3

Some applications on Earth replace wires with radio communication, standard stuff like Bluetooth. It saves mass and maybe complexity.

Is this done for intra-satellite communication and deep space probes too, i.e. within a spacecraft as replacement for wires? And on the space station or other human space flight missions? Are there special problems making it less desirable up there?

LocalFluff
  • 26,981
  • 8
  • 82
  • 222
  • Well, maybe not Bluetooth, but it does seem that maybe some radio band could be reserved for this purpose. – called2voyage Oct 05 '16 at 20:42
  • Then I imagine that it probably wouldn't simplify things much, because each device would have to be equipped with a Bluetooth send/receive device. And each would have to be tuned to a certain frequency to limit noise both to other Bluetooth devices as well as the other sensors on the spacecraft. Ultimately, I would say that this would just be more points of failure without significant benefits. – Phiteros Oct 05 '16 at 20:48
  • @Phiteros I agree, but I suppose it could possibly have an incredibly limited range of applications. For example, a single wireless link between two non-wire-connected components of a satellite (can't imagine what). – called2voyage Oct 05 '16 at 20:49
  • @Phiteros And then he does mention the space station, and human-occupied vehicles, which could potentially have use for it. – called2voyage Oct 05 '16 at 20:50
  • 1
    @LocalFluff The more signals you introduce though, the less useful it will be. In a car, you typically have a 1 phone to 1 car connection. – called2voyage Oct 05 '16 at 20:58
  • 1
    What wireless connections exist in a car that aren't to devices carried by passengers? Why on earth would you introduce power-hungry EM noise sources if you didn't need to? – Russell Borogove Oct 06 '16 at 00:47
  • NFC would be very useful for robotic arms do you want me to expand on that ?. – Antzi Oct 06 '16 at 07:04
  • @RussellBorogove Cars was a bad example. The industry would like to get rid of wires there, but there's probably not too much communication going on. In factories radio like zigbee is used to reduce wiring, simplifying manufacturing and eliminating sources of failures in machines. – LocalFluff Oct 06 '16 at 09:31
  • @Antzi Yes, that would be interesting if you have a minute. An issue is that one still needs wires for electric power, and maybe one could use the same wire for communication. So how much is gained? – LocalFluff Oct 06 '16 at 09:33
  • 2
    On a factory floor, power used for wireless communications is a drop in the bucket to the total power budget -- not so in a satellite. It's also used for relatively high-level commands, which corresponds to a satellite's ground link, not internal communications. For most signal applications within a sat, you can use a serial bus to reduce the total amount of wiring. Wireless for intra-satellite stuff seems like solving a non-problem. – Russell Borogove Oct 06 '16 at 14:39
  • Interestingly, the AOBA-VELOX 3 cubesat which was released today from the ISS uses a wireless communication system http://space.skyrocket.de/doc_sdat/aoba-velox-3.htm – oefe Jan 16 '17 at 19:49

5 Answers5

15

Radio may reduce some complexity, but it will introduce problems that wiring doesn't have.

  • Limited bandwidth.
  • Interference. You can separate wiring so it doesn't influence each other. Can't do that with radio: each receiver can hear all transmitters.
  • You're introducing new points of failure. A radio transmitter or receiver contains loads of active components, giving many more failure points than a wire. A wire that's been properly installed is extremely unlikely to break. In my experience, radio links are always less reliable than wires.
  • A radio transmitter uses more power than sending information through a wire.
  • Timing problems. You've added electronics to the link, these take time so sending a signal takes longer. You may have to retransmit after a collision with another transmission, adding more delay.

Radio is not used in cars beyond the Bluetooth link for your phone. Cars use a system like CANbus, which reduces the amount of wiring you need: instead of a switched wire for every application, you just have a 12V lead and a network wire, where all components attached to the network can talk to each other.

Radio links are used for convenience, not to reduce complexity.

Hobbes
  • 127,529
  • 4
  • 396
  • 565
  • Plus, using more power means you have to get rid of additional waste heat. Trivial on Earth with a decent atmosphere (convection and in extreme cases add a heatsink and/or fan); not quite that easy in the vacuum of space. That's an obvious issue to those who know about it, but it's very easy to overlook for people not in the field, because on Earth getting rid of waste heat is basically a non-problem except in the most extreme situations. – user Oct 06 '16 at 14:58
  • 1
    This answer is generally correct, though I should add that the ISS does use wireless communication for a number of purposes, including crew/payload communications (e.g., wifi), and for vehicle structural health information (EWIS and IWIS systems). Given the size of the ISS, the number of articulating joints, and the amount of instrumentation that would otherwise require cables to go over those joints, the convenience/complexity argument gets a little mushy. – Tristan Oct 07 '16 at 14:25
  • 1
    @Tristan This comment needs more attention. He did after all ask about the space station and other human space flight missions. – called2voyage Dec 07 '16 at 20:27
6

Alright, let's put on our engineering caps. To have a wireless connection, we need:

  • Wires connecting the subsystem to the wireless transmitter
  • Wires connecting the transmitter to the power supply
  • An internal environment that's fairly well shielded against RF, but allows wireless signals in its volume
  • Wires connecting the receiver to the element the subsystem is trying to communicate with
  • Wires connecting the receiver to the power supply

And this is just for one-way communication between a single set of systems.

On the other hand, for wired communication, we need:

  • A wire

In the grand scheme of things, wired connections (Over the short distances inside a vehicle, anyway) are simpler, cheaper, more reliable, and lighter. Going wireless would add huge amounts of cost, complexity, mass, and potential failure points - all of which space programs like to avoid if possible.

UIDAlexD
  • 1,108
  • 8
  • 13
5

There's another reason that makes this a bad idea. Many satellites are communications satellites of some kind, receiving communication data from a long distance and rebroadcasting it. Those depend on having a very quiet band. Transmitters tend to leak RF energy outside of their band, and it would make it difficult to hear those far away signals if there are additional RF transmitters on the spacecraft.

PearsonArtPhoto
  • 121,132
  • 22
  • 347
  • 614
2

Satellites are unmanned, nobody and nothing onboard will appreciate the comfort of a wireless connection. Every electronic subsystem is fixed to its location inside the spacecraft. Therefore any needed interconnection is possible and done using wires, cables, serial busses but not wireless. Using wireless connections would only add unnecessary complexitiy, additional power consumption, additional heat, more failure points and more possiblities for design errors.

Any interconnection problem has to be solved as simple and reliable as possible, thus any wireless connection has to be avoided.

Uwe
  • 1,714
  • 8
  • 11
  • 1
    Satellites are anything orbiting the Earth, and in case you thought he was referring to unmanned orbital equipment, he specifically says in the post: "And on the space station or other human space flight missions?" – called2voyage Dec 07 '16 at 20:25
2

This is actually an open research area and despite everyone here seeming to think it's a bad idea, companies still want to do it. Why? Well the simplest reason is 10-15% reduction in mass due to cable elimination and the volume reduction. Both of these are extremely expensive in space applications and any savings that can be made are sought after.

user67081
  • 131
  • 1
  • 5
    That's actually pretty interesting if it is actually true. Can you support this with some kind of link or citation? What company actually wants to do this? In stackexchange it's usually required that factual information like this be supported, and it helps others who would like to read further. – uhoh Dec 07 '16 at 07:12
  • fyi I've just asked a related question. – uhoh Dec 07 '16 at 07:16
  • IEEE held a conference on wireless for space and extreme environments in 2015 and the next one is planned for 2017. – Andreas Dec 07 '16 at 08:06
  • 2
    Looking at the conference talks, one is clearly the subject of this question, I believe the focus is more on extreme environments, however. I will agree that this could be a thing, weight saving is a good thing, although much of that has been saved by using two sets of wires, one communication and one power, which greatly reduces cable requirements. – PearsonArtPhoto Dec 07 '16 at 13:01
  • 2
    I will also state that more references would be a good thing for this answer. – PearsonArtPhoto Dec 07 '16 at 13:02
  • Hi all, there are various papers on this subject, although to be fair it's still not a very mature field. I suggest search terms of "intra satellite wireless", or "wireless harness satellite". I'm not saying this will necessary become industry standard, but it is certainly of interest to some companies. – user67081 Dec 08 '16 at 05:42
  • Also I can't really post proof that it's of interest to companies, I have worked in the industry though which is why I believe that. Also I'm sure you can trace some of the funding on the papers back to specific companies if you really cared! – user67081 Dec 08 '16 at 05:50
  • I doubt that 10-15 % is the mass of communication and data transmission cables only, it may be the mass of all cables together, including cables for electric energy. If you replace some cables for data and control transmission, you still need to distribute the operating voltages per cables. – Uwe Dec 08 '16 at 11:18
  • 10-15% is just a number I've seen quoted by papers in the area, no idea how realistic it is, but they should be only talking about data/control cables and not power. Some of these wiring harnesses are very complex though so I could believe it. Also, not only do you have all the wires, you have all the connectors on the pcb, shielding, busses, cable management devices, and even potentially weight balancers. – user67081 Dec 08 '16 at 23:07