I suggest that you familiarize yourself with the previously posted materials on the Starlink (SL) project:
‣ Part 20. Internal structure of the SL terminal ‣ Part 21. SL and polarization problems ‣ Part 22. Problems of electromagnetic compatibility with other satellites. ‣ Part 23. Interim results of the RDOF auction ‣ Part 24. Laser Channels -2 ‣ Part 25. EPFD ‣ Part 26. First results. Part one - positive ‣ Part 27. First results. Part two - problematic
On April 16, 2021, SpaceX sent a bunch of applications to the FSS for different options for placing the terminal on moving objects, or in our opinion, this is SOTM (Communication n he Move). It should be noted that the ITU allocated a certain frequency range in the KU-band for the service, for example, our SCRF formulates this as follows:
“Allocate the radio frequency bands for use: 14000-14103 MHz, 14129-14149 MHz, 14175-14194 MHz, 14242-14500 MHz ( Earth-to-space) and 10950-11200 MHz, 11450-11700 MHz, 12500-12750 MHz (space-to-Earth) by subscriber earth stations of satellite communications (ESSS) operating as part of domestic satellite networks, and subscriber ESSS, ... and installed on mobile objects located within the territory of the Russian Federation for the purpose of organizing satellite communication networks (lines). "
I did not refer to the ITU tables, but, as a rule, the RF strictly follows its frequency distributions. I note that SpaceX as a whole follows the ITU rules when it comes to something flying like Starship.
Only 200 MHz out of a possible 500 ... By the way, note that another 30 MHz drop out due to the fact that in some places there are Astronomical Observatories, and there it is necessary to exclude operation at these frequencies. I would like to note that Starlink's subscriber-satellite channel width is 60 MHz, in total we have 8 60 MHz beams on the satellite, that is, in fact, a ban on 30 MHz is a ban on 1/8 of the capacity.
Also SpaceX for the first time says that for moving objects (cars) it creates a new type of terminal that will automatically control the levels of its radiation and, if exceeded, will automatically turn off within 100 ms.
Please note that there is one application for the entire range of 14..14.5 GHz, which corresponds to the range for terminals installed in houses, that is, fixed ones. I wonder if the FCC will go to meet SpaceX here or will strictly follow the ITU recommendations.
And finally, when will this happiness happen in the USA?
There is a cheerful tweet from Ilona
Like, you have to stand for another 3 months and hold out for three more ...
But let's try to believe his harmony with algebra.
So, in order to realize mobility, we need 100% coverage of the US territory with the service.
The territory of the continental United States without Alaska is 7.6 million km².
The area of the Starlink beam in nadir (diameter 24 km) is 450 km, but for a reserve we take the beam at an angle, the spot will stretch even 2 times, then the beam area becomes 1800 km², total coverage requires 4235 beams. Since there are 8 beams on a satellite, this is, respectively, 530 satellites.
"Bingo!" - the one who is not in the subject will say, but the one who is in the subject will scratch his turnip and say: "And the chain mail is a bit short!" - because the territory of the Earth between 53 parallels, according to my calculations, is 300 million square kilometers, that is, the share of satellites located over the United States is 7.66 / 300 = 2.5%
Well, let's assume that satellites flying over the ocean or Mexico, they can direct the beams sideways and up, and let it be as much as 4%. That is, we need to have 530/4% = 13250 satellites in the constellation ...
This is a different alignment. But, if my assumption, earlier expressed here, about "jumps" of one beam over different spots, is true, then if one beam serves 10 zones, 1320 satellites are enough, if 8 zones - then 1656 satellites.
That is, Musk's words about the service can be implemented this year.
But, let's understand, this means that ONE BEAM serves 10 zones with an area of 450 km² (sorry 1800 km²). The capacity of the beam on the line in the 250 MHz satellite-subscriber on a 3 bits / Hz (for the current signal / 9 dB noise), that is, 750 Mbps, and are divided in this case, all 10 areas ...
However, this is not heavenly delight, isn't it?
How can the situation be improved? Divide US zones / territory into permanent and temporary zones.
In permanent - there are subscribers who constantly consume the service, in temporary - only travelers, respectively, service in time zones is provided upon request, that is, when the terminal is turned on for the first time, it goes through a full cycle of installation and registration in the network and its zone where it is served by another models (less common than fixed zones). Given the presence of national parks and large desert mountain areas, this will allow more resource to be used for areas with a large number of subscribers.
An even tougher option is to refuse (maybe temporarily) the promise to provide service on 100% of the continental United States, but this is IMHO a bad idea.