Re: [ns] Fwd: help me fix this trace
Dear Sir,
Don't worry about that, I will help you step by step to get the correct
results, firstly use the attach awk script to extract results from you
simulation output file using the command:
awk -f metrics.awk out.tr
and let me see the results, if its not worked well, there is another soultion..
Regards,
Mohammed AbuHajar
MSc Computer Science
University of Northern Virginia
- Original Message
From: adi suryo wibowo <[EMAIL PROTECTED]>
To: Mohammed Abu Hajar <[EMAIL PROTECTED]>; NS2
Sent: Saturday, December 1, 2007 12:19:46 AM
Subject: Fwd: help me fix this trace
Hii ns-User
I was read in ns mailing list and I found your name in contribute helping other
peoples.
I truly need your help..
I was simulated my tcl script in ns2.30 using new_trace format
I have out.tr then I want to calculate 3 performance metric :
paket delivery ratio, normalized overhead and end to end delay.
I use my awk script to get desire value but I don't know that was not success.
I get this value in my parsing result
send = 8871.00
recv = 0.00
routingpkts = 682.00
PDF = 0.00
NRL = inf
Average e-e delay(ms)= nan
No. of dropped data (packets) = 322
No. of dropped data (bytes) = 171304
I am attach .tr file in my mail, please help me what's wrong..?
why my recv and PDF = 0.00?
My awk script was wrong? or
can you give me correct calculation with my out2.tr in my attachment ?
What meaning of : NRL = inf
Average e-e delay(ms)= nan
thanks for your reply.
STTTelkom Student at Indonesia
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hsBEGIN {
sends=0;
recvs=0;
routing_packets=0.0;
droppedBytes=0;
droppedPackets=0;
highest_packet_id =0;
sum=0;
recvnum=0;
}
{
time = $3;
packet_id = $41;
#= CALCULATE PACKET DELIVERY FRACTION=
if (( $1 == "s") && ( $35 == "cbr" ) && ( $19=="AGT" )) { sends++; }
if (( $1 == "r") && ( $35 == "cbr" ) && ( $19=="AGT" )) { recvs++; }
#= CALCULATE DELAY
if ( start_time[packet_id] == 0 ) start_time[packet_id] = time;
if (( $1 == "r") && ( $35 == "cbr" ) && ( $19=="AGT" )) {
end_time[packet_id] = time; }
else { end_time[packet_id] = -1; }
#= TOTAL DSR OVERHEAD
if (($1 == "s" || $1 == "f") && $19 == "RTR" && $35 =="DSR")
routing_packets++;
#= DROPPED DSR PACKETS
if (( $1 == "d" ) && ( $35 == "cbr" ) && ( $3 > 0 ))
{
droppedBytes=droppedBytes+$37;
droppedPackets=droppedPackets+1;
}
#find the number of packets in the simulation
if (packet_id > highest_packet_id)
highest_packet_id = packet_id;
}
END {
for ( i in end_time )
{
start = start_time[i];
end = end_time[i];
packet_duration = end - start;
if ( packet_duration > 0 ) { sum += packet_duration; recvnum++; }
}
delay=sum/recvnum;
NRL = routing_packets/recvs; #normalized routing load = routing load but
it differ from routing overhead
PDF = (recvs/sends)*100; #packet delivery ratio[fraction]
printf("send = %.2f\n",sends);
printf("recv = %.2f\n",recvs);
printf("routingpkts = %.2f\n",routing_packets++);
printf("PDF = %.2f\n",PDF);
printf("NRL = %.2f\n",NRL);
printf("Average e-e delay(ms)= %.2f\n",delay*1000);
printf("No. of dropped data (packets) = %d\n",droppedPackets);
printf("No. of dropped data (bytes) = %d\n",droppedBytes);
printf("Packet Loss [%]= %.2f \n",
(droppedPackets/(highest_packet_id+1))*100);
}
[ns] Multi-interface support howto, by Ramon Aguero
Dear everyone, did any one implement suceessful the extension of ns-2 for multi-channel from Ramon Aguero at: http://personales.unican.es/aguerocr "many thanks for Ramon Aguero for his Document." I have some problems at Chapter 5, where to add Listing 5.5 and Section 5.3. in routingAgent.cc. (AODV.cc) I choice the AODV ad hoc routing agent to change. thanks
[ns] my trace file attachment
Iam really sorry I sent it again because my attachment do not appear in ns
mail. This is my awk and out.tr in my attachment
BEGIN {
sends=0;
recvs=0;
routing_packets=0.0;
droppedBytes=0;
droppedPackets=0;
highest_packet_id =0;
sum=0;
recvnum=0;
}
{
time = $3;
packet_id = $41;
# CALCULATE PACKET DELIVERY FRACTION
if (( $1 == "s") && ( $35 == "cbr" ) && ( $19=="AGT" )) { sends++; }
if (( $1 == "r") && ( $35 == "cbr" ) && ( $19=="AGT" )) { recvs++; }
# CALCULATE DELAY
if ( start_time[packet_id] == 0 ) start_time[packet_id] = time;
if (( $1 == "r") && ( $35 == "cbr" ) && ( $19=="AGT" )) {
end_time[packet_id] = time; }
else { end_time[packet_id] = -1; }
# CALCULATE TOTAL AODV OVERHEAD
if (($1 == "s" || $1 == "f") && $19 == "RTR" && $35
=="AODV")routing_packets++;
# DROPPED AODV PACKETS
if (( $1 == "d" ) && ( $35 == "cbr" ) && ( $3 > 0 ))
{
droppedBytes=droppedBytes+$37;
droppedPackets=droppedPackets+1;
}
#find the number of packets in the simulation
if (packet_id > highest_packet_id)
highest_packet_id = packet_id;
}
END {
for ( i in end_time )
{
start = start_time[i];
end = end_time[i];
packet_duration = end - start;
if ( packet_duration > 0 )
{sum += packet_duration;
recvnum++;
}
}
delay=sum/recvnum;
NRL = routing_packets/recvs; #normalized routing load
PDF = (recvs/sends)*100; #packet delivery ratio[fraction]
printf("send = %.2f\n",sends);
printf("recv = %.2f\n",recvs);
printf("routingpkts = %.2f\n",routing_packets++);
printf("PDF = %.2f\n",PDF);
printf("NRL = %.2f\n",NRL);
printf("Average e-e delay(ms)= %.2f\n",delay*1000);
printf("No. of dropped data (packets) = %d\n",droppedPackets);
printf("No. of dropped data (bytes) = %d\n",droppedBytes);
}
-
Bergabunglah dengan orang-orang yang berwawasan, di bidang Anda di Yahoo!
Answers
[ns] Forward Analysis trace file
I was simulated my tcl script in ns2.30 using new_trace format under linux I have out.tr then I want to calculate 3 performance metric : paket delivery ratio, paket overhead and end to end delay. I use my awk script to get desire value but I don’t know that was not success. I get this value in my parsing result send = 8871.00 recv = 0.00 routingpkts = 682.00 PDF = 0.00 NRL = inf Average e-e delay(ms)= nan No. of dropped data (packets) = 322 No. of dropped data (bytes) = 171304 I am attach .tr file in my mail, please help me what’s wrong..? My awk script was wrong? or can you give me correct calculation with my out2.tr in my attachment ? What meaning of : NRL = inf Average e-e delay(ms)= nan - Bergabunglah dengan orang-orang yang berwawasan, di bidang Anda di Yahoo! Answers
[ns] ns-2 FAQ
The Network Simulator ns-2: Frequently Asked Questions Note: This FAQ is now available at the project wiki: http://nsnam.isi.edu/nsnam/index.php/Ns_Users_FAQ _The mailing is is now subscriber only---please see the FAQ entry on "How do I post to the mailing list? Why was my post rejected?" for details._ * _Where do I get ns?_ From the ns web site at http://www.isi.edu/nsnam/ns/ns.html and the download page http://www.isi.edu/nsnam/ns/ns-tests.html. * _What platforms does ns run on and what kind of hardware do I need?_ Please see "where to start" on the building ns web page: http://www.isi.edu/nsnam/ns/ns-build.html#start. * _What should I do if I have trouble downloading/extracting ns?_ This question is answered in detail at http://www.isi.edu/nsnam/ns/ns-problems.html#downloading. * _What should I do if I encounter problems building ns?_ Check: 1. the README that comes in the distribution (very brief), 2. the "installation problems, bug fixes and help" web page http://www.isi.edu/nsnam/ns/ns-problems.html, 3. the archives of the ns-users mailing list http://www.isi.edu/nsnam/ns/ns-lists.html, 4. post a bug report (see below) http://www.isi.edu/cgi-bin/nsnam/reportbug.cgi. * _What do I do after I successfully build ns?_ + Put the path to your ns executable into your PATH environment + Put the path to your otcl into your LD_LIBRARY_PATH environment + Put the path to your tcl library into your TCL_LIBRARY environment * _Where can I find documentation for ns?_ All documentation is linked from the main ns web page http://www.isi.edu/nsnam/ns/. Documentation includes a tutorial (originally from Marc Greis) and a reference manual (ns notes and documentation). * _Words, words, words... that documentation is nice, but where are some sample scripts I can start from?_ Many sample scripts can be found in the ns distribution in ~ns-2/tcl/ex and ~ns-2/tcl/test. * _What protocols does ns support?_ A lot! Almost all variants of TCP, several forms of multicast, wired networking, several ad hoc routing protocols and propagation models (but not cellular phones), data diffusion, satellite, and other stuff. See the documentation (described above) for details, or download ns and look. * _How do I know that ns correctly implements these protocols?_ Ns has validation tests that cover many protocols, see http://www.isi.edu/nsnam/ns/ns-tests.html. However, ultimately users are responsible for verifying that ns is accurate for their purposes---since we cannot foresee all the ways ns may be used, we cannot test all cases with all inputs. * _Are there any contributed/additional protocols not in the main distribution?_ Yes, please see the contributed code web page http://www.isi.edu/nsnam/ns/ns-contributed.html. The mailing list archives can also be helpful (see below). * _How should I get started doing something (like implementing a new protocol or trying an experiment)?_ We recommend that you look through the tutorial (see documentation, above), then start with an example program that is most similar to yours (in the tutorial, or in tcl/ex or tcl/test in the distribution), and then start changing things. * _What should I do to compile ns to reflect my changes if I've modified some .cc or .h files?_ go to ns directory and run "make" or "make depend; make" * _How do I subscribe to the ns-users mailing list? How do I search old list archives? I can't take any more---how do I get off this list?_ To subscribe or unsubscribe, see http://www.isi.edu/nsnam/ns/ns-lists.html. The list archive is at http://www.isi.edu/nsnam/ns/ns-lists.html. * _How do I post to the mailing list? Why was my post rejected?_ As of June 2004 the ns-users lists allow posts from _subscriber only_. If you're not a subscriber, your posts to the list will be rejected. (This is unfortunately necessary to dispose efficiently of spam; manual filtering is too expensive.) We realize that the list is high traffic, so if you wish to post to the list without receiving messages on it, please subscribe and select the _no-mail_ option for your subscription. For details about the mailing list including mailing-list specific FAQ (for example, what if you're subscribed but still can't post), please see http://www.isi.edu/nsnam/ns/ns-lists.html. * _Did my post appear on the mailing list? (Or, why didn't my post appear on the mailing list?)_ See the http://www.isi.edu/n
[ns] please help out
I am working on 802.11e as well. 1. For me the easiest answer was to not worry about the ns trace file, and create my own. I count throughput at the MAC level by printing bytes sent during the simulation, and use a perl script to get the statistics afterward. 2. There are some tools to help you do this. http://www.tkn.tu-berlin.de/research/evalvid/EvalVid/docevalvid.html I am using an H.264 profile, but accuracy isn't important so I just used a mostly equivalent pareto VBR source. I hope this helps! Will Spearman Clemson University > hi all > > I have a very urgent and desperate call for help. I am using NS2-31 and EDCA > patch of TU Berlin, for 802.11e. Please guide me how to: > > 1. get the MAC level throughput trace file (the TCL code etc.) > 2. how I can evaluate the behaviour of H.264 over NS2 for some video frames. > Probably I am looking for a trace file containing the type of frames > (I,P,B), thier time stamps and sizes? Any tool to do it? > > Thank you very much. > > very kind regards > B
Re: [ns] effect of distance on throughput
Dear Julie
I have wanted to simulate effects of distance on throughput, for example
throughput of a MS with 200m as distance from BS , differs from throughput of a
MS 50m from BS.
ns has not this distance implementation. Thus , I have changed propagation
model to get that. if you are interested about this , i can explain it to u
detailed.
Now, about your question, i have not checked 10Km or more ... but if you dont
get any throughput for this distance, I think you must to decrease CSThresh_
and RXThresh_ amount in tcl/ns-default.tcl file for
mac/wireless-phy.cc::SendUp function :
...
if (Pr < CSThresh_) {
pkt_recvd = 0;
goto DONE;
}
if (Pr < RXThresh_) {
/*
* We can detect, but not successfully receive
* this packet.
*/
hdr_cmn *hdr = HDR_CMN(p);
hdr->error() = 1;
.
hope this help.
Best
"Haiying (Julie) Zhu" <[EMAIL PROTECTED]> wrote:
Hello,
Got your email from the NS list.
Could you help me on the distance issue? As far as I set MS and BS more
than 200m, the current model of 802.11 doesn't work, even I changed the Pt-
to 10e20. Actually, I'd like to see the throughput for 10Km or even
further, assuming enough power. Could you tell me which document to modify
to get it work on long distance?
Thank you very much for your help.
Julie
--
Haiying Julie Zhu,
Research Engineer, Ph. D, Wireless Applications and Systems Research |
Ingenieur de recherche, Ph. D. Applications sans fil et recherche sur les
systèmes sans fil
Communications Research Centre, Industry Canada |
Centre de Recherches sur les Communications, Industrie Canada
3701 Carling, P.O. Box | B.P. 11490, Stn "H"
Ottawa, Ontario, Canada, K2H 8S2
Phone | téléphone: 613-993-0716
Fax | télécopieur: 613-990-7987
E-mail | courrier électronique: [EMAIL PROTECTED]
On WWW, WISELAB is at | Visitez notre site web: http://www.crc.ca/wiselab
Government of Canada | Gouvernement du Canada
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Tara Arezoumand
Sent: Tuesday, November 13, 2007 1:15 AM
To: [EMAIL PROTECTED]
Cc: ns-users@ISI.EDU
Subject: [ SPAM ] Re: [ns] effect of distance on throughput
Thanks a lot ,
But 2 question :
1- for simulation of it, is there distance function that measures distance
between 2 nodes in ns?
2- which parts or files compute "throughput , packet loss and delay " in
wireless simulations?
Regards
"Kamil H. Suleiman" wrote:
Hello Tara,
To the best of my knowledge, once a node is inside a hearing range, we
don't see the difference say between 20 and 30 meters from the BS.
Probably some contributed module might be able to simulate the fading
effect.
Best,
Kamil
On Mon, 2007-11-12 at 00:55 -0800, Tara Arezoumand wrote:
> Hi Dear All
>
> I simulated a wireless network , with a BS and three MS ...
> I want to get MS throughput changes on become near or distant to BS..
> Is there any function in ns that has designed it ? if not please advise
me..
>
> hope to any help...
> Thanks
> Tara
>
> __
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
-
Be a better pen pal. Text or chat with friends inside Yahoo! Mail. See how.
-
Be a better sports nut! Let your teams follow you with Yahoo Mobile. Try it now.
