How do I use C# to create a fingerprint of the floor plan of a building

Question

1.57/5 (3 votes)

See more:

I need to capture the floor plan of a building using the fingerprint method for indoor localization, to enable me locate a user in a building. The fingerprint map will be put on an X-Y co-ordinate graph, to make location easier. I plan on doing this with the C# programming language. And also would like to try magnetic fingerprinting method.

Any ideas of how I can do this or any other easier methods to going about this would be highly appreciated. Thank you for your contributions.

What I have tried:

I have checked various articles, but non seem to show how to go about, they basically just describe how their own system works.

Posted 14-Jan-18 0:55am

Stephen Dauda

Updated 14-Jan-18 6:10am

Add a Solution

1 solution

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Dave Kreskowiak · Answer 1 · 2018-01-14T06:10:00

Solution 1

There is no "easy method" for this. Each corporate implementation is proprietary and it cost them a LOT of money and research time to develop. As far as I can tell, there is no public implementation available.

My wife works for a company that does this very thing in large buildings, but not to the point of locating a person at a specific X,Y location in a room. You just know "this person or thing is IN such and such room". It's NOT cheap to do just at that level.

Your problem is going to be sensors. How are you going to physically determine where in the room a person is? "Fingerprinting" does you no good unless you've got data to process coming from some sensors.

Posted 14-Jan-18 6:10am

Dave Kreskowiak

Comments

Stephen Dauda 17-Jan-18 4:40am

I plan on using Wi-Fi access points as my sensors. So on the map i create, I should be able to store the positions of the Wi-Fi access points in my database, which i think is the fingerprinting, so the floor plan will be placed on an X,Y co-ordinate, and the Wi-Fi access points will be placed by there as static points (reference points). Then i would by myself store different positions on the map database as possible places where the user could be at.
And that i think is fingerprinting, storing each possible point on the database, with the Wi-Fi access points also stored on it, then using the database to estimate the position of the user.

I don't know if that makes sense to you?

But basically i need help with drawing a graph or co-ordinate on C# and adding the floor plan to it, so that it somehow merges.

Dave Kreskowiak 17-Jan-18 19:35pm

We tried a commercial version of something like that and found it wasn't accurate at all. You could get a position down to a rather large section of a floor, not a room. The problem is WiFi signals reflect off of everything, and everyone, making an accurate triangulation based on signal strength pretty much impossible. Stand with a laptop in a conference room and you'll get a signal strength from the access point. Now, put a second person in the room without moving the laptop and it'll get a different set of signal strengths to each access point!

There is no real "fingerprinting" here (Google for "audio fingerprinting" for what that really means.) It's just standing in a location, getting signal strengths from your access points (which can vary even when you're standing still!) and putting that point of data in your database associated with a location coordinate.

What you're going to get when you "search" on the strength data is a list of possible locations, probably wildly varying all over the floor/building plan. You search has to cover a range of values because you're not going to a solid, unchanging strength to search on.

Want to try it yourself? Stand in one place with a laptop running a signal strength monitor. You'll see that the strength changes constantly. Now, turn 180 degrees and keep watching the signal strength. It'll keep changing and vary by more than 20%.

Other than that major problem, you're just clicking a point on an image and saving the X,Y coordinate of that click. Add that to the database as a location with a unique identifier and start collecting your signal strength data. Any of that data you save to the database will have to be associated with the ID of the point you created when you clicked and saved the X,Y coordinate.

Stephen Dauda 21-Jan-18 4:14am

Okay. I understand.

What if I try using magnetic field intensity rather than the access points. The magnetic field intensity varies from position to position due to the varying amount of steel present in each area. This should be a good method to test out, right?

And if so, how can i go about collecting and storing the different magnetic intensity readings in different areas and storing them into a database.

Dave Kreskowiak 21-Jan-18 11:57am

LOL! You have the SAME problem. Do you have any idea how many magnets and electromagnets are running around an office floor, both stationary and mobile?

Also, you need magnetometers in every device you're going to track. Smart phones usually some them for a compass, but not every magnetometer is the same. Some have a resolution good enough to find North and that's about it. Laptops don't usually have them.

Stephen Dauda 23-Jan-18 8:36am

But, it isn't meant for laptops, but mobile devices. And i don't need it to be 100% optimal, but for it to work to give me a proof of concept.

Dave Kreskowiak 23-Jan-18 9:26am

You're not the first person to come up with this. There are plenty of papers on the subject. Pay close attention to the references each paper cites and look those up and read them.

This technology is not as simple as it sounds on the surface. Some of the papers are from companies that came up with their own implementation, but they also leave out certain details. For example, there's one paper that says "steel and concrete in the building don't interfere with positioning accuracy, but, make it more accurate". But what the paper does NOT say is that air handling units and elevator motors on intermediate floors and the roof above the top floor WILL screw with accuracy when they are running.

"magnetic floor plan fingerprinting"[^]

Stephen Dauda 29-Jan-18 5:13am

Okay. So which technology method do you advise that I use to complete the project?

Dave Kreskowiak 29-Jan-18 8:28am

There lies the problem. Inside buildings, it's a chaotic sea of electromagnetic radiation. About the only thing that works reliably are RFID portals.

Stephen Dauda 31-Jan-18 6:10am

But aren't those a little tricky to implement?

Dave Kreskowiak 31-Jan-18 7:32am

Compared to what you've been talking about, it's a LOT easier to implement.

Stephen Dauda 5-Feb-18 8:00am

But it will involve me spending some money to get the required hardware for the RFID, right?

Dave Kreskowiak 5-Feb-18 8:25am

Unless you find a "reputable dealer" selling out of the back of a van in a parking lot, yes.

Stephen Dauda 12-Feb-18 9:36am

But it's for an undergraduate final year project, so spending money on the RFID hardware doesn't seem profitable

Dave Kreskowiak 12-Feb-18 12:01pm

Good luck with "fingerprinting" a room. You've got a lot of reading to do.

For a final year project, you can get away with this because you can always setup a controlled environment to get it to work, but, as an actual product, it's not going to fly.

Stephen Dauda 14-Feb-18 8:52am

Okay, I see. Can you recommend any articles that I can use to aid me, because I plan on using Trilatertaion, and the floor i plan to use as my use case only has 2 access points, and trilateration requires 3 access points.

Dave Kreskowiak 14-Feb-18 12:23pm

No, but you can Google for "trilateration wifi access points" to find a bunch.