SMART: An SVM-based Misbehavior Detection and Trust Management Framework for Mobile Ad hoc Networks

, , and

Due to a lack of pre-deployed infrastructure, nodes in Mobile Ad hoc Networks (MANETs) are required to relay data packets for other nodes to enable multi-hop communication between nodes that are not in radio range with each other. However, whether for selfish or malicious purposes, a node may refuse to cooperate during the network operations or even attempt to interrupt them, both of which are recognized as misbehaviors. In this paper, we describe an SVM-based Misbehavior Detection and Trust Management framework (SMART) to address the security threats caused by various misbehaviors. In SMART, the Support Vector Machine algorithm is used to detect node misbehaviors, which does not require any pre-defined threshold to distinguish misbehaviors from normal behaviors. In addition, a multi-dimensional trust management scheme is applied to evaluate the trustworthiness of MANET nodes from multiple perspectives, allowing the trustworthiness of each node to be described in a more accurate and effective manner. To validate the SMART framework, an extensive performance study is conducted using simulation. The results show that the SMART framework outperforms previous schemes: It handles a larger fraction of adversaries, and it is resilient when nodes are highly mobile. More importantly, it can detect adversaries that alter their behaviors over time.

  • 897851 bytes

  • 3627908 bytes

misbehavior detection, mobile ad hoc networks., multi-dimensional trust management, security



Downloads: 77 downloads

UMBC ebiquity