[Ndn-interest] Prefix registration in wireless
lixia at cs.ucla.edu
Thu Jul 20 00:03:23 PDT 2017
looks like no one has responded here, let me just throw in a few quick words:
> On Jul 12, 2017, at 12:13 PM, Adhy Satya <adhysatya820 at gmail.com> wrote:
> Hi all,
> Until now I couldn't find an answer for this question. I understand that NFD was developed for wired scenarios, but can be also used in the wireless scenario by allowing outgoing packets on the incoming faces.
regarding the issue of allowing an interest packet going in/out the same interface: yes, this has been a long recognized need from the earlier work on vehicle networking led by Prof. Giovanni Pau (you can find a few papers on this work if you just search his name on NDN publication list).
I recall this was discussed multiple times in NFD calls, though I lost track of the latest status on the implementations. Alex and Junxiao?
Regarding another issue raised in your earlier msg (In a wireless mesh scenario, how does NFD populate the RIB): there are multiple solutions, depending on specific app scenarios, some may fit better than others
- the vehicle networking papers develop some solutions
- Beichuan's group has done work with name/prefix discovery, you probably can find some paper from NDN publications
- there is also ongoing implementation (no publication yet) using geo location based forwarding for rooftop mesh network
> In my current scenario, I pre-register the prefixes and faces using the "nfdc register" command (I'm using NFD 0.5.0). In the event of a new node joining the network, does it need to register the remote faces in order to start communicating? Is there a workaround/node discovery service in NFD?
> Thank you
> On Sun, Jun 25, 2017 at 12:08 PM, Adhy Satya <adhysatya820 at gmail.com <mailto:adhysatya820 at gmail.com>> wrote:
> In a wireless mesh scenario, how does NFD populate the RIB? i.e., do nodes need to register all the nodes in the network for a certain prefix?
> Also, if a node joins the network, how does it start sending and receiving packets?
> Could you comment on that process please?
> Thank you,
> Ndn-interest mailing list
> Ndn-interest at lists.cs.ucla.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ndn-interest