Abstract
With wide applications like ocean surveillance, Underwater Acoustic Mobile Adhoc NETwork (UAMANET) becomes a double-edged sword for military and industry to secure underwater operations. UAMANET inherits vulnerabilities from 802.11-based MANET, which renders traditional cryptographic approaches defenseless. Trust Management Framework (TMF), maintaining confidence among participating nodes in MANET with metrics built from the communication activities, promises secure, efficient, and reliable access to terrestrial MANET. TMF cannot be directly applied to underwater environments due to marine characteristics that make it difficult to differentiate natural turbulence from intentional misbehavior. A machine learning method promises security in UAMANET with multi-domain trust, i.e., Multi-parameter Trust Framework for MANET (MTFM), by merging physical metrics - assessing motion of a cohort - into communication metrics. However, uncertainty of trust in harsh underwater environments affects the accuracy of trust evaluation. A novel Trust Model based on Cloud theory (TMC) solves the problem of trust uncertainty for UAMANET. This work complements MTFM with TMC to improve access control for UAMANET. By integrating the trust framework of communication and physical domains with the cloud model to combine two kinds of uncertainties: fuzziness and randomness, trust management is greatly improved for UAMANET. The paper proposes a trust model to defend UAMANET against the attacks with TMC-enhanced MTFM, i.e., Trust Management Framekwork with Multi-domain metrics and Cloud (TMFMC). The model takes a six-metric vector, the union of communication and physical characteristics, while calculate trust values from three levels: a node's own observations as direct trust, its one-hop relay as recommenders, and its multi-hop relay as indirect trust.