Taxonomy and Holistic Detection of Security Attacks in MANETs

Pedro García-Teodoro; Leovigildo Sánchez Casado; Gabriel Maciá-Fernández

Abstract:

This paper presents a study about security attacks in MANETs. Starting from the many contributions in this field, we attempt to provide a novel taxonomy using practical relationships among the attacks reported in the literature. On the basis of the cited taxonomy, we subsequently discuss and propose a holistic detection approach for detecting malicious behaviors in such environments. We argue the convenience of moving from systems with particular detection capabilities for specific services and/or protocols to such comprehensive schemes, in which it is possible to re-use the particular capacities exhibited to handle more general types of events and, thus, allow more robust and flexible detection solutions. As a case study, the experimental results obtained by a holistic detector for packet dropping attacks show the good detection capabilities of our approach.

Research areas: