Relation between advertising interval and number of packages received

[gdeblasio]

Hi all,

BLE newbie here.

I have a peripheral device (iBKS105 beacon) configured with Eddystone and iBeacon protocols, emitting on channels 37, 38 and 39. I set the advertising interval to 1000ms, and the sampling time to 1 sec. I also have a central device (nRF51 dongle) from which I do not know the scan interval and scan window. I capture packages through Wireshark. With the above data and theoretically speaking Wireshark should receive (for the sampling time=1sec) a maximum number of packets equal to 3 channels x 2 protocols = 6, however I receive a number of packages around of 26-30. How is this possible?

Is there a relation between advertising interval, scan interval, scan window, sampling time and the number of packages received?

Thanks in advance!

[haakonsh]

You should only see 3x2 packets at an advertising interval of 1sec. 

Are you sure the beacons are all from the same address? There's a lot of BLE devices advertising around us all the time and the sniffer will sometimes pick up packets from other devices. 

Can you share a sniffer log?

[gdeblasio]

Hello haakonsh,

Thank you very much for your help!

Yes, the beacons are all from the same address (no other BLE device around).

I didn't mentioned that the problem arises using nRF Sniffer v2 + Wireshark 2.4.6

Attached you will find .pcap files of two tests: 1 beacon iBKS105  + 1 dongle nRF51 and in both tests I have set these parameters: Tx 0dBm, Advertising Interval 1000ms. Sampling time 1 second.
Test1: Eddystone protocol, Slot 1 (UID) only. 5460.Test01-Eddystone.pcap
Test2: Eddystone + iBeacon protocols, Slot 1 (UID, UUID) only  4405.Test02-Eddystone+iBeacon.pcap

Thanks again and best regards!

G.

[gdeblasio]

Hello haakonsh,

With respect to receiving too many advertising packets, I solved the problem going back to nRF51 Sniffer v1.01 + Wireshark 1.10.14 + JLink v616a instead of using Sniffer v2 + Wireshark 2.4.6 + JLink v616c. Maybe the problem is due to the beta version of the Sniffer (v2).

Thank you very much for your support!

G.

[haakonsh]

Hey Gabriel,

I'm glad that you've solved your problem

I'd like to investigate whether there is a bug in the sniffer v2 or not. Do you mind sharing the .hex file of your project so that I can capture with my sniffer? Also what DK did you use as the sniffer?

Cheers,

Håkon.

[gdeblasio]

Hello Håkon,
At the moment I'm not developing software so I'm not using any DK: I'm just using the dongle to capture information from the frames coming from the beacon using Wireshark, such as the RSSI or the channel. The .hex file I used in these tests was downloaded from the Nordic website and is included in the Sniffer zip file v1.01. Anyway, I'll attach it to you just in case. sniffer_pca10031_51296aa.hex

I attached one image of the errors that Wireshark 2.4.6 was throwing using Sniffer v2 and a .pcap file in case they were of any use to you. Regarding the .pcap file, you will see that from frame 1 to frame 44 the time increment is completely different with respect to frame 45 onwards. Also you will see that from frame 45 onwards I receive the expected packets number, 1-3 packets per second, according to advertising interval set in the beacon (1000ms).
7356.test.pcap

If you need more information do not hesitate to ask me.

Thank you very much!

Cheers,
G.

[haakonsh]

Thanks a lot, I'll try to replicate it on my end :)