BSS Color making no changes in network performance

104 views
Skip to first unread message

Brociu

unread,
Oct 30, 2021, 8:20:17 AM10/30/21
to ns-3-users
Hello to You all

I'm using ns 3.33, and I want to conduct simple two-BSS (one Access Point, one Station in each BSS, total of 4 devices) simulation with BSS Color on and off and compare how some parameters (in example distance, RTS, aggregation and so on) combined with BSS coloring can improve network performance. I've got some more complex plans on expanding to more BSSes and much more devices in the future, but first I've to sort out issues with my current simulation.

Piece of code, responsible for colour implementation looks like this
deviceAp->GetHeConfiguration () -> SetAttribute ("BssColor", UintegerValue (1));

Sadly with my configuration (IEEE802.11ax 5GHz, channel 36, channel width 20MHz, STAs 12 meters apart from each other, APs 20 meter apart, AP-STA distance 4 meters), it seems that coloring didn't work. I've received same results in throughput, jitter and delay in every simulation which I'm conducting (30 seconds of simulator run, grabbing results from 10th to 30th second, filtering out possible trash which can occur durring first 10 seconds). To grab results I'm using FlowMonitor. I've changed distances, channel width several times.

To my great sadness, I've received simulation code from another person, which he believes was working well on ns 3.31, working on similiar basis as mine, but changing CCA Threshold from -82 to -62 with |2| step and distances between BSSes rising with 50m step from 0 to 350m - this gives a lot of simulations, but also when running this code with and without coloring, I came across SAME results for throughput, jitter and delay.

I've even added another line to the code, below previously mentioned:
deviceAp->GetHeConfiguration()->SetBssColor (ColorA);
with of course ColorA defined on top of my code as uint8_t ColorA = 1;
Nothing sadly has changed.

I'm totally tore apart, as I spent almost two months now sitting every other evening and even some nights, performing simulations and seeking for even small code mistake and it led me to no conclusions.

Am I doing something wrong? Is there significant issue with 3.33 version of ns and that's why I'm struggle?

Can someone point me at simplest BSS color implementation which actually works and has even slight difference in given results of performance? For now, giving all of my simulations, only conclusion of mine is that great invention of BSS coloring, isn't even working.

Looking forward to get some tips which will help me, I'm open to provide every data and info needed.

Cheers and thanks in advance

Ali Karakoç

unread,
Oct 30, 2021, 8:42:16 AM10/30/21
to ns-3-...@googlegroups.com
Hello,

BSS color does not increase the network performance on its own. This concept should be used within OBSS/PD mechanism. 

See this example:


Best regards,

Brociu <tekba...@gmail.com>, 30 Eki 2021 Cmt, 15:20 tarihinde şunu yazdı:
--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ns-3-users/537d6462-6d95-4ea9-8f78-ec9edf3613c4n%40googlegroups.com.

Brociu

unread,
Oct 30, 2021, 9:10:04 AM10/30/21
to ns-3-users
Thank You very much for Your's kind answer.

Of course I'm aware of it and my code sports OBSS/PD mechanism, in a very similiar way as in example You've provided. Altough Your message highlighted me one issue with my code, which probably leads to wrong results. I'm getting back to code and will be back with (I hope not) next issues.

Thanks!
Reply all
Reply to author
Forward
0 new messages