面向异质感知网络的水声MAC层信任评估与协同防御机制OA

Terminal nodes in dynamic heterogeneous sensing networks face computation,energy,and bandwidth con-straints,making heavyweight cryptography impractical;unreliable links may also blur channel anomalies and malicious behavior,weakening medium access control(MAC)-layer security.Thus,a feedback-driven MAC-layer trust evaluation and cooperative defense mechanism was proposed.Acknowledgment/negative acknowledgment(ACK/NACK)feedback served as a zero-cost trust primitive,and Dempster-Shafer(D-S)evidence theory fused communication status,environ-mental disturbance,and residual energy,with uncertainty modeling reducing channel-induced errors.Based on a suspi-cious node's receiver or sender role,the mechanism triggers vector-based forwarding rerouting or spare-channel isola-tion,forming a trust evaluation-role identification-defense loop.The physical-layer-independent framework was instanti-ated in underwater acoustic sensor networks using Bellhop acoustic-field modeling and α-stable noise.Simulations show superior detection accuracy,packet delivery,and throughput over representative baselines.