This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

sniffer time accuracy using ble_sniffer_1.0.1 and nRF51 dongle

Hi Guys,

Hopefully this is not a duplicate question.

I always try to search before I post but there are so many ways that people can ask a question, all technically correct.

Using ble_sniffer_win_1.0.1 software, it appears that the timing of the packets being reported in Wireshark is incorrect. I say this because i'm comparing to another sniffer (okay, you caught me -- the TI product).

For the Nordic sniffer, I'm using the nRF51 dongle.

As an example, I have a sniff I've done of simply listening to advertising from a board which is coming at 100mS intervals. Instead of showing a steady stream of between 100 and 103mS, on the wireshark trace, I show some intervals of just a few mS, and then the normal, expected 100mS or so after that.

!I was unable to drop the sniffer trace in line, the preview was okay, but the post lacked any formatting! I therefore attached the file instead.

sniff_report.txt

Is the strange timing expected? Just an outgrowth of using pipes?

Has anyone done anything to correct it? Could a different set of hardware help the problem? Could this be tied into the exact USB port I'm using, machine configuration or something else? I can certainly say that I have not been very careful in setting up this machine, but the TI sniffer dongle doesn't report the same instability in the timing.

Or is this not duplicated in your labs?

I'd like to be able to count on my sniffer tools, if at all possible, for ground-truth data.

I'll just answer your questions inline here.

Feels a little wrong, but not too sure about these forums and this should suit the purpose.

Yes, I did select the device. I had two devices running for this log session that I attached, I selected 1 for the TI device that was advertising and then brought up Wireshark with the w option.

Once that was done I recorded for a bit and then stopped the capture and zipped up the information you see attached here

logs.zip

So I guess from the response you've not seen this sort of strange timing. Your log for a device that is simply advertising at 100mS intervals, then, would be totally solid.

That's unfortunate because I don't know much that I can do differently.

Thanks for any pointers.

-thanks -e

Parents Reply Children
No Data
Related