Defense against packet collusion attacks in opportunistic networks. Issue 65 (March 2017)

• Record Type: Journal Article
• Authors: Alajeely, Majeed; Doss, Robin; Ahmad, Asma'a; Mak-Hau, Vicky
• Is Part Of: Computers & security. Issue 65(2017)
• Page Start: 269
• Page End: 282
• Publication Date: 2017-03
• Subjects: Opportunistic Networks -- Security -- Packet dropping attacks -- Denial-of-service -- Malicious node detection; Computer security -- Periodicals; Electronic data processing departments -- Security measures -- Periodicals; 005.805
• DOI: 10.1016/j.cose.2016.12.001
• Languages: English
• ISSNs: 0167-4048

Abstract

Abstract: Security is a major challenge in Opportunistic Networks (OppNets) because of its characteristics such as an open medium, dynamic topology, no centralized management and absent clear lines of defense. A packet dropping attack is one of the major security threats in OppNets since neither source nor destination nodes have control over the behaviour of intermediate nodes in the network. Consequently, the knowledge of where or when packets are/will be dropped is difficult to gather. In this paper, we present a novel attack and traceback mechanism against a special type of packet dropping attacks – packet collusion attacks, where the malicious node(s) drops some or all packets and then injects new fake packets in their place to mask the packet dropping. Our novel detection and traceback mechanism is based on the concept of a Merkle (or hash) tree and simulation results show it to be highly effective and accurate in terms of detecting attack instances and tracing back to the malicious node(s) in the network that is the attack source.