Frame error rate in WifiRemoteStationManager

217 views
Skip to first unread message

Morten Lindeberg

unread,
Mar 31, 2011, 4:43:17 AM3/31/11
to ns-3-users
Hi,

We are working on cross-layer enabled adaptations in MANETs, and
running all our simulations in ns-3. During the collection of link-
layer metrics for determining the state of wireless links, we came
across a method in wifi-remote-station-manager.h/cc which returns a
metric seem to represent the link state very well (GetFrameErrorRate
()). It seems to calculate a measure of frame error rate making,
however I cannot see anywhere in the literature where such an equation
is described for doing so.

If anyone can point us to some reference where this way of calculating
the frame error rate is described, we would be very happy! Any idea
anyone?

Best wishes,
Morten

Nicola Baldo

unread,
Apr 1, 2011, 4:27:42 AM4/1/11
to ns-3-users


On Mar 31, 10:43 am, Morten Lindeberg <mortenlindeb...@gmail.com>
wrote:
Have a look at these:

G. Pei and T. Henderson, "Validation of ns-3 802.11b PHY model"
http://www.nsnam.org/~pei/80211b.pdf

M. Lacage and T. Henderson, "Yet another Network Simulator"
http://cutebugs.net/files/wns2-yans.pdf

Regards,

Nicola

Morten Lindeberg

unread,
Apr 1, 2011, 8:34:43 AM4/1/11
to ns-3-users
Hi Nicola,

Although the papers give a great insight into how various PHY layer
parameters are calculated, I could not seem to find any mentioning of
the specific frame error rate calculation that I was looking for. I
guess I should continue searching :)

Thanks,
Morten

Konstantinos

unread,
Sep 14, 2011, 6:59:13 AM9/14/11
to ns-3-...@googlegroups.com
Dear Morten,

Have you found anything considering MAC Frame Error Rate ??

I'm also working on cross-layering designs and I want to use this metric but first I want to know how it is calculated etc.


Regards,
Konstantinos

Morten Lindeberg

unread,
Sep 14, 2011, 7:28:39 AM9/14/11
to ns-3-users
Hi,

By looking at the function, it seems they are using some kind of
exponentially weighted moving average, could not find anything in the
documents from Nicola. I guess it was just the developer of that added
this function for testing or debugging purposes.

Morten

Konstantinos

unread,
Sep 14, 2011, 8:36:11 AM9/14/11
to ns-3-...@googlegroups.com
Thank you Morten for that.

I want to include it in my design but I assume that I have to use a propagation model that creates some errors in the channel (e.g. nakagami) in order to see Frame Errors.

Also, this error rate is averaged over all links. If a node has 2 links and one is bad and one is good, you can not distinguish from MAC, you need PHY layer info.

Konstantinos
Reply all
Reply to author
Forward
0 new messages