WARP Project Forums - Wireless Open-Access Research Platform

You are not logged in.

#1 2018-Mar-26 21:15:52

why_here
Member
Registered: 2018-Mar-21
Posts: 8

Missed part of the reception in 802.11 Reference Design v1.4.0

Hi.
I use three WARP v3 and a phone for testing. Two WARP board are configured as AP, and the remaining WARP and phone are connected to the AP as STA. I use wlan_exp to have the AP send LTGs to connected STA. Then merge the three log files, and then I found that sometimes WARP nodes will miss the reception of frames.

444<---LTG---422(AP)       830(AP)---LTG--->phone

The sequence of send and received frames in the processed log file is as follows:

Code:

...
timestamp,Tx or Rx, pkt_type, node
11253.0,Tx,RTS,422
11276.0,Rx,RTS,444
11319.0,Tx,CTS,444
11349.0,Rx,CTS,422
11383.0,Tx,DATA,422
13061.0,Tx,ACK,444
13091.0,Rx,ACK,422

16741.0,Tx,RTS,830
16770.0,Rx,RTS,444
16774.0,Rx,RTS,422
16834.0,Rx,CTS,830
16869.0,Tx,DATA,830
18573.0,Rx,ACK,830
21241.0,Tx,RTS,422
21261.0,Rx,RTS,830
...

In the first part of the log above, node 830 missed all the frames. In the second part, nodes 444 and 422 missed the CTS and ACK frames.
Any advices?

Offline

 

#2 2018-Mar-27 15:31:41

chunter
Administrator
From: Mango Communications
Registered: 2006-Aug-24
Posts: 1171

Re: Missed part of the reception in 802.11 Reference Design v1.4.0

It's hard to say; there are lots of way for a packet to be lost and not show up in an Rx log. It could be some hidden node effect or other asymmetric reaction to external interference. If you can, controlling the experiment with RF cabling, attenuators, and power splitters might make things easier to analyze (though that's likely not possible with the use of a smartphone). Also, you should consider updating to a newer version of the design. We've made lots of improvements to the design in the last 2+ years since the release of v1.4. It's possible the new version will work better for this particular effect you are seeing.

Offline

 

Board footer