基于自主水下机器人的水下恶意节点检测算法OA

In order to address the issue of malicious nodes transmitting false data during underwater localization,which compromises the positioning accuracy of underwater wireless sensor network(UWSN)nodes,a malicious node detection algorithm(AUV-MNDA)is proposed based on autonomous underwater vehicle(AUV).The algorithm collects information such as the number of communication successes,false coordinates,and signal reception indication strength through the information interaction between the AUV and the nodes in the network,establishes three kinds of credibility indexes,namely,communication success rate,communication data authenticity and degree of communication non-delay,and then combines the information entropy of the credibility indexes to obtain a collection of credibility evaluation.Furthermore,considering the potential for malicious defamation by the AUV,the algorithm introduces a recursive optimization detection mechanism that weights the progressively refined node reputation evaluations,effectively detecting malicious evaluations and thereby enhancing positioning accuracy.Simulation results indicate that when the proportion of malicious nodes is below 50%,the AUV-MNDA algorithm can maintain a detection accuracy exceeding 90%.In addition,compared to the MNDA,LABTM,and TMIS algorithms,the AUV-MNDA algorithm decreases the underwater node localization error by 6%,8%,and 16%,respectively.