A recommendation management defense mechanism based on trust model in underwater acoustic sensor networks

• Published in: Future generation computer systems Vol. 145; pp. 466 - 477
• Main Authors: Zhang, Mengjie, Feng, Renhai, Zhang, Hehe, Su, Yishan
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.08.2023
• Subjects: Collaborative filtering; Fuzzy algorithm; Recommendation management; Trust management mechanism; Underwater acoustic sensor networks (UASNs); Trust management mechanism; Collaborative filtering; Fuzzy algorithm; Recommendation management; Underwater acoustic sensor networks (UASNs)
• ISSN: 0167-739X; 1872-7115
• DOI: 10.1016/j.future.2023.03.043

Underwater acoustic sensor networks (UASNs) have been applied in many civilian and military scenarios, but it is vulnerable to various security threats due to the broadcast transmission characteristics and the environment in which they are located. Trust management mechanism has been proven to be an effect way to improve network security. However, in the trust assessment process, some nodes may provide false suggestions, which will lead to trust conflicts and affect the normal operation of the trust mechanism. To screen out unreliable recommendations and dishonest nodes in the network and avoid potential dangers, a recommendation management trust mechanism based on collaborative filtering and variable weight fuzzy algorithm (CFFTM) is proposed in this paper. First, three kinds of evidence of trust: communication-based evidence, data-based evidence, and energy-based evidence are select as indicators. Then the variable weight fuzzy comprehensive evaluation algorithm is applied to calculate the direct trust value of the node. Secondly, in order to quantify the honesty ability of nodes, honesty degree is defined. Then the overall recommendation trust value of the node is obtained using the proposed collaborative filtering algorithm. The simulation results show that the mechanism can well filter out unreliable recommendations and improve the recognition rate and stability of the trust model under typical attack scenarios. •Application of malicious node detection based on filter trust mechanism in underwater acoustic sensor network.•Quantifying the ability of a node to be honest when acting as a third-party node.•Under the different number of malicious nodes and attacks, using the proposed trust model.•Evaluate and validate the effectiveness of the proposed mechanism.