Table of Contents
Fetching ...

Trust, But Verify, Operator-Reported Geolocation

Katherine Izhikevich, Ben Du, Sumanth Rao, Alisha Ukani, Liz Izhikevich

TL;DR

An in-depth analysis of operator-misreported geolocation is provided using a bandwidth-efficient methodology and a continually updated dataset of RIPE Atlas vantage points that misreport geolocation is released.

Abstract

Geolocation plays a critical role in understanding the Internet. In this work, we provide an in-depth analysis of operator-misreported geolocation. Using a bandwidth-efficient methodology, we find in May 2024 that only a small percentage (1.5%) of vantage points in the largest community-vantage point collection, RIPE Atlas, do not respond from their operator-reported geolocation. However, misreported geolocations disproportionately affect areas with limited coverage and cause entire countries to be left with no vantage points. Furthermore, the problem is escalating: within the past five years, the number of probes reporting the wrong location has increased ten-fold. To increase the accuracy of future methodologies and studies that rely upon operator-reported geolocation, we open source our methodology and release a continually updated dataset of RIPE Atlas vantage points that misreport geolocation.

Trust, But Verify, Operator-Reported Geolocation

TL;DR

An in-depth analysis of operator-misreported geolocation is provided using a bandwidth-efficient methodology and a continually updated dataset of RIPE Atlas vantage points that misreport geolocation is released.

Abstract

Geolocation plays a critical role in understanding the Internet. In this work, we provide an in-depth analysis of operator-misreported geolocation. Using a bandwidth-efficient methodology, we find in May 2024 that only a small percentage (1.5%) of vantage points in the largest community-vantage point collection, RIPE Atlas, do not respond from their operator-reported geolocation. However, misreported geolocations disproportionately affect areas with limited coverage and cause entire countries to be left with no vantage points. Furthermore, the problem is escalating: within the past five years, the number of probes reporting the wrong location has increased ten-fold. To increase the accuracy of future methodologies and studies that rely upon operator-reported geolocation, we open source our methodology and release a continually updated dataset of RIPE Atlas vantage points that misreport geolocation.
Paper Structure (20 sections, 7 figures, 2 tables)

This paper contains 20 sections, 7 figures, 2 tables.

Figures (7)

  • Figure 1: CDF of Error of Distance ---50% of misreported probes are over 900 miles away from their reported locations.
  • Figure 2: Number of violating probes over time---Using only RIPE measurements, the number of violating probes has increased at least tenfold from January 2019 (0.1% of all responding RIPE Atlas Probes) to May 2024 (0.5% of all responding RIPE Atlas Probes, since we do not have historical Ark calculations).
  • Figure 3: Number of Weeks Between SOI Violation and Geolocation Update---Of the probes that do not disconnect, it takes over 7 weeks for 50% of such probes to update their metadata. Notably, the SOI violation no longer occurs.
  • Figure 4: Distance Between Reported Locations Before and After SOI Violations---Roughly 80% of operators whose probes stopped violating updated their location by over 1K miles.
  • Figure 5: Moving from Germany to Brunei Case Study---The operator's probe reports a SOI violation between February 2022 and April 2023, before updating their location metadata from Germany to Brunei in April 2023.
  • ...and 2 more figures