krypton Gas Thrusters

For as small as the Starlink satellites are — in the “smallsat” class and weighing in at about 250 kg each — they’re packed with all sorts of fun stuff. As pointed out in the Real Engineering video below, each Starlink satellite is essentially a flying, solar-powered wireless router. Phased-array antennas on the Earth-facing side of the satellite will link to “user terminals,” the oft-described “pizza box” ground stations that will provide Internet services to groups on the ground. The satellite also has onboard Hall-effect krypton gas thrusters for station keeping and for the eventual de-orbit burn when the satellite passes its best-by date.

Perhaps the most interesting bit of tech onboard each satellite is a set of lasers. While none of the 180 or so Starlink satellites launched so far have been equipped with lasers, the intention is to use them for the all-important job of “backhaul” communications: the ability to link nearby satellites together optically to find a path between any two ground stations. This has significant throughput benefits over traditional terrestrial fiber-optic links, since the speed of light in glass is about half of that in a vacuum. In theory, Starlink connections have the potential to greatly reduce the latency that exists in terrestrial links. But of course the satellites need those lasers first, and they need to work.

The improved latency of Starlink is probably the key to understanding what Mr. Musk is trying to accomplish here. The ability to provide low-latency transcontinental connections could be incredibly lucrative, especially to the financial markets, where time is literally money. Given the lengths that high-frequency traders will go to shave a few milliseconds off a link, SpaceX could name their price for a reliable link that saves 30 milliseconds or more. Any of the other stated benefits of Starlink, like providing Internet access to underserved locations, will ride on the back of the waves of profit the service will unleash.

Let’s All Do the Hop to Hybrid Backhaul

But that leaves a question: what good is Starlink without the optical backhaul system? As mentioned, none of the satellites currently flying has lasers installed, so there’s no way to link them together directly. Can the constellation be used without the laser backhaul?

In a separate video, networking researcher Mark Handley brilliantly answers not only that question, but also provides us with a glimpse at how the Starlink network will probably work. He suggests that the laser backhaul could be replaced by hops between satellites and idle user terminals on the ground. Even with the overhead of switching and the increased distance compared to a direct laser connection between satellites, the connection would be profitably faster than a terrestrial fiber connection.

This scheme requires a decent number of user terminals to be effective, and they need to be fairly evenly distributed so that an efficient and low-latency path can be stitched together. This leaves the possibility that there will be some special offers made to people living in areas that need a ground station to fill in a hole in the network. It also leaves the problem of crossing the wide expanses of ocean, but as Dr. Handley points out, a ship or even a buoy anchored in the right place could serve the purpose.

Once Starlink satellites with lasers start to launch and the optical backhaul network begins to come online, will that spell the doom of the putative ground-based backhaul? Not necessarily, according to Dr. Handley. With some compelling simulations, he makes the case for using a combination of optical and ground relays to create a hybrid network that fills in the gaps better than an all-optical network. Hybrid backhaul seems to be especially beneficial for long north-south routes, and routes that pass over relatively unpopulated areas, such as northern Europe to southern Africa.

Granted, all of this is conjecture on the part of Dr. Handley, but given his background and the source material he’s working from, it all seems plausible. The details of the final Starlink system are likely to differ significantly from these simulations due to business imperatives, regulatory hurdles, and technical challenges as yet unknown.

But the whole architecture of Starlink is fascinating, and as much as we’re not looking forward to night skies littered with thousands of satellites, the technical achievements and engineering challenges of bringing such a system online are enough to hold our interest for a long time.