Page 2 of 2
Re: Details Regarding "Received packet radio metadata header"
Posted: Sun Jan 13, 2019 12:40 am
by thethinker
I'm still trying to find out how to read this clock. That's the only way I can think of which lets me find the offset between the HW clock and real clock. Please help!!
Re: Details Regarding "Received packet radio metadata header"
Posted: Wed Feb 13, 2019 7:42 pm
by thethinker
Hello ESP engineers,
Could someone PLEASE answer this?! I have been waiting for months.
Thanks
Re: Details Regarding "Received packet radio metadata header"
Posted: Thu Feb 14, 2019 3:50 am
by ESP_Sprite
At the moment, there is no way. I've asked the WiFi team if they would be willing to add one. Do note that at the moment, the timestamp is added in software, when the WiFi stack processes the packet, so you'll still not have the exact time the radio received the packet. I'll also ask if we can revert to hardware timestamps, for non-802.11B (so for G/N/AC/...) signals we should be able to do that.
Re: Details Regarding "Received packet radio metadata header"
Posted: Tue Apr 30, 2019 7:29 pm
by matteof93
thethinker wrote: ↑Wed Feb 13, 2019 7:42 pm
Hello ESP engineers,
Could someone PLEASE answer this?! I have been waiting for months.
Thanks
Hi, at the end I gave up on this method to retrieve the time at which the packet was received.
On the other hand I found out that using NTP is a good option and it works perfectly, basically I have 4 ESP32 that listen for PROBE REQUEST packets and each time the callback is triggered they get the system time as soon as possible (system time set previously with NTP).
I don't know how much this clock is skewed with respect to the real clock of a NTP server but I can tell you that all the 4 ESP32 give me very very similar timestamps as arrival time of the packet...in general the difference is within 0.05 seconds.