RE: Is there a limit on iBeacon advertising devices in an area?

Giannis Anastasopoulos — Tue, 26 Apr 2022 05:26:27 GMT

@daviddedwin

Thank you very much for your response. May I ask what tool did you use for the simulation?

Also, you are correct on your assumptions, that I am going to use the Uncoded PHY and the Legacy PDU.

RE: Is there a limit on iBeacon advertising devices in an area?

daviddedwin — Fri, 22 Apr 2022 18:57:47 GMT

Hi,

I ran a few simulations on the usecase and hopefully the results are helpful. Do accept the answer if it was helpful.

Thanks
David

The advertising interval = 4000ms
Packet size is 47 bytes with a bit rate of 1 Mbit/s
Max jitter is 10ms
Clock drift is 250ppm

Assumptions:

The nodes are assumed to be arranged equidistant from the receiver so all of them can equally interfere at the receiver, which would be a worst case scenario.The nodes are close enough so that packets are well within the receive sensitivity threshold.

The receiver is assumed to be in RX all the time except for the channel switch time, Packet RX complete time. If a typical scanner is used then the interval should be as close to the window as possible for eqivalent results and real results are likely to be worse than the simulated results.

Results:

Simulation for 2000 nodes running the broadcaster role with non connectable advertising (same as the iBeacon).
All 2000 nodes were detected in a worst case of 21 transmissions. This implies that the scanner needed aboiut 64 seconds, since the interval was set to 4000ms.
The channel utilization was about 18.4%.

Results:

Simulation for 4000 nodes running the broadcaster role with non connectable advertising (same as the iBeacon).
All 4000 nodes were detected in a worst case of 36 transmissions. This implies that the scanner needed aboiut 144 seconds, since the interval was set to 4000ms.
The channel utilization was about 37.6%

Results:

Simulation for 5000 nodes running the broadcaster role with non connectable advertising (same as the iBeacon).
All 5000 nodes were detected in a worst case of 48 transmissions. This implies that the scanner needed aboiut 192 seconds, since the interval was set to 4000ms.
The channel utilization was about 47%