NS-3 trace file packet ID

0
votes

I have a simple point to point UDP WiFi simulation in NS-3 that outputs data to a trace file. It provides lots of useful data but there is no information that gives a unique ID for each packet. I can't find anything in wireshark either when I open the pcap files.

I have output the results of my simulation to an ascii trace file and pcap files for both nodes but I can't find any packet identifier. I can see the sequence numbers of the packets but that's it.

I am new to NS-3 so I am not sure how to produce this information.

Here is some of the output from the trace file.

t 2.00082 /NodeList/0/DeviceList/0/$ns3::WifiNetDevice/Phy/State/Tx ns3::WifiMacHeader (DATA ToDS=0, FromDS=0, MoreFrag=0, Retry=0, MoreData=0 Duration/ID=0usDA=ff:ff:ff:ff:ff:ff, SA=00:00:00:00:00:01, BSSID=00:00:00:00:00:01, FragNumber=0, SeqNumber=0) ns3::LlcSnapHeader (type 0x806) ns3::ArpHeader (request source mac: 00-06-00:00:00:00:00:01 source ipv4: 10.1.1.1 dest ipv4: 10.1.1.2) ns3::WifiMacTrailer ()

Any suggestions are appreciated.

Thanks.

1
wifisimulationwiresharkpacket-capturens-3

1 Answers

1
votes

In case you might not be aware of this already, let me first point out what might seem to be the obvious but: "there is no such thing as unique packet id in real networks" and since pcap traces are designed to contain dumps of real packets in real networks, there is zero chance you will be able to find a unique packet id in a pcap trace generated by ns-3.

Now, ns-3 does contain a per-packet unique id that is available with the Packet::GetId method and you can trivially change the source code of the function that generates your ascii dump to add this in src/wifi/helper/yans-wifi-helper.cc. Grep for "Ascii".

Now if you want to know why it does not do this by default because it is so useful, I honestly can't remember but:

  • there is probably something related to the ns2 trace format that inspired this ascii format. Compatibility with existing tools might have been an issue

  • adding a packet id goes against the ns-3 philosophy of matching what real networks do