Desktop and mobile operating system fingerprinting based on IPv6 protocol using machine learning algorithms

Abstract

Operating system (OS) fingerprinting tools are essential to network security because of their relationship to vulnerability scanning and penetrating testing. Although OS identification is traditionally performed by passive or active tools, more contributions have focused on IPv4 than IPv6. This paper proposes a new methodology based on machine learning algorithms to build classification models to identify IPv6 OS fingerprinting using a newly created dataset. Unlike other proposals that mainly depend on TCP and IP generic features; this work adds other features to improve the detection accuracy. It also considers OSes installed in mobiles (Android and iOS). The experimental results have shown that the algorithms achieved high and acceptable results in classifying OSes. KNN and DT achieved high accuracy of up to 99%. SVM and GNB achieved 81% and 75%, respectively. Moreover, KNN, RF and DT achieved the best recall, precision, and f-score with almost the same as the achieved accuracy.