[ndnSIM] Testbed nodes geographical locations in mini-ndn

[Junxiao Shi]

 Dear folks

While I haven't used the visualization feature of Mini-NDN or ndnSIM, I do
know a thing or two about the coordinates.

Since results should not be affected and it seems this is a visual problem,
> if you need that for a paper, you can take a screenshot of the testbed from
> the testbed webpage and put it in your paper to demonstrate the topology.
>

ndnmap offers a JSON feed that contains WGS84 coordinates of every router
on the global NDN testbed.
Each WGS84 waypoint represents a point on earth's surface. They are
accurate to the city and may not indicate the exact location of the device.
http://ndnmap.arl.wustl.edu/json/geocode/
There's no schema for this feed, so you have to guess the semantics.


> Topologies in ndnSIM are created based on x, y coordinates.
>
ns-3 visualizer wants coordinates on a 2-dimension grid. You can't directly
use WGS84 coordinates because they represent points on a sphere. There are
many algorithms to project a world map onto a flat sheet of paper.
Visualizing these coordinates would look like http://ndnmap.arl.wustl.edu/

I am not sure if you can covert angle and radius to x, y coordinates
> somehow
>
Mini-NDN's "angle and radius" are hyperbolic coordinates. It's neither
WGS84 nor 2-dimension grid, and has nothing to do with geolocation. They
are calculated based on network connectivity (especially BGP AS-level
topology) by the math gurus at Northeastern University.
Visualizing these coordinates would look like NDN retreat 1703
Geohyperbolic Routing and Addressing Schemes
<https://www.caida.org/workshops/ndn/1703/slides/ndn1703_ivoitalov.pdf>
slide 4.

In terms of results, I do not think it will make any difference (the thing
> that will make a difference is the link delay).
>
The bottom image on http://ndndemo.arl.wustl.edu/ has the link delay of
every link on NDN testbed. They are measured with IP ping when
commissioning a node.

<https://www.caida.org/workshops/ndn/1703/slides/ndn1703_ivoitalov.pdf>
Yours, Junxiao
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndnsim/attachments/20190126/02ef1424/attachment.html>