An Optimization Method for the Geolocation Databases of Internet Hosts Based on Machine Learning

Wang, Ting; Xu, Ke; Song, Junde; Song, Meina

doi:10.1155/2015/972642

Cited by 5 publications

(3 citation statements)

References 22 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our use-case scenario we do not assume to have access to the IP address. Other methods are restricted to a limited geographical area (such as China [11] [12], or Europe/US [13] [14] [15] [16]) and need information like network topology to model the speed. All these information are not accessible in the context of targets hidden behind a proxy, as ours.…”

Section: State Of the Artmentioning

confidence: 99%

ImMuNE: Improved Multilateration in Noisy Environments

Champion

Daciér

Chiapponi

2022

2022 IEEE 11th International Conference on Cloud Networking (CloudNet)

View full text Add to dashboard Cite

Identifying an attacker is a key factor to mitigate ongoing attacks. To evade localization, a single compromised machine can hide for months behind millions of available residential IP proxies. Without knowing the IP address of the machine, registration-based geolocation methods cannot be applied.Measurement-based methods have been proposed to estimate the location of a target without using its IP address. These methods use Round Trip Time (RTT) values and network speed modeling. They estimate a distance between the target and other observation points with known locations, called landmarks. However, most of these methods require additional information, whether it is on the topology of the network or the characteristics of the landmarks.In this paper, we present ImMuNE, a measurement-based technique which can estimate a location with only a few Round Trip Time measurements between a target and landmarks, even when some of these measures are inflated by temporary network congestion.Leveraging a previously made measurement campaign, we present promising results based on 11 millions TCP connections collected over a period of 4 months.

show abstract

Section: State Of the Artmentioning

confidence: 99%

ImMuNE: Improved Multilateration in Noisy Environments

Champion

Daciér

Chiapponi

2022

2022 IEEE 11th International Conference on Cloud Networking (CloudNet)

View full text Add to dashboard Cite

show abstract

“…Shavitt and Zilberman [12] proposed a landmark evaluation method based on the point of presence (PoP) level network analysis, which uses the majority voting algorithm to determine the city-level location of landmarks, and greatly improved their location accuracy. Wang et al [13] established an evaluation machine learning model based on the routing strategy and evaluated the city-level location of landmarks. ey significantly improved the accuracy of landmarks based on the reduction of evaluation costs.…”

Section: Introductionmentioning

confidence: 99%

Evaluator: A Multilevel Decision Approach for Web-Based Landmark Evaluation

Yin¹,

Yang²,

Liu³

et al. 2020

Security and Communication Networks

View full text Add to dashboard Cite

Street-level landmarks are an important basis for street-level IP geolocation, and the web-based landmark is one of the main sources of street-level landmarks. Considering the existing street-level landmark evaluation methods having low accuracy and strict constraints, this paper analyses the causes and evaluation idea of invalid web-based candidate landmarks and proposes Evaluator, a web-based landmark evaluation approach. Evaluator adopts the idea of the decision tree to filter invalid landmarks layer by layer and comprehensively estimates the quantitative reliability of candidate landmarks with public data and services to obtain reliable landmarks. This paper proposes the domain name system (DNS) distributed query algorithm to effectively resolve all IP addresses of a domain name, which provides data support for Evaluator to filter candidate landmarks. Meanwhile, this paper also proposes a reverse verification algorithm to obtain all domain names of an IP address, which provides an important reference to calculate the reliability of a reliable landmark. In addition, gradient descent is used to assess the parameters of the reliability estimating model, which effectively improves the robustness of Evaluator. Experiments show that reliable landmarks from Evaluator reduce the geolocation error of 100 targets in Hong Kong from 7.30 km to 3.91 km, compared with the landmark verifying method (LVM), one of the latest web-based landmark evaluation methods. Moreover, Evaluator significantly improves the evaluation coverage based on the same geolocation accuracy with street-level landmark evaluation (SLE), one of the latest landmark evaluation methods.

show abstract

“…They correct candidate landmarks with incorrect city-level locations and greatly improve the accuracy of landmarks. Wang et al [9] established an evaluation model based on routing strategy and machine learning to evaluate the city-level location of the landmarks extracted from the IP location databases, and significantly improved the accuracy of the landmarks. Shavitt and Zilberman [10] proposed a landmark evaluation method based on POP-level network analysis, which votes to determine the city-level location of the landmarks in a POP-level network, and greatly improves the accuracy of the landmark location.…”

Section: Introductionmentioning

confidence: 99%

Street-Level Landmark Evaluation With Upper Error Bound

2019

View full text Add to dashboard Cite

Street-level landmarks are the basis of high-precision IP geolocation, and the location-error of landmark affects the accuracy of geolocation result. However, existing landmark evaluation methods cannot determine the error range of landmarks. Therefore, a street-level landmark evaluation algorithm that can estimate the upper bound of landmark error is proposed. Firstly the city of candidate landmarks are verified by IP location databases. Secondly, candidate landmarks are grouped through their last-hop routers, and then divided into several clusters by E-Apriori algorithm based on their location. Thirdly, the Land-mark reliability probability model is used to calculate the probability of the last-hop router location range. Finally the upper error bound of the landmark is determined by the position of the last-hop router. By verifying 503 reliable land-marks evaluated by the algorithm in Hong Kong, we find that the algorithm can determine the upper error bound of landmarks, and the accuracy reaches 100%. We test the algorithm based on 100 accurate landmarks and 400 unreliable landmarks, and find that our algorithm can evaluate 84 accurate landmarks and 1 invalid landmark, i.e., it achieves an accuracy of 98.8%. Finally, we take landmark evaluation experiment based on 50,000 candidate landmarks in Hong Kong and Zhengzhou respectively. The results show that geolocation errors decrease obviously using our reliable landmarks, and the mean error of 100 targets in Hong Kong is reduced from 4.18 km to 2.78 km.

show abstract

An Optimization Method for the Geolocation Databases of Internet Hosts Based on Machine Learning

Cited by 5 publications

References 22 publications

ImMuNE: Improved Multilateration in Noisy Environments

ImMuNE: Improved Multilateration in Noisy Environments

Evaluator: A Multilevel Decision Approach for Web-Based Landmark Evaluation

Street-Level Landmark Evaluation With Upper Error Bound

Contact Info

Product

Resources

About