Indoor WiFi positioning App using Nodemcu's as accesspoints

Thread Starter

Damien De Silva

Joined May 20, 2017
28
I want to build and an app that tells the location of a smartphone using only WiFi. I will have to use Nodmcu's as the access points. It would be a great help if you guys can give me an idea how this could be done.
Thanks a lot in advance !!!!!!!!!
 

Yaakov

Joined Jan 27, 2019
1,612
We’ll need more information. Is it telling the phone where it is? Telling someone else where the phone is? What is the resolution requirement? Which mobile platforms does it have to work with. How many NodeMCU boards will you have and what coverage do you expect?

Basically, you will be looking for MAC addresses and depending on requirements, possibly RSSI.

Here’s a project that might help: https://github.com/kalanda/esp8266-sniffer
 

Thread Starter

Damien De Silva

Joined May 20, 2017
28
We’ll need more information. Is it telling the phone where it is? Telling someone else where the phone is? What is the resolution requirement? Which mobile platforms does it have to work with. How many NodeMCU boards will you have and what coverage do you expect?

Basically, you will be looking for MAC addresses and depending on requirements, possibly RSSI.

Here’s a project that might help: https://github.com/kalanda/esp8266-sniffer
Thanks a lot for the reply !!!!!!!! I want to tell the phone where it is. If the fine final product can get an accuracy of 0.1meters it would be perfect. At the moment only for Android. 3 NodeMCUs to do triangulation probably. 15 ft * 15 ft area.
 

Glenn Holland

Joined Dec 26, 2014
705
Kinda off the subject, but also somewhat related to your idea.

The elevator industry attempted to use a form of position tracking using a radio transmitter at the top of the shaftway and a receiver on the elevator car. It used an algorithm similar to GPS tracking to locate the vertical position of the elevator platform and final stopping which must be +/- 1/4 inch from the floor.

The radio tracking would eliminate the floor selector in the control room which is operated by a perforated rape running over a pulley and attached to the elevator car frame. In addition to the radio tracking, there were also redundant sensors to determine if the elevator platform was within a safe distance from the floor to allow the doors to unlock and open.

That idea was never implemented on a commercial basis, but it was pretty clever anyway.
 
Last edited:

Sensacell

Joined Jun 19, 2012
2,567
Kinda off the subject, but also somewhat related to your idea.

The elevator industry attempted to use a form of position tracking using a radio transmitter at the top of the shaftway and a receiver on the elevator car. It used an algorithm similar to GPS tracking to locate the vertical position of the elevator platform and final stopping which must be +/- 1/4 inch from the floor.

The radio tracking would eliminate the floor selector in the control room which is operated by a perforated rape running over a pulley and attached to the elevator car frame. In addition to the radio tracking, there were also redundant sensors to determine if the elevator platform was within a safe distance from the floor to allow the doors to unlock and open.

That idea was never implemented on a commercial basis, but it was pretty clever anyway.
You have to wonder why it wasn't commercially implemented.
Usually because it's not robust, or suffers other fatal flaws.
 

Glenn Holland

Joined Dec 26, 2014
705
You have to wonder why it wasn't commercially implemented.
Usually because it's not robust, or suffers other fatal flaws.
Elevators carry people billions of passenger miles per week (second only to automobiles) and the industry has a policy of using only technology that's been proven "super safe" and 99.999% reliable.

I suspect that radio tracking may have been abandoned because of the risk of EMI/RFI and/or glitches that could interrupt operation. Most elevators are designed to operate on a "'walk away safe" philosophy with no human intervention (except for calling the service company) in case of a problem.

Remote monitoring is allowed, but over-riding any safety critical function is prohibited unless done by a person actually in the control room or another location where the passenger areas are in direct view.
 

Sensacell

Joined Jun 19, 2012
2,567
I don't think it's possible to achieve 10 cm positioning accuracy this way, not going to happen with any reasonable hardware.

Radio waves travel at 300,000,000 meter per second, that means about 3.3 nanoseconds per meter.
This implies you must measure the time of arrival to 0.33 nanosecond resolution to resolve 10 cms.
 

Yaakov

Joined Jan 27, 2019
1,612
I don't think it's possible to achieve 10 cm positioning accuracy this way, not going to happen with any reasonable hardware.

Radio waves travel at 300,000,000 meter per second, that means about 3.3 nanoseconds per meter.
This implies you must measure the time of arrival to 0.33 nanosecond resolution to resolve 10 cms.
He would be using RSSI, not time of flight to do the triangulation, but I think the resolution is still not going to be enough to get that sort of position accuracy.

I think the only way to see how well it works is test. The program I linked should have the basic capabilities necessary. I’ve done presence detection with the ESP8266 and similar code, but I wasn’t trying to do positioning.

RSSI will get him something, but frankly, I can’t tell just how much.
 

Sensacell

Joined Jun 19, 2012
2,567
He would be using RSSI, not time of flight to do the triangulation, but I think the resolution is still not going to be enough to get that sort of position accuracy.

I think the only way to see how well it works is test. The program I linked should have the basic capabilities necessary. I’ve done presence detection with the ESP8266 and similar code, but I wasn’t trying to do positioning.

RSSI will get him something, but frankly, I can’t tell just how much.
Ok makes more sense now-
And that "something" will be highly variable, depending on the orientation of antennas, building materials and many other vagaries.
 

Yaakov

Joined Jan 27, 2019
1,612
Ok makes more sense now-
And that "something" will be highly variable, depending on the orientation of antennas, building materials and many other vagaries.
Yes, but it can be calibrated. The effect of people moving around is more of a problem. Especially with only three nodes.
 

John P

Joined Oct 14, 2008
1,808
I know he said "only WiFi", but if the phone could emit an audio signal that could be picked up by the external processors, or vice versa, the project might work in an uncluttered environment. Time of flight for a sound wave is quite manageable.
 
Top