HMIPv6 in NS-2
 

Hi, I have installed fhmip extension in ns-allinone-2.33 . for HMIPv6 i just un blocked define map-mod in map-reg.cc and blocked all other defines and fhmip-simula.tcp is runing. i wrote the awk to calculate the end-to-end delay for packet but the result is strange. my code is:
BEGIN {

seqno = -1;
droppedPackets = 0;
count = 0;
num = 0;
flag = -1;
}
{
if ($1 == "+" && $5 =="tcp" && seqno < $12){
seqno = $12;
temp = seqno +1 ;
flag = 1;
} else if (seqno == $12) {
flag = -1;
}
if (seqno >-1 ){

if ($1 == "+" && $5 =="tcp" && flag == 1){
start_time[seqno] = $2;
packetID[seqno] = seqno;
time[seqno] = $2;

} else if(($7 == "tcp") && ($1 == "r") && $6 == seqno) {
end_time[seqno] = $2;

}
} ###if seqno > -1

}


END {

for(i=0; i<=seqno; i++) {
if(end_time[i] > 0) {

delay[i] = end_time[i] - start_time[i];
print time[i]" " delay[i] ;
count++;
} else
{
delay[i] = -1;
}
}

}
could anybody please have a look and see if it is correcr or not please?

many thanks

Hannaneh

handover latency
 

I have installed sesons FHMIP1.3.1 extension in ns2.31.I am able to run the simulation properly for mipv6,hmipv6,fmipv6,fhmipv6 (by following same instrucion set).Now i want to compare teh results interms of throughput,handover latency,loss rate?
Can anybody guide me in right direction?

Handover Latency
 

look at this paper "A Comparison of Mechanisms for Improving Mobile IP Handoff Latency for End-to-End TCP" which has been wrote by the person who wrote the extension for NS-2.

Handover Latency
 

Thanx Hannaneh, i have checked the paper but there is no hints on how to calculate the handover latecy (i.e. the last packet send from PAR and the first packet recived from the NAR) using awk.Can you please guide me on this?

there is a file named fhmip_plot_seq.pl which calculats the handover delay and shows the plot.

awk scripts
 

I have already checked the "plot_seq.pl" file it is for "Sequence number extraction" only. It creates two separate file for "send" and "receive" for each node and shows the sequence numbers in it.I have tried the mohit's awk scripts[http://mohittahiliani.blogspot.com/2...-for-ns2.html] but the results are strange "pdr" and "throughput" is very less for fhmipv6 as compared to other schemes (i.e mipv6,hmipv6,fmipv6).Have you tried that? also there is "end to end delay" which is different from handover latency.Can you please help me on this?

Regarding handover delay, there is clear specification about it in the paper: the time between the first time that packet has re-transmited and the last time that packet was send by sender, this time is clear in send0.log. you need to find the seq no that has been re-transmitted. the plot is base on seq no and time.

trace analysis
 

I have checked the plot_seq results but not able to get the packet which is transmitted from PAR.Can you please help me on this?.I am attaching here with fhmipv6.tcl (tcl script), plot_seq.pl (plot file), send_log(output of plot_seq file) .I am using udp for tarce analysis.

Attachment 7499

Attachment 7500

Attachment 7502

Hi Sangram,

Could you please explain how you successfully installed the FHMIP1.3.1 extension ns2.31,I would really appreciate as I am having all sorts of problems using FHMIP2.31 patch by Pedro on ns2.31.Thanks.

FHMIP1.3.1 Installation
 

Hello silvapearl ,

I have installed seasons FHMIP1.3.1 extension in ns2.31.

# Download links :
1) http://translate.google.co.in/transl...%26aid%3D16409
[Note: On "www.baisi.net" you can find lot of good resources related to ns2 installation,simulation.]

2) http://www.pudn.com/downloads103/sou...ail422234.html

# Installations instructions are clearly mentionded in pdf document..just rename pdf document to "FHMIP1.3.1.pdf" and you can view it (also you translate it with google for more clear understanding).

If you still found some problem, i can send extension just email me.

For your information i have tried this on winxp + cygwin + ns2.31 + fhmip1.3.1

Hi Sangram,

Thank you very much for your response. I have indeed successfully installed it and run the fhmip_simula.tcl scripts successfully.I however have another issue,when I want to run fmip or hmip or fhmip,i need to block out certain this out in the mip-reg.cc file and then run the tcl script again,this works just fine. But I cant make any sense of the trace files,I cant tell the difference in the results of the different(mip,fmip or fhmip) schemes I am running. I even checked out the fhmip_plot_seq.pl,that didnt make much sense either.

Please how do I evaluate the performance of these schemes e.g in terms of plotting graphs,etc.

Sorry to overwhelm you with so many questions,I am pretty new to ns.

Basically I am asking the next thing to do in order to determine delay and packet loss of the 3 different schemes(mip,fmip and fhmip) and then plot and compare these.

Script analysis
 

Hope this will help you ,
http://translate.google.co.in/transl...ails%2F3551405

Thank you once again Sangram for your response.I have visited the site,but because it is being translated by google,its a bit difficult making sense of what is written.
I have a few more questions,have you now been able to successfully compare the latency,throughput,loss rate etc, of mip,hmip,fmip and fhmip? If you have can you please attach the awk script you used.

The mohit script you mentioned above did you make modifications to it before you used it?

Also the "send_log" file you attached above,where is this located,because I see it defined in the plot_seq.pl file but I don't see the "send_log" file anywhere.

Thanks,I look forward to your response.

You need to have TCP (Not UDP)traffic between MN and CN as the plot_seq.pl just filter TCP and shows the sequence numbers. plot_seq.pl just works for Handover latency , for other ones (i.e. packet lost, ...) you need to write your own script. but i recommend you first try to get some results for handover latency and compare it with the result in the paper which was written by the person who wrote the fhmip patch.
hope it works :)