Take the 2-minute tour ×
Android Enthusiasts Stack Exchange is a question and answer site for enthusiasts and power users of the Android operating system. It's 100% free, no registration required.

I'm try to gather data on the feasibility of my next Android project in which I would like to use GPS indoors. Knowing that GPS wasn't built to be accurate indoors, I'm trying to decide if the project is worthwhile.

I have a rough estimate of my required accuracy (10 meters). The location is the middle of a major city.

So I'm wondering if there is any data out there that relates to the GPS systems on Android phones that would help me with this research.

Thanks.

Edit:

Thanks for the answers so far folks. Good reads. I realized that there are a lot of factors affecting GPS data. Even though I may be "shooting for the moon" with my accuracy requirements, I'm hoping that my unique situation might help achieve my goal indoors.

I intended to use GPS + wireless network data (Assisted GPS) and try to get the best accuracy possible. My user will also be in the middle of a major city, which I would imagine would have many cell towers. Whether or not I can achieve this with current/up-and-coming Android technology is really what I would like to determine, so my search for actual data representing valid tests of Android devices continues.

If you have any links to data of this sort, please do share.

Thanks again all!

share|improve this question
    
I think it is better if you widen the scope of your technologies. GPS is not necessarily the best tool to satisfy your requirements. And of course, you need to explain your requirement a bit better. Do you really need absolute positioning? or is relative positioning sufficient? is it possible to assume the ability to scout the area for wifi AP/cell tower positions? Is the device intended to be used together with another hardware that have fixed positions? –  Lie Ryan Mar 26 '11 at 7:19

3 Answers 3

10 m? You'd be lucky to get that kind of accuracy outdoors.

I'm the developer of Car Dashboard, which is a car home replacement that includes GPS features like speedometer, etc. I have the default minimum GPS accuracy set to 200 m which seems to work pretty good for most users (I also have it adjustable in the settings in case they have issues). On average, I'd say indoors near a window will get you about 50-100 m accuracy. Outside with nice conditions should get you below 50 m. Indoors without any visual sight to the sky is going to be a dead zone. The signal is just simply not going to reach you and if it does, it is going to be highly inaccurate.

This is the resource I use to describe GPS accuracy in my application:

GPS accuracy is affected by a number of factors, including satellite positions, noise in the radio signal, atmospheric conditions, and natural barriers to the signal. Noise can create an error between 1 to 10 meters and results from static or interference from something near the receiver or something on the same frequency. Objects such a mountains or buildings between the satellite and the receiver can also produce error, sometimes up to 30 meters. The most accurate determination of position occurs when the satellite and receiver have a clear view of each other and no other objects interfere.

share|improve this answer
    
Edited post with comment, thanks! –  user1827 Mar 25 '11 at 23:00

I found this from GPS Tracklog. It's from 2005 but my guess is that the technology is roughly (if not exactly) the same.

They come through my roof (plywood and spanish tiles on top) just fine too, and they also make it through typical fiberglass boat construction. Most materials are not good absorbers of microwaves, but anything with a high water content (incl. people) or metal will block the signals. As a general rule, anything that will heat up readily in a microwave oven will also absorb the GPS signals although the frequencies are a little different (1.5 GHz for GPS and a little over 2 GHz for ovens). So as long as the wood and shingles on your roof are dry they’ll pass the signals ok, but of course insulation with a metal foil layer would block them. The physical reason why things with water absorb the signals is that water molecules are highly polarized with the oxygen end having a negative charge and the end with the two hydrogens having a positive charge. When the GPS signals go through water the oscillating electrical field of the microwaves cause the water molecules to rotate back and forth and this absorbs energy from the signals.

share|improve this answer
    
Edited post with comment, thanks! –  user1827 Mar 25 '11 at 23:00

Under limited circumstances, it is possible to use the accelerometer, compass, and gyroscope (and of course, Wifi, GPS, and cell tower) to increase the accuracy of GPS, by using dead reckoning/path integration.

share|improve this answer
    
Thanks for the link. My user will be walking so I imagine it might be pretty tough (if at all possible) to implement something like this. –  user1827 Mar 26 '11 at 17:48
    
@user1827: Even if the user is walking, it is still possible. David Saachs of InvenSense gave a Google TechTalks on Sensor Fusion (youtube.com/watch?v=C7JQ7Rpwn2k) back in August, skip to 23:17 to the section about calculating position through acc/comp/gyro, he talks about how to get position, its limitation, the difficulties, and how to circumvent their limitations. In particular, if your user is mostly walking, you can use pedometer algorithm to filter out the drift from the double integration. The technology is said to be on Gingerbread, but I don't have a reliable source for it. –  Lie Ryan Mar 26 '11 at 19:13

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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