Trust Based Security Framework for Secure Routing in Mobile Ad hoc Networks

Abstract

Trust Based Security Framework for Secure Routing in Mobile Adhoc Networks Mobile Ad hoc Networks (MANETs) are infrastructure-less networks without central authority. Messages among distant nodes are exchange using multi-hop communication with an assumption that all nodes are trustworthy. However, aforementioned assumption is beyond reality in a network, such as MANETs. A relay node may misbehave by dropping or altering messages either to save resources or due to the malicious intention. To cater for the aforementioned security issues, range of trust-based security schemes has been proposed in the literature to detect and mitigate the adverse e ect of the misbehaving nodes in MANETs. Most of the existing trust-based security schemes in MANETs lack comprehensive and adaptive trust criteria. Proposed schemes are based on single trust attribute, such as packet loss. Each packet loss is considered as an indicator of possible attack by a malicious node. However, there can be several reasons of packet loss, such as interference, queue over ow, and node mobility. Moreover, single trust attribute may cause the bootstrapping problem, which refers to the time required for the evaluation of the trustworthiness of a node. Because of the bootstrapping problem, malicious nodes remain undetected for longer period of time, thus providing more opportunities to malicious nodes to carry out adverse activities. Furthermore, use of static trust parameters, such as trust threshold, weights assignment to rst-hand and second-hand reputation information, and trust update frequency in existing trust-based security schemes may result in high false positives rate, low malicious node detection rate, and network partitioning. To address the aforementioned challenges, we have presented an Adaptive Trustbased Security Framework (ATSF) that is based on the comprehensive trust criteria with adaptive trust parameters. The ATSF is based on the multi-attribute trust criteria with a mechanism to identify the real underlying causes of packet losses. Moreover, the ATSF adapts and update the trust parameters, such as xi trust threshold, weights assignment to rst-hand and second-hand reputation information, and trust update frequency, based on the run-time network conditions. We provide an implementation of ATSF on a standard routing protocol, the Optimized Link State Routing (OLSR) protocol, and evaluate it through extensive simulations with NS-2 under varying network conditions. We have evaluated the e ectiveness and performance of the ATSF in comparison to existing trust based security schemes under di erent network parameters and con gurations. The experimental results show that the ATSF achieves 98{100% detection rate of malicious nodes with only 1{2% false positives. Moreover, ATSF achieves 75{90% packet delivery ratio compared to 65{80% of static scheme with slight increase in end-to-end delay and energy dissipation of the nodes.