How many beacon devices can be scanned by one gateway？

RE: How many beacon devices can be scanned by one gateway？

ovrebekk — Wed, 01 Sep 2021 06:36:47 GMT

The chart simply calculates the statistic probability that two packets will overlap in time, and makes the assumption that in the case of two packets overlapping both packets will be lost.

In practice this might be slightly pessimistic, and you could have situations where two packets overlap and one of them is still successfully received (if one of them have significantly higher signal strength for instance), but in real life you could also have interference from other sources causing packet loss that the excel sheet is not able to predict.

The excel sheet should only be used as an indication of what kind of self interference to expect. It has not been verified with thousands of nodes in a practical test.

Best regards
Torbjørn

RE: How many beacon devices can be scanned by one gateway？

Clever Huang — Tue, 31 Aug 2021 07:48:28 GMT

hi ovrebekk，

Thank you very much for your reply, I got a lot of useful information. But there is one more question I would like to ask. What is the specific basis for the calculation results in the chart? thanks again.

RE: How many beacon devices can be scanned by one gateway？

ovrebekk — Mon, 30 Aug 2021 17:18:09 GMT

Whenever several beacons are active in the same area you will have a chance of packet collisions, and this chance increases with the number of beacons in the area, and how often they each are advertising (the advertising interval).

I answered a similar question some time back, and at that time I made an excel sheet to try and estimate the amount of packet loss caused by having multiple beacons within range:

devzone.nordicsemi.com/.../PacketLossCalculator.xlsx

In that document you can enter the packet length in microseconds, which depends on how much advertising data you are sending, the number of beacons in the area, and the total time over which the calculations are applied.

The advertising interval is defined as a percentage of the total time, and the graphs try to visualize what is the best pick of advertising interval in order to minimize packet loss and/or increase beacon reception on the scanner.

The first graph visualizes overall packet success rate (how many of the advertise packets are actually received).

The second graph is probably more important, and shows for how many of the beacons the scanner will receive at least one packet within the total time. Inversely, for how many beacons are all packets lost because of interference within this time interval.

As a tip, if you are actually planning to have up to 10000 beacons advertising in an area, you can increase reception rate considerably by using three receiver chips on the scanner side so that you can scan on all three channels at the same time.

Best regards
Torbjørn