Wifi light controller

Looking into using this product to create my own AP which does not seem to be able to be purchased by a vendor ready out of the box.

We have lights which are controlled over wifi. This is a very dense layout. 1000 lights in an area of 200ft x 250ft. Then multiply this by 10 for some environments. Most of the problems we have run into are the limitations put on radios to ensure the user experience is good, Our purpose does not need high bandwidth as we are just talking about controlling the lights.

Our lights are on the 2.4 frequency. We have found AP’s which have 3 x 2.4 radios and we are able to get 192 devices connected to each of the 3 radios simultaneously. This means we can get 576 devices per AP and only need 2 AP’s for 1000 lights.

Another aspect that has interested me is if possible to shrink the channel bandwidth down from 20Mhz to 10Mhz or 5Mhz. Given that we’re NOT looking for bandwidth, but we ARE looking for maximum number of clients (per AP, but also for our total area of install ). We’re thinking it might be useful to REDUCE the channel width. This way we could move from 3 non-interfering channels for our install to 6 or even 11 non-interfering channels possibly allowing us to use a RouterBoard with 4 radios increasing the total to 768 per AP. if not more.

Then you’ve come to the right place :slight_smile:

We have lights which are controlled over wifi. This is a very dense layout. 1000 lights in an area of 200ft x 250ft. Then multiply this by 10 for some environments. Most of the problems we have run into are the limitations put on radios to ensure the user experience is good, Our purpose does not need high bandwidth as we are just talking about controlling the lights.

Our lights are on the 2.4 frequency. We have found AP’s which have 3 x 2.4 radios and we are able to get 192 devices connected to each of the 3 radios simultaneously. This means we can get 576 devices per AP and only need 2 AP’s for 1000 lights.

Another aspect that has interested me is if possible to shrink the channel bandwidth down from 20Mhz to 10Mhz or 5Mhz. Given that we’re NOT looking for bandwidth, but we ARE looking for maximum number of clients (per AP, but also for our total area of install ). We’re thinking it might be useful to REDUCE the channel width. This way we could move from 3 non-interfering channels for our install to 6 or even 11 non-interfering channels possibly allowing us to use a RouterBoard with 4 radios increasing the total to 768 per AP. if not more.

You’re right, reducing channel bandwidth will provide you with a wider useable spectrum.

For this to be possible, you’ll need to check that both AP and wireless station radios specific chipsets support desired channel width.

Also, using TDMA (specifically Nv2 in routerOS implementation) will allow for simultaneous wireless client count to scale better and also prevent the hidden node problem; this will force you to use same brand for AP and wireless stations.

Number of simultaneously connected stations is not limited by any means, but for physics; a single, bad connectivity station can drive the AP radio mad and hinder its performance with all the rest of stations. TDMA helps into achieving more simultaneously connected stations to perform stably too.

Without knowing more specifics about this project I couldn’t be 100% sure, but most of the times is better, and cheaper, to use several (i.e. 6 single radio APs) instead of two, three radios APs. It allows to place them more homogenously, and as you add CAPsMAN, managing all of them centrally will be as convenient and easy as a single one.

As I understand it each light has its own integrated wifi module. This rules out nv2/tdma. Regarding 10mhz channels they are a great idea but have to be supported by the light.