Paper Info
Title
A Trust-Based Malicious Detection Scheme for Underwater Acoustic Sensor Networks
Abstract
Underwater acoustic sensor networks (UASNs) have been widely applied in the fields of maritime and underwater industries and national defense. However, due to the unattended deployment environment of UASNs, the sensor nodes are vulnerable to malicious attacks and are easily compromised to be malicious nodes. In recent years, trust models are proved as an effective and efficient tools for identifying malicious nodes possessing valid identity information. We propose a trust-based malicious identification scheme (TMIS) for UASNs. First of all, the impact of underwater environment on communication trust is quantified, which makes communication trust effectively reflect the behavior of the attacks that cause communication failure such as selective forwarding attacks. Second, communication traffic is exploited to effectively reflect the behavior of the attacks that transmit or receive an abnormal number of packets, such as DOS attack. Third, we train the prediction model with SVM and K-means++ algorithms. Finally, two trust update mechanisms are proposed to cope with the dynamic environment of UASNs and On-Off attacks. The simulation results show that TMIS can effectively identify malicious nodes in complex underwater environment compared to the other three kinds of identification schemes. In particular, the larger the rate of malicious nodes is, the better TMIS performs relatively.
Year
DOI
Venue
2022
10.1007/978-3-031-06791-4_34
Artificial Intelligence and Security
Keywords
DocType
ISSN
UASNs, Malicious node detection, Trust mode
Conference
0302-9743
Citations 
PageRank 
References 
0
0.34
0
Authors
5
Name
Order
Citations
PageRank
Liang Kun100.34
Sun Shijie200.34
Huang Xiangdang300.34
Yang Qiuling400.34
Naixue Xiong52413194.61