8
\$\begingroup\$

Is it possible to passively pipe GPS satellite signals indoors via a waveguide, or 2 antennas connected by a coax? If so, with what type of waveguide, antennas, coax, etc. should I try and experiment? Thanks.

\$\endgroup\$
  • \$\begingroup\$ Is the location you are retransmitting have no GPS and so you are wanting to simulate gps in the building? \$\endgroup\$ – Kellenjb Dec 13 '10 at 0:28
  • 1
    \$\begingroup\$ I've often wondered if it would be possible to augment GPS with land-based "satellites" to give finer resolution than the space ones for nearby locations, but that would still be backwards compatible with an old GPS receiver. \$\endgroup\$ – endolith Dec 13 '10 at 5:43
  • 2
    \$\begingroup\$ @endolith - yes, there were articles about GPS "pseudolites" even back in the late 1990's. \$\endgroup\$ – Chris Stratton Dec 13 '10 at 6:04
  • \$\begingroup\$ @endolith, @Chris, WAAS? \$\endgroup\$ – Nick T Dec 13 '10 at 14:38
  • 1
    \$\begingroup\$ @Nick T - Close, but not quite. WAAS isn't compatible with old GPS receivers as Endolith specified. \$\endgroup\$ – Kevin Vermeer Dec 14 '10 at 18:19
11
\$\begingroup\$

We've done this for a development project that involved GPS reception. We just wanted to be able to get a GPS signal indoors at our engineering office, for basic development of the GPS drivers.

We put a good quality GPS antenna on the roof, with coax going down into the office below. On the ceiling of the office, the coax connected to a GPS-band RF amplifier, which output through an antenna also on the ceiling.

It was a low power amplifier, so it only radiated an effective signal over a short distance within the office space--less than 10 metres. But that was enough to serve our purposes.

The other equipment we used sometimes was a GPS simulator, which simulated a bunch of satellites, and could convince a GPS unit that it was at any location of our choosing. It radiated through an antenna, again, at low signal levels so it disrupted only a small area. And we only used it indoors, within our offices.

| improve this answer | |
\$\endgroup\$
  • 4
    \$\begingroup\$ Sounds like you passed up a perfectly good excuse to get a window seat. \$\endgroup\$ – Nick T Dec 13 '10 at 16:35
  • 4
    \$\begingroup\$ A window seat may not be the solution :-( One problem that I have seen is that most modern double glazed units use coated glass (such as Pilkington K-Glass). This has a metallised coating on the two inner surfaces which gives an attenuation of more than 40dB in the GPS frequency range. You often get a better signal through the window frame than the glass. \$\endgroup\$ – uɐɪ Dec 14 '10 at 10:20
  • \$\begingroup\$ I have done exactly the same thing. Sometimes the radiating antenna is in a RAM box that we use for testing the receivers. \$\endgroup\$ – uɐɪ Dec 14 '10 at 10:23
6
\$\begingroup\$

You could use a large antenna on the roof, pipe the data through an RF channel and then radiate it into the building with another antenna.

Expected Problems

If you place an antenna on the roof and pipe signals into the building and then receive them in the building, you are effectively making the point you are receiving at the antenna upstairs. Once they hit this point they are piped in series into the building.

The noise will be a problem, each antenna will lose power and the cabling with have loss. I would suggest an RF amplifier connect at the antenna end on the roof, if not, one inside will still give an improvement. This will make getting a lock much easier (if you do have problems).

Some suggestions

I would suggest coax cable, running a waveguide is not very feasible. A waveguide will be easier to pipe the signal in with, but will make your power amplifier hard.

Your antenna on the roof needs to have a wide receive angle. This will take some research, you will probably want a high effective area to receive as much power as possible.

Inside you will want an antenna that is very directed at the location you want your device for the best receive. I would suggest a horn antenna.

I did this quickly, I will spend more time on it later if I can, I think this gives a general idea. Getting this to work will not be easy. Please let me know if there is something specific I can add to help.

| improve this answer | |
\$\endgroup\$
  • 3
    \$\begingroup\$ Wouldn't I have to be concerned about the legality of any RF amplifier transmitting at GPS frequencies, even though an indoor antenna? That's why I asked if some passive means might work. \$\endgroup\$ – hotpaw2 Dec 13 '10 at 2:12
  • 1
    \$\begingroup\$ @hotpaw2, I forget things like that. I cannot easily find any information about the band which means you should not radiate at it. When you do it passively you should be able to avoid someone noticing, but in realize absorbing and re-radiating a signal can violate these rules also if they can tell you are doing it. Now, it is not feasibly for you to passively radiate a signal into a building and someone notice it outside. \$\endgroup\$ – Kortuk Dec 13 '10 at 3:14
  • 1
    \$\begingroup\$ @Kortuk, How can passive re-radiation possibly be illegal, since, if I remember my physics correctly, almost every metal object of about the right size re-radiates to some degree? \$\endgroup\$ – hotpaw2 Dec 13 '10 at 5:04
  • 1
    \$\begingroup\$ @hotpaw2, If it is illegal, I'd would imagine the regulation would be aimed at intent. I wouldn't be worried about it for meter-sized broadcast radii, but I'm not from the US. \$\endgroup\$ – tyblu Dec 13 '10 at 5:37
  • 1
    \$\begingroup\$ @hotpaw2, If you are doing this for iPhone apps, they use cell towers for their first lock, and you will find that you are getting into a mess faking towers and GPS. I would suggest you take code that is considered "proven" and treat the interface as function and when you need fake locations you fake the results from that interface. \$\endgroup\$ – Kortuk Dec 13 '10 at 16:47
5
\$\begingroup\$

We provide equipment that will bring the GPS signal indoors. Yes, you need a passive antenna and an active antenna - you have to be able to calculate the correct amount of cable and possibly the use of an amplifier. All of this is regulated by the FCC. Unless you are a Federal agency, part of the military or using this in an anechoic chamber, you have to apply for a license from the FCC.

| improve this answer | |
\$\endgroup\$
3
\$\begingroup\$

This seems like a lot of work, for very little gain, to me. You would have to subtract the lengths of each of the waveguides from the distances calculated by the GPS receiver electronics. In fact it's more complicated than that: the signal would reflect around inside the wave guide, making the apparent length longer than the actual length.

My suggestion: put the receiver on the roof, in a weather-proof housing, and then send the coordinates wherever you need them, wired or wirelessly.

Edit: I think I didn't understand the point of the OPs question. I thought he was interested in putting GPS in a moving vehicle like an RV. If Ian's assumption is correct and OPs goal is to test a GPS from indoors, then my suggestion won't really work either. I think the only solution possible would be to use a piece of coax with an external antenna. But make sure the coax is just the right length for the frequencies used - GPS signals are very weak and any line losses would certainly cause problems.

| improve this answer | |
\$\endgroup\$
  • 1
    \$\begingroup\$ This is not a lot of use if you are trying to test a GPS receiver indoors within a system. The whole point is that you need the GPS being tested to convert the RF into the position data. The fact that it will produce a solution based on the position of the external antenna is not normally a problem. \$\endgroup\$ – uɐɪ Dec 14 '10 at 13:49
  • 1
    \$\begingroup\$ @Ian: I think you misunderstand my comments about subtracting the apparent length of the waveguide from the distance calculations. Each satellite is in a different position in the sky, so if the waveguide is say 50 ft long, then the distances to the satellite will be 50 ft too long in different directions for each satellite. I doubt the GPS receiver would be able to lock on at all in that situation, because it would be calculating distances that don't line up, and therefore make no sense. I don't think it would calculate the position of the antenna - it simply wouldn't lock at all. \$\endgroup\$ – Eric Cox Dec 14 '10 at 16:36
  • 1
    \$\begingroup\$ @Eric Cox : If this were true, then how do the active GPS repeaters (sold for International markets or shielded test room use) work at the end of their long coax connections? \$\endgroup\$ – hotpaw2 Dec 14 '10 at 22:07
  • 1
    \$\begingroup\$ @Eric. The signals from the satellites all arrive at the outside antenna with their different path lengths and hence signal delays up to that point. The onward transmission of the signal through whatever transmission medium used (waveguide/co-axial cable/RF amplifier) will ideally only add a consistent time delay to all of the signals. This will make the GPS receiver, that can only see the re-radiated signals, think that it is actually outside and a couple of nanoseconds earlier. \$\endgroup\$ – uɐɪ Dec 15 '10 at 8:11
  • 1
    \$\begingroup\$ @Eric - how can the receive antenna tell that the signals are radiating out from the retransmitting antenna in "divergent directions". It will only (ignoring reflections) receive the direct line signal between the two antennas. It is exactly the same as having a remote antenna and a long coax \$\endgroup\$ – uɐɪ Jan 4 '11 at 12:08
2
\$\begingroup\$

Note: I think the other answers for this are better than mine. Just wanted to throw this out there, especially if you use an active device.

This is effectively building a GPS repeater. To my knowledge, at least in the US, there are regulations as to when/where you can use one.

It can be done, there are devices to do it if you look around (how else would people design devices that use GPS if they had to be outside all the time???), but you need to meet regulatory standards and the rules so as to not interfere with others.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ The way that GPS receivers get designed is with a GPS simulator system, as far as I know. National Instruments offers one, among other companies, but they are not cheap. Something on the order of 30-40k, as I recall. But, if you are shipping 100k+ units of your GPS chip, then the 40k for a GPS simulator isn't too bad. \$\endgroup\$ – mjcarroll Dec 14 '10 at 17:33

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.