Detecting electronic signals at a distance

Thread Starter

liontamer

Joined Dec 11, 2011
2
I'm a children's author, and know nothing about electronics, so need a bit of guidance. Please excuse me if my questions are laughable.
I'm writing a futuristic novel, so I'm talking about what might be possible in the future, rather than what is possible now. So - is it possible to detect electronic signals at a distance? i.e. could someone theoretically track down an artificial intelligence through the signals it is sending?
Basically what I've got is an AI that is hiding on a ship. I want it to shut itself down so that its enemies can't track it.
Does this make sense?
 

Adjuster

Joined Dec 26, 2010
2,148
How do you suppose a radio works? Seriously though, most electronic equipment, including computers, will emit a certain amount of electromagnetic radiation. This can for instance result in radio interference - you may have noticed that putting your radio close to your lap-top does not give the best reception.

Listening-in detectors have been proposed for various purposes from detecting TV licence evaders to espionage. The effectiveness of some of these things may be exaggerated, but they at least have some possible basis in reality.

Although systems can be screened to minimise such radiation, it is hard to suppress it completely, and usually not cost-effective to do so, unless for some purpose such as secret operation. One thing that might work against this idea in a ship is that a metal hull may provide a degree of screening from the nosey detector, but perhaps you could work that into your story, for instance taking refuge in the bowels of the ship.
 

jimkeith

Joined Oct 26, 2011
540
Well every computer (artificial intelligence) has a system clock--they range in frequency from perhaps 100mHZ to 10gHZ--most new desktops run around 2gHZ or so. This signal is radiated as low level RF and can be picked up with the proper receiver--in this case a spectrum analyzer--the signal spectrum analysis could easily detect all kinds of nuances like non-standard frequencies thus making it possible to detect alien stuff--each associated harmonic and side tone appears as a unique blip on the spectrum display--the dish antenna makes it very directional thus making it possible to locate the source, but the presence of steel bulkheads (as in ships) introduces great complexities (reflections) as well as do other sources of RF (like shipboard radar, communications and imbedded systems) thus making it difficult to locate--and the air ventilation ducts could cause very interesting wave guide like channels for RF--so it becomes a great cat and mouse game. In futuristic fiction, the spectrum analyzer would be standard equipment on military vessels.

http://en.wikipedia.org/wiki/Spectrum_analyzer

Microcontrollers that run on batteries have a sleep mode so if it is idle while waiting for a command it will turn off its clock generator until the proper interrupt occurs thus saving battery power--this would make it almost impossible to locate and would cause great frustration if it awoke occasionally for brief periods of time--something like a beeper tormentor that can be placed over a ceiling tile in someone's office cubicle--did you hear that???

http://embedded-lab.com/blog/?p=3237

http://forum.allaboutcircuits.com/showthread.php?t=47466
 
Last edited:

tracecom

Joined Apr 16, 2010
3,944
Basically what I've got is an AI that is hiding on a ship. I want it to shut itself down so that its enemies can't track it.
Does this make sense?
Cool idea! And of course, the problem is, who or what is going to turn it back on when the danger is past? Perhaps the young hero of the story?
 

praondevou

Joined Jul 9, 2011
2,942
Provided the AI is a real robot, not a hybrid between human and machine or something similar it will be quite easy for it to shutdown and hide itself - considering nowadays tracking methods/possibilities.

A robot could power down so much leaving only an ultra low power microprocessor working that it would be almost impossible to detect it , with equipment available today.
I wonder how the AI would decide when it is a good moment to recover from sleep mode, it would have to have at least some surveillance functions running.

However in the future it may be possible to have a function in every room on the ship that detects if total air volume inside has changed, or if something got in or out without permission. I mean we see in futuristic movies that all doors are automatic, for a cargo ship there are probably electronic locks on the door / gates that give permission to enter only to the robot/humanoid/humans that have the right to do so.
If I was a space ship designer I could also implement weight sensors on the floor , this way I would be absolutely sure that there is nothing hidden I don't know about. In this case the AI would have to hover, but then it wouldn't be shut down, right? Unless it's hanging from the ceiling, but then it wouldn't be hidden... :D

Many scenarios are imaginable... If this is a "dumb" ship and you only want to hide from enemies, deep sleep mode , as you said, would be a good option.
 

tracecom

Joined Apr 16, 2010
3,944
However in the future it may be possible to have a function in every room on the ship that detects if total air volume inside has changed, or if something got in or out without permission. I mean we see in futuristic movies that all doors are automatic, for a cargo ship there are probably electronic locks on the door / gates that give permission to enter only to the robot/humanoid/humans that have the right to do so.
If I was a space ship designer I could also implement weight sensors on the floor , this way I would be absolutely sure that there is nothing hidden I don't know about. In this case the AI would have to hover, but then it wouldn't be shut down, right? Unless it's hanging from the ceiling, but then it wouldn't be hidden... :D.
If you haven't read "The Cold Equations," a sci-fi short story by Tom Godwin, you should. It uses some of the techniques you describe.
 

Adjuster

Joined Dec 26, 2010
2,148
Another way to hide is to join a crowd. Then there is the possibility of a decoy, something like a radio transmitter deliberately emitting the signals being searched for.
 

thatoneguy

Joined Feb 19, 2009
6,359
Don't forget the thermal signature if he is hiding in a ship. Also, energy: spaceships are rather advanced, even our little toys today, let alone in a future with AI, so energy is conserved by not keeping oxygen or heat in unoccupied areas (Heinlein). The main system would know who is where, unless the hideout manages to hide in a crowd, as mentioned above. There are the usual ventilation systems one can crawl through (Enders Game), but operation would be erratic.

I hate it when sci-fi authors mess up stuff like that, such as hiding out in a cargo hold, which might have a 100% nitrogen atmosphere to preserve the cargo.

Detecting neutrinos is a new way sci-fi has been introducing location methods, but those wouldn't be emitted from an android AI(Ringo, Drake, Weber, many others). Generally only fusion/fission/ and <unknown plot point here> create neutrinos, which we can only currently detect miles deep inside unused mines using tanks of solvent (Homestake gold mine, South Dakota)

Robots can see far infrared and short UV, the entire spectrum if desired. So an Android probably could as well.

Lots of other electronic signatures, display screens, especially holographic laser scanner types, or just system clocks emit radio in the kHz to Ghz range. It is unknown at this time whether they can transit a force field or not. Sometimes yes, sometimes no. Star Wars had force fields that allowed light to pass, but not lasers, colored light, perhaps it was a polarizing filter as well?

I'd be happy to look at a rough draft and point out anything stupid silly. At least take the time to read Wikipedia to get ideas about anything you will reference, such as Radio Frequency (also look up Van Eck Phreaking). In addition to the very latest DoD company promo videos for less lethal weapons, sensor suites, etc. A lot of what was sci-fi in the 80's is real today, because the authors kept current on what the DoD was interested in having built. Railguns are a worn out concept, though, pretty much a given in any sci-fi book.
 

hgmjr

Joined Jan 28, 2005
9,027
With an AI then there is the possibility that the entity can mimic a signal that is generic to the environment it find itself in and therefore cloak itself so as to blend into its surroundings.

hgmjr
 

Thread Starter

liontamer

Joined Dec 11, 2011
2
Thank you all so much for your very helpful answers. You've clarified a number of things for me. This is where I am now:
The AI has a powerful and distinctive electromagnetic signature. This is what its enemies are pursuing, and are tuned into. So although the ship itself (sea-going, not space) presumably also emits electromagnetic radiation, they won't pick this up because they are not tuned to it. Make sense so far?

Before it shut itself down, the AI also calved off little bits of itself to create maintenance, surveillance and protective systems for the ship and its human crew. One of these systems is supposed to wake it up after a certain amount of time, but with salty air, seawater etc things go wrong. Which, as you guessed, Tracecom, is where the hero comes in.

Can anyone see problems with this? I'll check out Wikipedia as you suggest, Thatoneguy, but am sure to have more questions as I go, and at some stage will want to run relevant bits of the novel past some knowledgeable readers. And sorry, Praondevou, I won't be able to post it up here, as my publisher would not like it!

If you want to see what I've written previously, have a look here. (These current books are fantasy rather than SF.) http://www.randomhouse.com/kids/keeperstrilogy/
 
Top