As for when... it will probably depend a lot on semantics, which particular Mars-bound hardware can be called "Starlink". They won't exactly be 2nd hand satellites from the LEO constellation, they'll be traditional mission-specific relay sats with incrementally more Starlink-like features and mission extensions.Take the "Starlink" semantics out of it, set a specific criteria for surface coverage or interplanetary bandwidth for the whole Mars-relay fleet. Or if you want, only the "SpaceX" Mars-relay fleet.
I think it is likely that there will eventually be a constellation of cheap satellites in low/medium Mars orbit, launched and operated by SpaceX, using technologies developed under the "Starlink" brand, providing (amongst other things) point-to-point ground communication services. It will probably be in place before or at roughly the same time as the first human presence on Mars.
Quote from: Robotbeat on 01/06/2024 04:22 pmPeople always like to make the claims about how hard stuff is without actually checking the numbers. Check the numbers!!!Check the conops.It's not just about the Delta V, it's also about the impulse and trajectory design.To insert at Mars with SEP, you need a trajectory that has a really low rate of closure with the planet because the thrust provided by the SEP thruster is so low. A conventional high-thrust insertion does all of its burn in minutes or tens of minutes, where the same delta-V from SEP takes days or weeks.Then there's the whole power and thermal issue from being at 1.4 to 1.6 AU, and not being able to close a telecom link with Earth using the existing bus.Starlink is not a Mars-capable bus without enough changes to make it an entirely new spacecraft.
People always like to make the claims about how hard stuff is without actually checking the numbers. Check the numbers!!!
Quote from: steveleach on 01/07/2024 11:25 amI think it is likely that there will eventually be a constellation of cheap satellites in low/medium Mars orbit, launched and operated by SpaceX, using technologies developed under the "Starlink" brand, providing (amongst other things) point-to-point ground communication services. It will probably be in place before or at roughly the same time as the first human presence on Mars.Probably before, I mean SpaceX could sell “Martian Starlink” services to probes from national space programs for the price of several million per year. The data rate will be far superior and much cheaper than setting up your own data relay system. When starship is flying and SpaceX is already building thousands of SL satellites per year it’s not like the cost would be huge… a few customers could probably pay the whole venture off.
Quote from: ZachF on 01/07/2024 02:33 pmQuote from: steveleach on 01/07/2024 11:25 amI think it is likely that there will eventually be a constellation of cheap satellites in low/medium Mars orbit, launched and operated by SpaceX, using technologies developed under the "Starlink" brand, providing (amongst other things) point-to-point ground communication services. It will probably be in place before or at roughly the same time as the first human presence on Mars.Probably before, I mean SpaceX could sell “Martian Starlink” services to probes from national space programs for the price of several million per year. The data rate will be far superior and much cheaper than setting up your own data relay system. When starship is flying and SpaceX is already building thousands of SL satellites per year it’s not like the cost would be huge… a few customers could probably pay the whole venture off.I think the initial constellation will be multi-purpose, serving as both GPS and comms. The Earth GPS system uses 24 satellites (plus spares) in MEO orbits and provides global coverage. For Mars, each satellite would only provide at most the same throughput as one Starlink satellite, but because of the sparseness of customers, the bandwidth per customer would be higher than here on Earth. The satellites might also have some imaging capabilities.
Quote from: DanClemmensen on 01/07/2024 03:05 pmQuote from: ZachF on 01/07/2024 02:33 pmQuote from: steveleach on 01/07/2024 11:25 amI think it is likely that there will eventually be a constellation of cheap satellites in low/medium Mars orbit, launched and operated by SpaceX, using technologies developed under the "Starlink" brand, providing (amongst other things) point-to-point ground communication services. It will probably be in place before or at roughly the same time as the first human presence on Mars.Probably before, I mean SpaceX could sell “Martian Starlink” services to probes from national space programs for the price of several million per year. The data rate will be far superior and much cheaper than setting up your own data relay system. When starship is flying and SpaceX is already building thousands of SL satellites per year it’s not like the cost would be huge… a few customers could probably pay the whole venture off.I think the initial constellation will be multi-purpose, serving as both GPS and comms. The Earth GPS system uses 24 satellites (plus spares) in MEO orbits and provides global coverage. For Mars, each satellite would only provide at most the same throughput as one Starlink satellite, but because of the sparseness of customers, the bandwidth per customer would be higher than here on Earth. The satellites might also have some imaging capabilities.I think there will be a Mars GPS (Martian Positioning System - MPS?), but I think it will be a little later. and use different satellites.GPS needs both satellites and ground stations (with extremely known location and elevation) This wont be practical until you have at least 3 landers (or people on the ground) to do that. 6 MPS Satellites in high orbits would probably be sufficient for Mars. Starlink (on Mars) would need ~50 satellites for a global coverage.If however they main starlink sats are the same as for Earth, then the high powered relays could also act as the MPS sats.
Quote from: waveney on 01/07/2024 04:00 pmQuote from: DanClemmensen on 01/07/2024 03:05 pmQuote from: ZachF on 01/07/2024 02:33 pmQuote from: steveleach on 01/07/2024 11:25 amI think it is likely that there will eventually be a constellation of cheap satellites in low/medium Mars orbit, launched and operated by SpaceX, using technologies developed under the "Starlink" brand, providing (amongst other things) point-to-point ground communication services. It will probably be in place before or at roughly the same time as the first human presence on Mars.Probably before, I mean SpaceX could sell “Martian Starlink” services to probes from national space programs for the price of several million per year. The data rate will be far superior and much cheaper than setting up your own data relay system. When starship is flying and SpaceX is already building thousands of SL satellites per year it’s not like the cost would be huge… a few customers could probably pay the whole venture off.I think the initial constellation will be multi-purpose, serving as both GPS and comms. The Earth GPS system uses 24 satellites (plus spares) in MEO orbits and provides global coverage. For Mars, each satellite would only provide at most the same throughput as one Starlink satellite, but because of the sparseness of customers, the bandwidth per customer would be higher than here on Earth. The satellites might also have some imaging capabilities.I think there will be a Mars GPS (Martian Positioning System - MPS?), but I think it will be a little later. and use different satellites.GPS needs both satellites and ground stations (with extremely known location and elevation) This wont be practical until you have at least 3 landers (or people on the ground) to do that. 6 MPS Satellites in high orbits would probably be sufficient for Mars. Starlink (on Mars) would need ~50 satellites for a global coverage.If however they main starlink sats are the same as for Earth, then the high powered relays could also act as the MPS sats.Starlink is not needed until there are multiple ground locations, and they can serve as ground station locations. Starlink requires precise location information, basically the same as GPS, to operate. Today, that information comes from GPS receivers embedded in the user equipment. After a user station is already in the net, it could use the net instead of GPS, but why?I actually have a patent on deployable ground stations, not user stations, for a constellation. I'm not happy with that patent since IMO the alleged co-inventor and the patent attorney messed it up. However, the reading I did as part of this "invention" means I learned a bit about the requirements. And yes, the crew that deploys a ground station uses GPS.
As someone with ~35 patents to my name (all telecoms)...Starlink wouldn't need GPS on Mars. On Mars you don't start with territories and the need to treat different users differently. I can see merit in having both integrated in some way. Martian GPS would I think only be of use once explorers and rovers start covering serious distances on Mars, it would be a complete overkill for the first people there.Starlink would be of use as soon as it is deployed to handle traffic from the existing infrastructure there. For example: MRO (HiRise) is heavily constrained as to how much it can observe each day, to keep it within its ability to send the results back.
Quote from: waveney on 01/07/2024 05:18 pmAs someone with ~35 patents to my name (all telecoms)...Starlink wouldn't need GPS on Mars. On Mars you don't start with territories and the need to treat different users differently. I can see merit in having both integrated in some way. Martian GPS would I think only be of use once explorers and rovers start covering serious distances on Mars, it would be a complete overkill for the first people there.Starlink would be of use as soon as it is deployed to handle traffic from the existing infrastructure there. For example: MRO (HiRise) is heavily constrained as to how much it can observe each day, to keep it within its ability to send the results back.GPS is not about geopolitical boundaries, although it is also used for that.I spent about 40 years as a systems programmer and system architect in fixed communications. I then shifted to GEO satcomms for 9 years before shifting to LEO satcomms.The biggest change from fixed comms is that satcomms uses TDMA. TDMA requires extremely tight control in the time domain to keep two transmissions from stepping on each other at the satellite. The speed of light is 300 meters per microsecond. As the distance from sender to satellite varies, the sender must vary the time offset so the signal reaches the satellite at it's allocated time. For GEO this is (almost) a fixed offset for each terminal location. For LEO, this offset is adjusted by feedback continuously after the terminal is in the net. However, to enter the net or switch to the next satellite the terminal must know where it is to a close approximation, which is why the terminals have GPS.
Starlink wouldn't need GPS on Mars. On Mars you don't start with territories and the need to treat different users differently.
Mind you, a GPS satellite at its core is a atom clock (those can be made remarkably small nowadays) that is broadcasting its local time and position. That would not necessarily constitute a trivial modification to Starlink satellites (on top of the bigger solar PV panels), of course.
Quote from: whvholst on 01/08/2024 06:13 pmMind you, a GPS satellite at its core is a atom clock (those can be made remarkably small nowadays) that is broadcasting its local time and position. That would not necessarily constitute a trivial modification to Starlink satellites (on top of the bigger solar PV panels), of course.I would be very surprised if the Starlink satellites do not have highly accurate ("atomic") clocks, probably rubidium clocks. These should be good enough if they add the sophisticated corrections based on a master clock at the central base. The true magic of GPS is the waveform that encodes the data such that inexpensive receivers can use it. All of the satellites transmit continuously on the same frequency, using extreme spread codes so a receiver listening on the frequency can decode all of the satellites simultaneously from the same channel.
Quote from: DanClemmensen on 01/08/2024 06:51 pmQuote from: whvholst on 01/08/2024 06:13 pmMind you, a GPS satellite at its core is a atom clock (those can be made remarkably small nowadays) that is broadcasting its local time and position. That would not necessarily constitute a trivial modification to Starlink satellites (on top of the bigger solar PV panels), of course.I would be very surprised if the Starlink satellites do not have highly accurate ("atomic") clocks, probably rubidium clocks. These should be good enough if they add the sophisticated corrections based on a master clock at the central base. The true magic of GPS is the waveform that encodes the data such that inexpensive receivers can use it. All of the satellites transmit continuously on the same frequency, using extreme spread codes so a receiver listening on the frequency can decode all of the satellites simultaneously from the same channel.Moreover, there have been a number of research papers published recently which demonstrate how to use existing Starlink signals for positioning even without the cooperation of SpaceX. A direct-sequence spread spectrum beacon with a long enough spreading code is a clock signal; combine a few of them with information about the position and trajectory of the transmitters and you can start making position estimates.One such paper I found recently: Nabil Jardak, Ronan Adam: "Practical Use of Starlink Downlink Tones for Positioning"https://www.mdpi.com/1424-8220/23/6/3234With help from SpaceX -- particularly around improving estimates of the satellite orbits -- they could do better.(that said, returning to the main topic of this post, I think it's going to be a while before we see some form of Starlink in orbit around Mars; there are many other more important things to set up first).
Quote from: launchwatcher on 01/09/2024 04:07 pmQuote from: DanClemmensen on 01/08/2024 06:51 pmQuote from: whvholst on 01/08/2024 06:13 pmMind you, a GPS satellite at its core is a atom clock (those can be made remarkably small nowadays) that is broadcasting its local time and position. That would not necessarily constitute a trivial modification to Starlink satellites (on top of the bigger solar PV panels), of course.I would be very surprised if the Starlink satellites do not have highly accurate ("atomic") clocks, probably rubidium clocks. These should be good enough if they add the sophisticated corrections based on a master clock at the central base. The true magic of GPS is the waveform that encodes the data such that inexpensive receivers can use it. All of the satellites transmit continuously on the same frequency, using extreme spread codes so a receiver listening on the frequency can decode all of the satellites simultaneously from the same channel.Moreover, there have been a number of research papers published recently which demonstrate how to use existing Starlink signals for positioning even without the cooperation of SpaceX. A direct-sequence spread spectrum beacon with a long enough spreading code is a clock signal; combine a few of them with information about the position and trajectory of the transmitters and you can start making position estimates.One such paper I found recently: Nabil Jardak, Ronan Adam: "Practical Use of Starlink Downlink Tones for Positioning"https://www.mdpi.com/1424-8220/23/6/3234With help from SpaceX -- particularly around improving estimates of the satellite orbits -- they could do better.(that said, returning to the main topic of this post, I think it's going to be a while before we see some form of Starlink in orbit around Mars; there are many other more important things to set up first).You can in fact use the Starlink signals for positioning and timing, but that's because the Starlink satellties themselves know their own positions and time with extreme accuracy, very like GPS satellites. However, it's hard to begin listening to a Starlink satellite in the first place unless you know where it is and where you are, because your Starlink antenna is directional in order to get the gain it needs and because the satellite's transmission is directional to get the gain it needs. You can only receive signal from a satellite that is transmitting in your general direction and that your antenna is looking at. By contrast, a GPS satellite transmission is omnidirectional, and your little GPS receiver's antenna is also omnidirectional. This means you receive from every GPS satellite in line of sight in any direction, but with an extremely weak signal. That's where the magic extreme spread codes come in. They allow GPS to send a relatively small amount of data over this extremely weak link.
Nabil Jardak, Ronan Adam: "Practical Use of Starlink Downlink Tones for Positioning"https://www.mdpi.com/1424-8220/23/6/3234With help from SpaceX -- particularly around improving estimates of the satellite orbits -- they could do better.
Therefore, the frequency error state has captured most of the satellites’ orbit and clock errors, preventingtheir total propagation into the user position state. The remaining position error is mainlydue to measurement noise combined with weak satellite geometry.
