A telecommunications question.

[Kerbal01]

How does sattelite communications work? Specifically the ground transmission and sattelite recieving and re-transmission, also what governs what the dishes are made of and shaped like. My (limited) understanding is that the ground station sends a signal (some band of radiation?) to a sattelite in some sort of orbit (LEO,MEO,GEO,SSO,Molynia etc..). This sattelite then sends the signal back to a different ground station. How accurate is this description?

[Laie]

Found a nice primer; rather than quote it, here's the link: http://www.daviddarling.info/encyclopedia/C/communications_satellite.html

[TythosEternal]

Satellite communication (SATCOM) can take many forms, but the most common description is a space-based microwave-band transponder (or network of transponders) that serves to relay information between two or more users (who could be on the ground, at sea, in the air, in space, etc.).

Beyond this, there is a very diverse array of implementations. One of the simplest is a geostationary repeater, which simply receives, filters, and retransmits any waveform it "hears" on a particular band. Users can then "bounce" a signal "off" of the repeater to transmit to other users outside their line-of-sight.

More complex SATCOM implementations can feature cross-band transmission, support for multiple services, cross-links to other satellite in the same constellation, and even (if you don't mind shelling out major cash) complete point-to-point coverage from a large network of low-earth satellites (a la Iridium).

SATCOM networks can be robust and flexible. Networks exist for the transmission of phone conversations, radio channels, television networks, and many more. Most leverage a range of microwave bands (Ku and Ka are two of the most common). Putting such communications in space has a number of advantages, including coverage and cost (believe it or not, a handful of communications satellites are frequently both much easier and much cheaper than, for example, laying down a new trans-oceanic fiber network).

Edited September 22, 2014 by TythosEternal
Typos, is course!

[sgt_flyer]

Regarding satellite dishes forms and materials, the key is mostly to have the dish's inside layer made of reflective material - then the shape is put into play, to concentrate those reflected waves on a single point, while allowing a bit of freedom regarding at which angle the waves hits the dish. (So you don't need absolute precision - just being pointed in the direction of the emitter is enough. The receiver is placed at the collection point. (One widely used shape is the parabollic shape for this reason)

for satellites on orbits different than geostationnary, you might have to turn the antennas on the ground to track the satellites - while a satellite in a constellation of sats can be spin stabilised, use reaction wheels, or use gravity gradient to keep their relative orientation from each other - so their antennas keep pointing in the good direction.)

[cantab]

Usually one frequency/channel/band is used for Earth-satellite, and another is used for satellite-Earth. Then the satellite can simply retransmit everything as it recieves it. Satellites use various types on antenna but I don't think they often have tight-beam dishes like the ground stations.

For satellites not in GSO the system needs to handle Doppler shift of the transmissions.

[Idobox]

A lot has already been said, but I want to add a few details.

Satellites communicate by radio, usually in the UHF or microwave frequencies (300MHz to 60GHz for civilian applications, higher is too expensive and restricted).

The higher the frequency, the easier it is to focus your beam, so very high frequencies are often used for the uplink (Earth to space).

An important characteristic of an antenna is its gain. An antenna with a low gain, for example the ones in your phone, will receive and emit roughly the same in every direction. An antenna with high gain will send all the energy in a small cone, and amplify signals coming from there too. That's why antennas are usually very high gain, both on Earth (the satellite is a small target, power is expensive), and on the satellite (Earth is also a small target from GEO)

Dish antennas used to be very popular because they are quite easy to build. They work on the same principle as a telescope, using a parabolic reflector to concentrate the signal. There are a few variants, like the Cassegrain that get better gain (more powerful) in the same size. To reflect radiowaves, you need anything conductive, for satellites its often silver plated aluminium or titanium, because they are very light, and silver is the best metallic conductor there is. Earth based dishes are usually made of steel because it is cheaper, but high end stuff can be made of copper, possibly silver or gold plated.

An common alternative on modern satellites is antenna arrays: a bunch of small antennas all connected to each other. They have performances comparable to dishes of the same area, but are flatter and sturdier. It is also easier to tweak them to beams of arbitrary shapes, but it's more difficult to get good performances over a wide bandwidth. There are many ways to build this type of antenna, but the easier one is to use a printed circuit board.

A common reason to use a beam with a weird shape is to send signal in one place and not another. For example, for a TV satellite, you don't want to waste your power sending a signal in the middle of the ocean, or to a country where you don't plan to sell your services. To do that, you need to use each antenna with a specific phase, that you find using magic (math irl but beyond the scope of this post).

Then you have modulation/demodulation, encryption and all sorts of other details. The big question is how to encode information on a radiowave, and there are lots of ways to do it, with different advantages and issues.