Detection of black hole nodes in mobile ad hoc network using hybrid trustworthiness and energy consumption techniques

• Main Author: Mustafa, Ahmed Sudad
• Format: Thesis
• Language: English
• Published: 2017
• Online Access: http://psasir.upm.edu.my/id/eprint/70994/1/FK%202017%2012%20-%20IR.pdf; http://psasir.upm.edu.my/id/eprint/70994/

Mobile ad-hoc network (MANET) is an evolving technology that is utilized in different applications (i.e. military surveillance, personal network, etc.) and developed in the recent years. Nodes in MANET are capable of functioning as a router for data communication. MANET devices do not require central management, capable of self-organizing/ healing through persistent reconfiguration. Any MANET node requires a protocol in order to communicate with its neighbor within its transmission range. Ad hoc on-demand distance vector routing protocol (AODV) is a commonly used protocol in MANET. AODV is a reactive protocol that offers relatively low routing overhead since the nodes utilizing this protocol operates only when a route is requested. However, AODV suffers severely from the black hole attacks where the attacker node advertise itself as having the optimum path leading to the destination node by varying some essential parameters. Therefore, detecting the black hole in the network is substantial since MANET depends on the cooperation between adjacent nodes. In this thesis, a hybrid detection algorithm mechanism has been proposed which combines two detection algorithms based on nodes’ trustworthiness and energy consumption in a parallel manner in order to detect the black hole nodes. An empirical testing approach was utilized here where several scenarios have been implemented and investigated in order to find the optimal settings. Network simulator (NS2) simulation findings demonstrate that the trust based algorithm achieves an average packet delivery ratio (PDR) of 87.3%, end to end delay (EED) of 7.47 ms and black hole detection accuracy of 90%. On the other hand, the detection algorithm based on the energy consumption achieves PDR of 91.6%, EED of 14.03 ms and detection rate accuracy of 93%. The hybrid technique offers decent average PDR of 94.7, EED of 8.62 ms and improved black hole detection rate accuracy of 96%. Furthermore, the hybrid technique offers reduced end to end delay with relatively high PDR when compared with two recent works.