Details about PhyRxDrop

162 views
Skip to first unread message

Rukna

unread,
Feb 23, 2015, 2:54:31 PM2/23/15
to ns-3-...@googlegroups.com
Dear NS-3 users,

Can anyone tell me what is exactly PhyRxDrop count? I am trying to calcualte the number of packet drop in physical layer. I hook PhyRxDrop to calculate it. When I print print the PhyRxDrop count, it is always quite high (~40). However, I just send one packet at each second. I do not under what is going on?

Thank you

Konstantinos

unread,
Feb 23, 2015, 3:03:09 PM2/23/15
to ns-3-...@googlegroups.com
Hi Rukna,

From the API:
  • PhyRxDrop: Trace source indicating a packet has been dropped by the device during reception
It depends on where you hook it. If you use the '*' (wildcard) for NodeList, it will fire for every node, so you might see large number of Drops even with small number of packets.
Also, depending on the MAC and the type of packet, you might have also re-transmissions, so a single application packet might be dropped several times at Phy.

Rukna

unread,
Feb 23, 2015, 3:19:56 PM2/23/15
to ns-3-...@googlegroups.com
Thank you for your reply.

Annie

unread,
Apr 19, 2017, 2:44:37 AM4/19/17
to ns-3-users
Hi Konstantinos,

According to your post, the PhyRxDrop will be triggered by retransmitted packets, right? May I ask will flow monitor count the original packet and its retransmission as ONE packet, or it counts the original packet and retransmission as TWO/multiple packets? Thanks in advance.


在 2015年2月23日星期一 UTC-5下午3:03:09,Konstantinos写道:

Konstantinos

unread,
Apr 19, 2017, 3:07:48 AM4/19/17
to ns-3-users
Dear Annie,

Please do not spam the list with the same question on several threads, especially when these threads are old and not relevant.
PhyRxDrop, as mentioned above, will be triggered when a packet (any packet, could be re-transmitted) is dropped at the NetDevice (PHY layer) during the reception. This is probably due to low SINR.

FlowMonitor monitors packets at IP layer. MAC-layer re-transmissions or fragmentations that could potentially generate increased number of packets on the wire, are not counted, because at the end MAC should return one IP packet to upper layer.
Reply all
Reply to author
Forward
0 new messages