An Enhanced Dissection of Attacks in Wireless Sensor Networks

Abstract

Developing reliable solutions against attacks in wireless sensor networks (WSNs) requires dissecting the attacks to understand activities and strategies exploited by the attackers. Authors who proposed this comprehension consider active attacks and attacks occurring on specific layers but lacks to investigate relevant aspects such as the physical and logical components involved in the attack, routing protocols exploited and the position of attacker. We propose in this paper, a more complete Unified Modelling Language (UML) characterization of attacks which represent static and dynamic aspects of the attacker activities. Sixteen popular attacks have been studied and classified based on similarities and differences. For each attack, it is able to identify data that are exchanged, layers which are traversed, sequence of activities involved and components which are exploited within each attack. As a road-map to design countermeasures, resemblances and divergences are identified and discussed. A theoretical comparison with similar works has demonstrated that this study is effectively complementary.