When determining your location using ONLY Wi-Fi signals?

i will work on a project that is trying to determine your position using Wi-Fi signal strength from several access points. I was wondering if anyone knew about any similar projects or any articles on this topic.

If anyone cares: his research project at my university. the application is written, used as a playing field to develop a new Wi-Fi antenna, which is better suited for this type of use. I am only working on the coding part of the project. oh and writing it using C #, which is optional


Clarification: Its ONLY Wi-Fi. There is no GPS goodnes for us, because it must work indoors. The software must determine your location using the known locations of the access points and their signal strengths to tell you where you are. its accuracy is currently about 4-5 meters of accuracy, in addition, I already have a working prototype, and I'm just wondering if something like this was done before or if anyone has any tips or ideas for the project /about the project

+4
source share
7 answers

Having given up the first two answers, where they need to use GPS and A-GPS in the first and well-known Wi-Fi networks in the second, I answer:

it sounds simple, but first you need to do homework, a survey.

  • you will need to measure and create an oval shape (in paper) with points and percentages of all wifi routers in the camp.
  • When, imagine that you compile 2 information about routers, you are ready to go.
  • Get current Wi-Fi points and signal strength from a user laptop / device and query the database using these values.
  • provide the user with the current location.

Example:

  • on campus, you measured that at this point you need to have about 55% of the signal strength from WiFi router 1 and 25% from WiFi router 2.

To use all of this in C #, you should start with this code draft article to get the strengths of the signal. Then simply use the return values ​​with your data that you measured before the first survey began.

Hope this helps :) At least that was what I would do to approach this issue.

+6
source

We did this in a project that already determines the distance from access points, but without signal triangulation (already discussed here in other answers).

I have a recommendation from "man, I am sorry that I didn’t have to go back and do this", it would be to spend extra time on 2 areas:

  • A simple and repeatable calibration method using multiple data points . For example, the transition from “very close” to “curiously closed” will be much more than “really far” to “really really far”. It will not be a linear slope.
  • Data smoothing. When you move, the signal strength will be disproportionately different from your movement (due to obstacles in the way). This will make your results more accurate if you take a moving average from the last 5-10 samples of signal strength, and not just take the last sample.
+4
source

You may like the Google Gears Geolocation API . I don't know how easy it is to use it with C #, but it can be useful / relevant.

(Disclaimer: I work at Google, but had nothing to do with the geolocation API.)

+1
source

I would not have thought that signal strength would work well: too many things interfere with it (for example, walls). But you can try to measure ping time, especially if you threw a hundred pings in the AP and took the average. Most of the time, the AP will digest the ping and generate the pong, but will also make up the transit time for the signals. If the AP response time is consistent, you should be able to subtract it and, therefore, calculate the distance to the access point. According to famous hyperbolic mathematicians, you can find your position from 3 or 4 access points.

Pavel.

+1
source

Skyhook It depends on the people registering the location of their access points.

0
source

Goto Wifi in a positioning system (WPS), where we use several Wifi access points to track location. it is implemented by many systems, such as Google, horizon, etc., providing additional security privacy. I, too, not knowing what I tried to prevent this technique. At best, we can simplify our complexity in a variety of ways - by introducing INNOVATIVE into your existing infrastructure. I recommend that you move on to managing and using the wifi router add-on before pursuing this.

0
source

You can try the FIND3 project, here the documentation is https://www.internalpositioning.com/doc/ , its API can be used for this purpose and you can use the learning algorithm to constantly study your position with accuracy. They also have mobile apps (Android only). I want to do something like this.

0
source

All Articles