Uncertainty analysis framework for trust based routing in MANET

Abstract

MANET (Mobile Ad hoc Network) has distinctive characteristics like node mobility, broadcast nature of wireless communication, and peer to peer data transfer. Due to these characteristics uncertainty is an integral part of MANET behavior. Uncertainty quantification is important for performance evaluation and better decision making in MANET. Packet dropping is a serious issue in MANET which degrades the performance of the network to a great extent. The packet dropping attack increases uncertainty in the network communication and eventually of the MANET. Trust based routing mechanisms are useful for neutralizing packet dropping attack. There are many existing trust based routing protocols which uses uncertainty for precise trust calculations. Thus, trust and uncertainty are closely related to each other. The paper proposes Uncertainty Analysis Framework (UAF) for MANET, it calculates the network Belief, Disbelief, and Uncertainty (BDU) values. The UAF framework integrated into different trust variants of AODV protocol which use direct trust, indirect trust, and global trust. The research work discovers impact of different trust models on MANET BDU using various mobility models. Experimental analysis shows that, trust based routing protocols show average 3 % gain in packet delivery ratio and at least 5 % increase in the network belief.