I could use some math help. What I want to know is the minimum number of Starlink satellites needed so that a ground station always has one 550 km satellite in view at various latitudes with various numbers of planes partially populated.
Quote from: Joseph Peterson on 12/09/2018 10:42 pmI could use some math help. What I want to know is the minimum number of Starlink satellites needed so that a ground station always has one 550 km satellite in view at various latitudes with various numbers of planes partially populated.I can't give you a simple equation for that. I wrote a matlab script to figure it out (attached). You set some parameters: orbit altitude, number of rings, number of satellites per ring, and orbital inclination, and it generates plots like the one attached, which is for the following parameters. orbit altitude: 550 km orbit inclination: 53 and 37 degrees number of planes: 20 number of satellites per plane: 30This configuration gives continuous coverage from 25 to 56 degrees N, but has minimum coverage gaps around 10 N and 22 N.
Quote from: Joseph Peterson on 12/09/2018 10:42 pmI could use some math help. What I want to know is the minimum number of Starlink satellites needed so that a ground station always has one 550 km satellite in view at various latitudes with various numbers of planes partially populated.I can't give you a simple equation for that. I wrote a matlab script to figure it out (attached). You set some parameters: orbit altitude, number of rings, number of satellites per ring, and orbital inclination, and it generates plots like the one attached, which is for the following parameters. orbit altitude: 550 km orbit inclination: 53 and 37 degrees number of planes: 20 number of satellites per plane: 30This configuration gives continuous coverage from 25 to 56 degrees N, but has minimum coverage gaps around 10 N and 22 N.From there, you just have to fool around with the numbers a bit. This config is the minimum number of sats I came up with at 550 km altitude. The minimums are fairly sensitive to the relative phasing of the satellites within their rings, and there may be some optimization there I haven't figured out.
For those wanting an example of what RF inter-satellite links look like in FCC documents, here are a couple of constellation filings that have them:Iridium NEXT (SAT-MOD-20131227-00148)Audacy (SAT-LOA-20161115-00117)Look for Inter-Satellite Service or ISS beams in the documentation (both narrative and Schedule S). If you find similar entries in the SpaceX filings then please let me know, I must have overlooked them.
The above gives average sats visible - is it possible to simply plot minimum sats? (no matlab)
If you are assuming a mixed constellation of planes at different inclinations, it probably makes more sense at least initially to ask about coverage of a constellation at 550km, with only 53 degrees. In other words, when can a partially launched constellation be of use.
A back of the envelope calculation indicated that you can get a fair coverage down to approaching 30N or so with 53 degree inclination satellites, but with only 12 of the 24 nominal planes populated.Covering most of the population of the USA and Europe.
Look for Inter-Satellite Service or ISS beams in the documentation (both narrative and Schedule S). If you find similar entries in the SpaceX filings then please let me know, I must have overlooked them.
Quote from: speedevil on 12/11/2018 09:19 amThe above gives average sats visible - is it possible to simply plot minimum sats? (no matlab)Yup. The dashed line shows minimum sats visible, and the solid line is average sats visible.QuoteIf you are assuming a mixed constellation of planes at different inclinations, it probably makes more sense at least initially to ask about coverage of a constellation at 550km, with only 53 degrees. In other words, when can a partially launched constellation be of use.I tried that first. You have to choose between covering Europe with at least 1, covering the US with at least 1, or using a lot more satellites.550 km is a lot harder than 1050 km. At 1050 km the spots are so much bigger that the number of satellites to get min 1 coverage goes way down.QuoteA back of the envelope calculation indicated that you can get a fair coverage down to approaching 30N or so with 53 degree inclination satellites, but with only 12 of the 24 nominal planes populated.Covering most of the population of the USA and Europe.That's not what my Matlab script shows. At 12 planes you'd open up a lot of gaps at the equator that would continue up through at least San Francisco.
Could you run the Matlab script again for the higher orbital plane satellites? It will IIRC be the first one put in place, and for those of us at higher latitudes provides better coverage.
Space Exploration Technologies Corp., Hawthorne, California, has been awarded a $28,713,994 competitive, firm-fixed-price, other transaction agreement for experimentation per the advanced research announcement, FA8650-17-S-9300. This agreement allows for experimentation in the areas of establishing connectivity, operational experimentation, and special purpose experimentation. Experimentation will include connectivity demonstrations to Air Force ground sites and aircraft for experimental purposes. For the proposed Phase 2, the awardee proposes to perform experiments in two other key areas: early versions of a commercial space-to-space data relay service and mobile connectivity directly from space to aircraft. Work will be performed in Hawthorne, California, and is expected to be completed by June 18, 2021. Fiscal 2019 research, development, test and evaluation funds in the amount of $19,167,989 will be obligated at the time of award. Air Force Research Laboratory, Wright-Patterson Air Force Base, Ohio, is the contracting activity (FA8650-19-9-9320). (Awarded Dec. 19, 2018)
I'm not sure this news bit is about Starlink. But it is definitely not about rockets or launches.However, it's 29 millions - so should be some place for it here DoD Contracts for Dec. 19, 2018:
Quote from: smoliarm on 12/20/2018 04:09 pmI'm not sure this news bit is about Starlink. But it is definitely not about rockets or launches.However, it's 29 millions - so should be some place for it here DoD Contracts for Dec. 19, 2018:Thank you!!This ties into earlier reports of Air Force Research Lab interest and preliminary tests, FCC issuing permits for same, and the DARPA Blackjack program looking into using LEO constellation satellite buses for milsats (implied: wolves hiding among the sheep)
Quote from: docmordrid on 12/20/2018 05:08 pmQuote from: smoliarm on 12/20/2018 04:09 pmI'm not sure this news bit is about Starlink. But it is definitely not about rockets or launches.However, it's 29 millions - so should be some place for it here DoD Contracts for Dec. 19, 2018:Thank you!!This ties into earlier reports of Air Force Research Lab interest and preliminary tests, FCC issuing permits for same, and the DARPA Blackjack program looking into using LEO constellation satellite buses for milsats (implied: wolves hiding among the sheep)I don't think this is related to Blackjack, Blackjack is DARPA and more about using production line to build DoD's own satellites, this is Air Force Research Lab testing the bandwidth. So there is the potential for SpaceX to win more in Blackjack.