Improving the accuracy of measurement-based geographic location of Internet hosts

Network Resilience Analysis: Review of Concepts and a Country-level. Case Study

“…An important observation is made on the correlation of router and population densities, which has been confirmed and elaborated further in [19]. Contrary to such a passive approach, [30] suggests the deployment of active probes and application of a precise-location algorithm based on triangle inequality, where distance is represented by packet delay. The algorithm operation is based on a set of nodes with a preciselyknown geographical location (a.k.a.…”

IP geolocation in metropolitan areas

“…Delay based geolocation techniques use two sets of nodes: a) probe nodes, which initiate pings to the other nodes, and b) landmark nodes, which respond to pings sent by the probe nodes [27,23]. Both the probe nodes and the landmark nodes have known locations.…”

On the real impact of Path Inflation in networks under production

“…Basically, it is expected that hosts placed in a similar geographical distance present similar delays measured typically by means of the ping tool. Such correlation has been found in some regions of the Internet, essentially North America and west Europe [13], but the precision is limited in the rest of the world. As the target of our work is to span all the possible destinations in the world, this precision depending on the area represents a significant caveat.…”

“…The former class includes mechanism based on the delays between a set of reference nodes-landmarks-and the target node. Examples of this are [11], [13], [14]. This approach is based on the linear correlation between the delay in the networks and the geographical distance between the objective and the set of landmark nodes.…”