Got a new toy

https://play.google....ifisolver&hl=en

I need to figure out scaling and a few other things, but here's a simple example of one access point at work.

That's actually really interesting, I didn't know there was anything like this on the market

If you use Ekahau you can do actual signal testing based on moving about the location marking yourself on a floorplan, it generates a heatmap of the signal strength in the building.

Oh that's cool. Other radio signals in the vicinity - usually neighbours' WiFi - can cause destructive interference, too. So unless it can factor that in some way, it won't be an accurate representation.

Oh that's cool. Other radio signals in the vicinity - usually neighbours' WiFi - can cause destructive interference, too. So unless it can factor that in some way, it won't be an accurate representation.

Well since it measures the strength/clarity of the received signal by the device, it's going to factor in anything that can affect communication between the device and the AP

Well since it measures the strength/clarity of the received signal by the device, it's going to factor in anything that can affect communication between the device and the AP

This app runs a simulation, it says nowhere that it performs any actual measurements. It requires you to use a pre-determined image of a floor-plan, where the app uses edge-detection to automatically find the walls. You configure the materials, the location of the router, and it simulates away.

The demo simulation the page links to shows the propagation of waves from the router, towards the outskirts of the property. However, when the simulation first begins, there are no other waves depicted anywhere on the image until the radiation from the router propagates to the outside of the image. 

It says 1px maps to 1cm, so going by its recommended maximum image size of 500px, that gives us a max distance of five metres. It takes light approx. 17 nanoseconds (0.000000017 seconds) to travel that distance, so I'm not sure an android phone has enough time-resolution to make precise enough measurements to correctly determine how external radio sources will interfere.

Plus, it being a 2D simulation, there's an extra dimension being ignored. Radiation propagates spherically from the source, so waves will be reflecting off of the floor/ceiling and then interfering with itself.

That looks like a pretty cool system, but I don't think it's $3000 useful. 

This app runs a simulation, it says nowhere that it performs any actual measurements. It requires you to use a pre-determined image of a floor-plan, where the app uses edge-detection to automatically find the walls. You configure the materials, the location of the router, and it simulates away.

 

The demo simulation the page links to shows the propagation of waves from the router, towards the outskirts of the property. However, when the simulation first begins, there are no other waves depicted anywhere on the image until the radiation from the router propagates to the outside of the image. 

 

It says 1px maps to 1cm, so going by its recommended maximum image size of 500px, that gives us a max distance of five metres. It takes light approx. 17 nanoseconds (0.000000017 seconds) to travel that distance, so I'm not sure an android phone has enough time-resolution to make precise enough measurements to correctly determine how external radio sources will interfere.

 

Plus, it being a 2D simulation, there's an extra dimension being ignored. Radiation propagates spherically from the source, so waves will be reflecting off of the floor/ceiling and then interfering with itself.

I feel like it's a 2d representation of 3d information, sort of like a topographical map. It doesn't particularly matter if it shows height, as long as it shows the end signal strength after it's been reflected off of all surfaces too your location. 

That looks like a pretty cool system, but I don't think it's $3000 useful. 

 

 

I feel like it's a 2d representation of 3d information, sort of like a topographical map. It doesn't particularly matter if it shows height, as long as it shows the end signal strength after it's been reflected off of all surfaces too your location. 

He uses a 2D matrix to save computational complexity. Even then, in order to store the amount of information required to fully represent the data in memory would have been in excess of 60GB. He had to employ several techniques to reduce the computation complexity by a good 100 times, and also reduce the amount of memory required.

The method he uses breaks the floorplan down into a two-dimensional grid, each pixel representing a 1x1 cm block. The equation considers the four nearest neighbours of each point in the 2D grid.

A 3D grid would essentially exponentiate the amount of work required to simulate the wave propagation.

The full details of the working of this app can be seen on the developer's website, here.

This app runs a simulation, it says nowhere that it performs any actual measurements. It requires you to use a pre-determined image of a floor-plan, where the app uses edge-detection to automatically find the walls. You configure the materials, the location of the router, and it simulates away.
 

Actually, they have a heatmapper, which is the only software I'd ever heard of from them, so when I said Ekahau that's what I meant.
https://www.ekahau.c...apper/overview/

That tool requires you to walk through the premises and takes data on available access points and WiFi strength etc

Oh, when I posted the following, it was in context with the OP, not your post. 

Oh that's cool. Other radio signals in the vicinity - usually neighbours' WiFi - can cause destructive interference, too. So unless it can factor that in some way, it won't be an accurate representation.

The moral of the story? Always quote the post you're replying to.