SpaceX Low Latency Starlink Satellite Network Will Be Massively Profitable

by

SpaceX has US Federal Communications Committee (FCC) approval to build a constellation of 4,425 low Earth orbit communication satellites. It will use phased array antennas for up and downlinks and laser communication between
satellites to provide global low-latency high bandwidth coverage.

Mark Handley, University College London built a simulator based on public details from the FCC
filings to understand the latency properties of the network.

They evaluate how to use the laser links to provide a network and look at the problem of routing on this network. They conclude the SpaceX Starlink network can provide lower latency communications than any possible terrestrial optical fiber network for communications over distances greater than about 3000 kilometers.

Delay is Not an Option: Low Latency Routing in Space

As network bandwidths have increased, latency has emerged as being the limiting factor for many networked systems, ranging from the extremes of high-frequency trading, to the more mundane effects of latency on VoIP, online gaming, and web performance. Fundamentally, once traffic engineering has mitigated congestion and buffer bloat has been addressed, for wide-area traffic the remaining problem is that the speed of light in glass simply isn’t fast enough.

In Starlink’s initial phase, 1,600 satellites in 1,150 km altitude orbits will provide connectivity to all except far north and south regions of the world. A second phase adds another 2,825 satellites in orbits ranging from 1,100 km altitude to 1325 km, increasing density of coverage at lower latitudes and providing coverage at least as far as 70 degrees North. Finally, in an additional FCC filing SpaceX proposes launching an additional 7,518 satellites in approximately 340 km VLEO orbits. Mark only examined the LEO constellation.

The inverse square law suggests that received power on Starlink could be as much as 2000 times greater than on the EDRS (European Data Relay System). In 2014, the European Data Relay System (EDRS) achieved 1.8 Gb/s from LEO to geostationary earth orbit (GEO), across a distance of 45,000 km. Most Starlink distances will be 1000 km or less.

It seems probable that free-space laser link speeds of 100 Gb/s or higher will be possible.

A dense LEO constellation like Starlink has two main advantages over terrestrial networks. First, it can connect almost anywhere, however remote. Second, the speed of light in a vacuum os 47% higher than in optical fiber. The ability to connect anywhere is important, but we speculate that providing low-latency wide area communication will be where the money to maintain and operate such a network is made, connecting cities that are already well connected using optical fiber, but with lower latency as a premium service.

Already there are new private microwave relay links between New York and Chicago, London and Frankfurt, and London and Paris. These links have relatively low capacity compared to fiber, but are of high enough value to the finance industry to be worth building new low latency links.

Simulations show dense LEO constellations have very many paths available, and many of them are of similar latency. This allows groundstations to be much more conservative about when they move traffic back to the lowest delay path, using timescales much longer than the latency of the broadcast load reports, so avoiding instability. This is an interesting direction for future routing work on dense LEO constellations.

59 thoughts on “SpaceX Low Latency Starlink Satellite Network Will Be Massively Profitable”

  2. The “RTT” abbreviation in the CHART is not defined anywhere. I hate to troll, but as we learned in 5th grade science, abbreviations should be defined on first use, and CHARTS should also have all of their axes labelled. These figures are pig-slop as far as I’m concerned, but it’s still better labelled etc. than the “exhibits” commonly found in finance, so you got them beat, Brian Wang, congrats.

  3. Boca Chica? You can’t get to the inclinations you need from there. You’d wind up launching straight up the Tennessee Valley to get to 53 degrees, or over heavily populated parts of the Yucatan and Central America.What are the “plans for another” launch complex? They’ve got two pads at Canaveral, one at Vandenberg, and the BC stuff in the early stages. Am I missing something? More importantly, am I missing something that can make a contribution to launch capacity before 1Q2024?

  4. Boca Chica? You can’t get to the inclinations you need from there. You’d wind up launching straight up the Tennessee Valley to get to 53 degrees, or over heavily populated parts of the Yucatan and Central America.

    What are the “plans for another” launch complex? They’ve got two pads at Canaveral, one at Vandenberg, and the BC stuff in the early stages. Am I missing something? More importantly, am I missing something that can make a contribution to launch capacity before 1Q2024?

  5. You do realize that they have almost completed an entirely new launch complex and have plans for another? It was never possibly to do the number of launches they wanted from existing infrastructure and if it wasn’t for their earlier mishaps the new launch complex might have already been in use.

  6. Communism would be a highly evolved form of socialism and socialism is based on the society being so democratic that people are in direct ( the state doesn’t own it) ownership of the factories and companies they work for. Venezuela is being avoided because it’s a capitalist hell hole with a media so free that it can ask the USA/CIA to help overthrow their government and not be taken off the air. North Korea is of course just a state capitalist dictatorship so again not enough democracy around those part to be called anything with an S or C in it. I wish people would not believe authoritarian and capitalist propaganda and would read what the original modern socialist thinkers had to say and hoped to achieve.

  7. China is a authoritarian state capitalist society and Cuba and the USSR was the same. The idea that Venezuela or North Korea are socialist are equally fanciful and quite proves the point that capitalism and it’s propaganda systems rules the world. I suppose it does help that the dictators in all or many of these countries claimed to be socialist to help hide their crimes but at least the people living there were not as easily fooled as foreigners. So in closing socialism is an impossibility without democracy and no one has or is arguing that China or North Korea is or that the USSR used to be.

  8. “…but the low latency priority on StarLink may be a differentiator.”It’s not a differentiator if all your competitors have access to it. Then it’s just something that everybody has to have to do business. That’s not good news for the HFT firms–but it is for SpaceX.

  9. “…but the low latency priority on StarLink may be a differentiator.”

    It’s not a differentiator if all your competitors have access to it. Then it’s just something that everybody has to have to do business. That’s not good news for the HFT firms–but it is for SpaceX.

  10. Space force can fuck right off though. We don’t need weapons in space, it should be kept purely peaceful. Especially if it’s only one nation with a huge arsenal.

  11. He’s said publicly that he doesn’t want to have an IPO for SpaceX before a Mars supply route is established, because that would mean having other shareholders, and a distortion of his goals based purely on profit.

  12. Space force can fuck right off though. We don’t need weapons in space, it should be kept purely peaceful. Especially if it’s only one nation with a huge arsenal.

  13. He’s said publicly that he doesn’t want to have an IPO for SpaceX before a Mars supply route is established, because that would mean having other shareholders, and a distortion of his goals based purely on profit.

  14. The hardware becomes easier/cheaper, leveling the playing field versus shortwave HFT trading, but the low latency priority on StarLink may be a differentiator. Traders paying for priority QoS could make mad money for SpaceX. HFT is well known for paying stupid amounts of money for marginal speedups. About the only way they could tighten it up would be putting those StarLink flat antennas on the roof of exchange datacenters (which could fit in the center of the roof areas, rather than roof edge locations for beaming microwaves horizontal to other regional datacenters), and that’s a limited amount of real estate area which the exchange could charge a pretty penny for. This, along with perhaps rights/leases/exemptions for roof access by existing HFT link equipment effectively grandfathering space claims would make for additional exclusivity.

  16. Of course, the more the low-latency tech is commoditized, the less competitive advantage the HFT guys can acquire. Weirdo tricks off the ionosphere are incredibly cumbersome, but that’s a feature, not a bug, if your competitor can’t do it. With Starlink, the playing field is level. HFT’s **hate** level playing fields.

  17. I’m still trying to figure sending URLs in Vuukle, but there’s a SpaceNews article titled “Satellite antenna companies divided on near-term feasibility of cheap flat panels”, from a couple of weeks ago. Let’s just say that there’s a certain amount of skepticism about that price.

  18. The hardware becomes easier/cheaper, leveling the playing field versus shortwave HFT trading, but the low latency priority on StarLink may be a differentiator. Traders paying for priority QoS could make mad money for SpaceX. HFT is well known for paying stupid amounts of money for marginal speedups. About the only way they could tighten it up would be putting those StarLink flat antennas on the roof of exchange datacenters (which could fit in the center of the roof areas, rather than roof edge locations for beaming microwaves horizontal to other regional datacenters), and that’s a limited amount of real estate area which the exchange could charge a pretty penny for. This, along with perhaps rights/leases/exemptions for roof access by existing HFT link equipment effectively grandfathering space claims would make for additional exclusivity.

  19. Just an update, after thinking about it for a while and listening to other commentary:What I have above seems like it’s probably way too grandiose. A lot of the speculation elsewhere is that this will be as simple a mod to an existing S2 as they can get away with. Likely it’ll be just some aerosurfaces and TPS for a test article. Some think that it’ll be on a flight-qualified S2, so they can still launch a payload on the mission. Others think it’ll be a dedicated test launch.Either way, it seems unlikely that this thing will be big enough to accommodate 52 Starlinks.

  20. Pretty cool stuff. Never heard of these laser links in connection with Starlink before so I did some digging:There don’t seem to be many companies offering this technology but a company called Mynaric just published that it has signed an MoU regarding supply of laser communication links for an undisclosed satellite constellation. Maybe there is a link? Any thoughts?

  22. IPO? I’m pretty sure Musk regrets the fact Tesla is publicly owned. Also – if he is sincere in his plan to colonize Mars then SpaceX will NEVER offer an IPO.

  23. No mor0n, people avoid Venezuela and NK exactly because they are communists hellholes.And the Chinese still emigrate to the US if they can, because China is run by totalitarians.

  25. Of course, the more the low-latency tech is commoditized, the less competitive advantage the HFT guys can acquire. Weirdo tricks off the ionosphere are incredibly cumbersome, but that’s a feature, not a bug, if your competitor can’t do it. With Starlink, the playing field is level. HFT’s **hate** level playing fields.

  26. I’m still trying to figure sending URLs in Vuukle, but there’s a SpaceNews article titled “Satellite antenna companies divided on near-term feasibility of cheap flat panels”, from a couple of weeks ago. Let’s just say that there’s a certain amount of skepticism about that price.

  27. 1. Launch the network2. Make it clear that this isn’t the internet, just that it can carry internet traffic as well as phone traffic3. Demand $1billion per annum from Netflix, Facebook, Amazon, Google to allow their traffic passage through the network4. Profit.

  28. Just an update, after thinking about it for a while and listening to other commentary:

    What I have above seems like it’s probably way too grandiose. A lot of the speculation elsewhere is that this will be as simple a mod to an existing S2 as they can get away with. Likely it’ll be just some aerosurfaces and TPS for a test article. Some think that it’ll be on a flight-qualified S2, so they can still launch a payload on the mission. Others think it’ll be a dedicated test launch.

    Either way, it seems unlikely that this thing will be big enough to accommodate 52 Starlinks.

  29. HFT groups apparently not only use private microwave relay links, some also rely on shortwave ionosphere bouncing for more “direct” long range comms. Some are still in use for transatlantic arbitrage reasons, but the variability of the ionosphere is problematic, especially at very long range using multiple bounces. If the up/cross/down links have very few nodes and process at low latency, HFT firms working exchanges on both sides of the Atlantic will pay a pretty penny for the privilege of using a fast lane on StarLink. search for”Shortwave Trading | Part I | The West Chicago Tower Mystery – Sniper In Mahwah & friends”for more information.

  32. SpaceX-Starlink Secret Plan:1. Goal: Create high-return, proprietary new business by building low-latency internet access to entire planet in support both stationary and mobile platforms at lowest system cost, lowest user equipment cost and lowest operating cost.2. Create a reusable, low-cost, launch capability that cannot be matched by any competitor, giving SpaceX-Starlink a 4-5X launch cost advantage. Plan for 60 and 120 spacecraft per launch, plus on-orbit spares. Second stage reusability could drop costs even more.3. Design and produce spacecraft in very high-volume to give 10X+ lower costs when compared to historic spacecraft manufacturing.4. Bring in Department of Defense (DoD) in support of Starlink’s inherent high-redundancy, high-survivability communications network.5. Use cost, vision and speed advantages – plus DoD – to convince FCC to allocated bandwidth to Starlink, despite competitive challenges. Commit to early deployment to support FCC rationale.6. Move with “SpaceX speed” to put initial network in place…then begin to harvest massive cashflow…and launch a SpaceX IPO.

  33. No mor0n, people avoid Venezuela and NK exactly because they are communists hellholes.

    And the Chinese still emigrate to the US if they can, because China is run by totalitarians.

  34. People don’t avoid Venezuela and NK because of socialism. They avoid them because American capitalism is trying to destroy them and, where possible, starve them into submission.One socialist country that is becoming more attractive is China. Sometime between 2020-2025 every Chinese will have a home, a job, plenty of food, education, safe streets, health and old age care.  On that day there will be more suicides and executions, and more homeless, poor, hungry and imprisoned people in America than in China.By 2021, 450,000,000 urban Chinese will have more net worth and disposable income than the average American, their mothers and infants will be less likely to die in childbirth, their children will graduate from high school three years ahead of–and outlive–American kids.Early in the movie “Crazy Rich Asians” a Chinese-Singaporean father admonishes his young kids to finish their dinner, saying, “Think of all the starving children in America.”That day is fast approaching.

  35. 1. Launch the network
    2. Make it clear that this isn’t the internet, just that it can carry internet traffic as well as phone traffic
    3. Demand $1billion per annum from Netflix, Facebook, Amazon, Google to allow their traffic passage through the network
    4. Profit.

  36. Likely both.There’s an open question concerning how expensive the ground stations will be. If I had to guess, they’ll mostly be selling ISP transit and backhaul to begin with, with expensive ground stations for early adopters or individuals who are really desperate and willing to pay the premium. But the cost of ground stations should drop fairly quickly.

  37. There’s a tweet out from Elon today that SpaceX plans to test a “mini-BFS” as a new second stage for the F9. This is potentially huge news for Starlink.Right now, both phases listed above were licensed in 3/2018. Per FCC rule, half the birds have to be on-station in 6 years, or the end of 1Q2024. Given that all the orbital planes are multiples of 25, that means that SpaceX has to launch 2225 satellites by then, or they’ll lose the license to launch the rest of the constellation.BFR won’t be ready in time. Falcon Heavy is fairing-limited. So it’s likely that F9 is going to be the platform of choice.Per the FCC application, the Starlink sats reenter in a deployed configuration that’s 4 x 1.8 x 1.2 m for the body (more for the solar panels). I’m betting that the stowed configuration will be 3 m long. If that’s the case, then you need to stretch the existing fairing by 4.5 m if you want to launch 26 birds (25 + 1 spare) at a time. That would be about 15 t to a 53-degree orbit @ 1150 km altitude, which seems doable with a reusable F9.But that’s 89 launches.Assuming they don’t start to deploy until 3Q2019, that means that they’re looking at about 5 launches per quarter–20 per year–in addition to the launches they provide their paying customers. Given that Canaveral hardly ever exceeds 20 launches/year and Vandenberg has handled about 9/year recently, and that SpaceX is sharing those ranges with ULA and, soon, Blue Origin, there’s a pretty serious problem here. No doubt range improvements will help somewhat, but it’s still really tight.In addition to the cadence problem, stretching the F9 fairing by 4.5 m is a lot for a vehicle that’s already longer and skinnier than most. I’m not sure that the second stage would take the bending moment for a fairing stretched by that much, during a small guidance deviation.But a mini-BFS might solve both the cadence and fairing problems.First, mounting a 5 m-diameter mini-BFS on top of the 3.66 m core seems doable, and should make the stage a lot stiffer. But second, they just might be able to launch 50 birds + 2 spares at a time, using a 3-stick-reusable Falcon Heavy.This requires a much, much, **much** larger fairing than exists today, but that might be possible with this mini-BFS. And it halves the number of launches, effectively only needing about 10 FH launches a year to meet the license requirements.

  39. HFT groups apparently not only use private microwave relay links, some also rely on shortwave ionosphere bouncing for more “direct” long range comms. Some are still in use for transatlantic arbitrage reasons, but the variability of the ionosphere is problematic, especially at very long range using multiple bounces. If the up/cross/down links have very few nodes and process at low latency, HFT firms working exchanges on both sides of the Atlantic will pay a pretty penny for the privilege of using a fast lane on StarLink.

    search for

    “Shortwave Trading | Part I | The West Chicago Tower Mystery – Sniper In Mahwah & friends”
    for more information.

  40. Don’t bother. Socialists pester about the evils of capitalism in America, but you’ll never see them fleeing to enjoy the safety and prosperity of Venezuela or North Korea.

  41. I agree. As a space exploration fan we are stoked. Even the US military is juicing the competition. Funding all sides. Space force, etc.

  42. SpaceX-Starlink Secret Plan:

    1. Goal: Create high-return, proprietary new business by building low-latency internet access to entire planet in support both stationary and mobile platforms at lowest system cost, lowest user equipment cost and lowest operating cost.

    2. Create a reusable, low-cost, launch capability that cannot be matched by any competitor, giving SpaceX-Starlink a 4-5X launch cost advantage. Plan for 60 and 120 spacecraft per launch, plus on-orbit spares. Second stage reusability could drop costs even more.

    3. Design and produce spacecraft in very high-volume to give 10X+ lower costs when compared to historic spacecraft manufacturing.

    4. Bring in Department of Defense (DoD) in support of Starlink’s inherent high-redundancy, high-survivability communications network.

    5. Use cost, vision and speed advantages – plus DoD – to convince FCC to allocated bandwidth to Starlink, despite competitive challenges. Commit to early deployment to support FCC rationale.

    6. Move with “SpaceX speed” to put initial network in place…then begin to harvest massive cashflow…and launch a SpaceX IPO.

  43. Its still unclear if Starlink will sell its services directly to the consumer or if they will sell bandwith to the local provider and he will do the last mile stuff. Correct me if I am wrong.

  45. I hope he is right about the need But I was one of those investors of Global Star which went bankrupt. I would like to see the transmit receiver systems use on earth. I would also like to see how this is going to work with billing taxes and international boundaries. Countries like China with strict limits on free speech etc.

  47. Boring company might be valued at multiple Billions if its projects keep going, SpaceX might be making 100+Billion by mid-late 2020s just from low latency networks and Tesla’s driving AI hardware is several years ahead of competition. Each of those is a massive business moat. If it all goes at this pace Musk might be in the race with Bezos for the first Trillionaire crown. Space fans cant really loose either, 2 space champions are doing fantastic financially; bodes well for the early space economy (fingers crossed 😉 )

  48. People don’t avoid Venezuela and NK because of socialism. They avoid them because American capitalism is trying to destroy them and, where possible, starve them into submission.

    One socialist country that is becoming more attractive is China. Sometime between 2020-2025 every Chinese will have a home, a job, plenty of food, education, safe streets, health and old age care.  On that day there will be more suicides and executions, and more homeless, poor, hungry and imprisoned people in America than in China.

    By 2021, 450,000,000 urban Chinese will have more net worth and disposable income than the average American, their mothers and infants will be less likely to die in childbirth, their children will graduate from high school three years ahead of–and outlive–American kids.

    Early in the movie “Crazy Rich Asians” a Chinese-Singaporean father admonishes his young kids to finish their dinner, saying, “Think of all the starving children in America.”

    That day is fast approaching.

  49. Likely both.

    There’s an open question concerning how expensive the ground stations will be. If I had to guess, they’ll mostly be selling ISP transit and backhaul to begin with, with expensive ground stations for early adopters or individuals who are really desperate and willing to pay the premium. But the cost of ground stations should drop fairly quickly.

  50. There’s a tweet out from Elon today that SpaceX plans to test a “mini-BFS” as a new second stage for the F9. This is potentially huge news for Starlink.

    Right now, both phases listed above were licensed in 3/2018. Per FCC rule, half the birds have to be on-station in 6 years, or the end of 1Q2024. Given that all the orbital planes are multiples of 25, that means that SpaceX has to launch 2225 satellites by then, or they’ll lose the license to launch the rest of the constellation.

    BFR won’t be ready in time. Falcon Heavy is fairing-limited. So it’s likely that F9 is going to be the platform of choice.

    Per the FCC application, the Starlink sats reenter in a deployed configuration that’s 4 x 1.8 x 1.2 m for the body (more for the solar panels). I’m betting that the stowed configuration will be 3 m long. If that’s the case, then you need to stretch the existing fairing by 4.5 m if you want to launch 26 birds (25 + 1 spare) at a time. That would be about 15 t to a 53-degree orbit @ 1150 km altitude, which seems doable with a reusable F9.

    But that’s 89 launches.

    Assuming they don’t start to deploy until 3Q2019, that means that they’re looking at about 5 launches per quarter–20 per year–in addition to the launches they provide their paying customers. Given that Canaveral hardly ever exceeds 20 launches/year and Vandenberg has handled about 9/year recently, and that SpaceX is sharing those ranges with ULA and, soon, Blue Origin, there’s a pretty serious problem here. No doubt range improvements will help somewhat, but it’s still really tight.

    In addition to the cadence problem, stretching the F9 fairing by 4.5 m is a lot for a vehicle that’s already longer and skinnier than most. I’m not sure that the second stage would take the bending moment for a fairing stretched by that much, during a small guidance deviation.

    But a mini-BFS might solve both the cadence and fairing problems.

    First, mounting a 5 m-diameter mini-BFS on top of the 3.66 m core seems doable, and should make the stage a lot stiffer. But second, they just might be able to launch 50 birds + 2 spares at a time, using a 3-stick-reusable Falcon Heavy.

    This requires a much, much, **much** larger fairing than exists today, but that might be possible with this mini-BFS. And it halves the number of launches, effectively only needing about 10 FH launches a year to meet the license requirements.

  52. Don’t bother. Socialists pester about the evils of capitalism in America, but you’ll never see them fleeing to enjoy the safety and prosperity of Venezuela or North Korea.

  54. Its still unclear if Starlink will sell its services directly to the consumer or if they will sell bandwith to the local provider and he will do the last mile stuff. Correct me if I am wrong.

  56. I hope he is right about the need But I was one of those investors of Global Star which went bankrupt. I would like to see the transmit receiver systems use on earth. I would also like to see how this is going to work with billing taxes and international boundaries. Countries like China with strict limits on free speech etc.

  58. Boring company might be valued at multiple Billions if its projects keep going, SpaceX might be making 100+Billion by mid-late 2020s just from low latency networks and Tesla’s driving AI hardware is several years ahead of competition.

    Each of those is a massive business moat. If it all goes at this pace Musk might be in the race with Bezos for the first Trillionaire crown.

    Space fans cant really loose either, 2 space champions are doing fantastic financially; bodes well for the early space economy (fingers crossed 😉 )

Comments are closed.