Why is the power of a satellite phone not 100 kilowatts, but 2-5 watts?

Question

Space

Example:
For the operation of a conventional cell phone, in a cell with a radius of 35 km, as you know, you need about 100 milliwatts of power.
Knowing that satellite phones "finish off" to satellites in geostationary orbits (35,000 km), and that the power varies inversely with the square of the distance , it follows that the
power of a satellite phone should be about 100 kilowatts , instead of 2-5 watts as it is on really.
So, why does a satellite phone have such a low power?
Small calculation:
35,000,000 (m) / 35,000 (m) = 1000 the difference between the distances of the cell and the satellite,
then we square 1000 and multiply by 0.1 watts (the power of the phone required to "break through" 35 km)
we get 100 kilowatts (100,000 watts) of required power for a satellite phone.
There are a couple of hints:
1) For space communication, the range from 1.5 to 30 centimeters is optimal, all other waves have problems with the atmosphere. Outside the atmosphere of the ionosphere, there is almost no attenuation.
2) The earth
is round 3) At least 2 reasons

Answer 1

I can be wrong, but I assume that:
1) There is always a direct line of sight between the satellite and the satellite phone, not always between the phone and the BS.
2) The range of 35 kilometers in GSM is determined not by the signal strength, but, by the EMNIP, by the delay in the propagation of radio waves, because there is a fixed duration of one bit and an interval between timeslots (some number of empty bits), and 35km is the distance that radio waves fly over the time of one timeslot + the time of these empty bits, so the power would be enough for a greater distance.
3) Geostationary satellites are not always used, some communication systems use low-flying satellites located at an altitude of 780 kilometers
4) Different frequencies: for a signal with a frequency of 10 GHz, the attenuation in the ionosphere is <0.0001 dB, for a signal of 1 GHz (GSM - 900 MHz, 1800 MHz) < 0.01 dB, that is, 100 times higher.

Answer 2

The main mistake is that cellular satellites do not fly in geostationary orbits. For example, the orbital height of the Irridium satellites is 780 km.

Answer 3

Well... There are several reasons.
Firstly, not so much transmission speed is required. The voice, of course, will be terrible, since the codec will capture a couple of hundred, thousands of hertz in frequencies, but that is what it is. EMNIP, there is something in the region of a couple of tens of kilobits per second, at most. However, a lot has already changed.
Secondly, as already mentioned above - direct visibility. This changes a lot, and centimeter waves calmly pass through the atmosphere. In addition, the satellites, of course, have good antennas of a couple of meters, the right antennas, so to speak.
Finally, a couple of watts is a lot. Modern methods of searching for a signal in noise are amazing, they are looking for it where there is literally nothing but noise.