Reduction of error rate in Sybil attack detection for MANET

Mobile ad hoc networks (MANETs) require a unique, distinct, and persistent identity per node in order for their security protocols to be viable, Sybil attacks pose a serious threat to such networks. Fully self-organized MANETs represent complex distributed systems that may also be part of a huge complex system, such as a complex system-of-systems used for crisis management operations. Due to the complex nature of MANETs and its resource constraint nodes, there has always been a need to develop security solutions. A Sybil attacker can either create more than one identity on a single physical device in order to launch a coordinated attack on the network or can switch identities in order to weaken the detection process, thereby promoting lack of accountability in the network. In this research, we propose a scheme to detect the new identities of Sybil attackers without using centralized trusted third party or any extra hardware, such as directional antennae or a geographical positioning system. Through the help of extensive simulations, we are able to demonstrate that our proposed scheme detects Sybil identities with 95% accuracy (true positive) and about 5% error rate (false positive) even in the presence of mobility.