12
$\begingroup$

I was reading up on fly-by-wire development, and I saw a short section about fly-by-wireless technology. It seems like a great idea, with the potential to lower costs, weight and complexity. I can see a possible scenario where it could be an issue, though:

  1. Two aircraft are very close together (e.g. on a runway or flying in formation).
  2. One pilot transmits commands through the aircraft's fly-by-wire system to other parts of the aircraft.
  3. The other aircraft accidentally receives the signal because it is so close.
  4. Things get very bad very quickly.

The thing is, I haven't been able to find any technical specifications regarding fly-by-wireless systems, and I have no idea if the transmission would be powerful enough to reach the other craft, nor if it would then be interpreted as actual data sent from that aircraft's pilot.

Is this cross-interference between fly-by-wireless systems possible? If so, how can it be mitigated?

$\endgroup$
  • 4
    $\begingroup$ I assume such a crucial system would be digital, in which case this wouldn't be a problem. E.g. a computer on a wifi network with multiple users doesn't receive pieces of data intended for the other computers. $\endgroup$ – Chris Mueller Mar 13 '15 at 21:57
  • $\begingroup$ @ChrisMueller Actually it does receive the data and acts on it or discards it. By 'act on it' I mean that it typically waits while the network is busy before it transmits its own data. $\endgroup$ – dcorking Mar 16 '15 at 19:45
  • 2
    $\begingroup$ @dcorking While that may be true, the computer still knows that the data isn't intended for it. $\endgroup$ – Chris Mueller Mar 16 '15 at 20:47
10
$\begingroup$

Cross-interference between aircraft is a high unlikely event because all commercial aircraft designs have to pass DO-160 environmental testing requirements. Among the DO-160 testing specification is EMI/EMC Testing. These tests include Radiated Emission, Interference and Immunity Testing. Part of these tests are to answer "Two aircraft are very close together" and "The other aircraft accidentally receives the signal because it is so close." Below is a picture of an anechoic chamber used to test aircraft.

enter image description here

Fly-by-wire allows the aircraft control system to access and command monitoring and control systems. The monitoring and control systems have a specific address similar to an internet protocol (IP) address. This enables the aircraft control system to address specific devices.

Aircraft testing protocols are similar to automobile testing protocols except more stringent. But all engineering designs are not foolproof. To mitigate risk most engineers use a process called DFMEA (Design Failure Mode Effect Analysis). Unfortunately there are still a few unfortunate incidents. The most recent examples are the loss of MH370 and Toyota vehicle recalls for sudden unintended acceleration.


References:

$\endgroup$
9
$\begingroup$

To add to what Russell said, it would take a very stupidly designed protocol to allow the comands from one aircraft to control the other, even assuming that cross-reception is good.

As a everyday example, think of a bunch of people talking on cell phones standing near each other. One person accidentally receiving another's conversation just doesn't happen. This has to be deliberately designed into the protocols, but that's no different from designing the landing gear robust enough to hold the weight of the aircraft. It's just something you'd do as a normal part of the design.

My fear about this kind of system is not cross-communication, but overpowering interference. You have to make sure that the other nearby aircraft's transmissions didn't overwhelm your receivers to the point where they can't hear your signals anymore. That shouldn't be too hard to do either assuming everyone plays nice.

However, guarding against deliberate outside interference would not be so easy. The other aircraft, if designed to do so, could deliberately radiate orders of magnitude more power to your receivers than your own transmitters do. The protocols would make it essentially impossible for them to take over your aircraft, but your controls would no longer function. This is basically like a DoS (denial of service) attack.

I doubt we will see these kinds of systems widely deployed, at least in commercial and military aircraft, until they are considered safe from this kind of attack. Otherwise, a terrorist with a steerable parabolic dish antenna could cause some serious trouble.

There are other alternatives to using lots of heavy copper cable for communication, like fiber optics and various kinds of multiplexing.

$\endgroup$
  • $\begingroup$ I probably should have indicated it in the question, but I was worried about interference, as you mentioned. Thanks for addressing that. $\endgroup$ – HDE 226868 Mar 14 '15 at 21:57
  • 2
    $\begingroup$ I would guess that wireless first will be used for receiving measurements from large arrays of not-too-critical sensors. Structural health monitoring sensors, for example. $\endgroup$ – Nick Alexeev Mar 15 '15 at 8:58
  • $\begingroup$ +1 For the discussion of DoS attacks on this type of system. $\endgroup$ – Chris Mueller Mar 16 '15 at 20:49
8
$\begingroup$

Modern communications systems are able to achieve essentially any desired data rate and degree of message integrity in any well-enough-defined environment. It's "just a matter" of the degree of effort, complexity and $ that are needed.

Designing a system to meet any desired data rate, number of users, and integrity is thus "just a matter of engineering". Proximity, signal strength, interfering signals, ... are 'just part of the system specification'.

Things go bad when the specification and reality agree to differ and/or when properly failsafe design is not used. "Defining the problem" is arguably the hardest and most important part of such a system.

$\endgroup$
5
$\begingroup$

No, crossed communications would be a total non-issue.

Think of it like a home wireless network. All your devices at home can talk to each other on your WiFi network, and the same is true for your neighbour. But your devices can't talk to your neighbour's devices because they are on a different wireless network.

Such wireless networks in planes would have to be far more robust than your typical home network, as there are other potential issues:

  • Accidental interference from another craft that doesn't cross-communicate between craft, but potentially blocks communication on one or both networks.
  • Deliberate interference (jamming) of the communications by for example a terrorist on the plane or nearby.
  • Deliberate interception or spoofing of communications to hi-jack the plane. This would require hacking into the secure network and would therefore be much harder than jamming.

Some ideas to mitigate these problems might be:

  • Directional antennae on the wireless equipment. This could help to make jamming or interception much harder because other signals would be blocked so any would-be hacker would need to have their equipment physically located directly between the antennae.
  • Multi-part encryption keys and certificates in the devices and stored on a card held by the pilot. This would allow the equipment to verify commands were coming from a verified source (the pilot/cockpit) and also ensure that nobody else could listen in or intercept/spoof the commands.
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.