A Malicious Node Identification Strategy with Environmental Parameters Optimization in Wireless Sensor Network
- PDF / 2,119,398 Bytes
- 20 Pages / 439.37 x 666.142 pts Page_size
- 60 Downloads / 163 Views
A Malicious Node Identification Strategy with Environmental Parameters Optimization in Wireless Sensor Network Zhijun Teng1,2 · Baohe Pang2 · Mingyang Sun3 · Luying Xie2 · Liwen Guo4 Accepted: 29 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In the wireless sensor network (WSN), nodes show a low forwarding rate under a poorquality links environment and a resource-constrained state. The malicious nodes imitate this forwarding behavior, which can selectively forward date, eavesdropping, or discarding important dates. The traditional reputation model is challenging to identify with this kind of sub-attack nodes. To address these problems, a malicious node identification strategy based on time reputation model and environmental parameters optimization (TRM-EPO) is proposed in the WSN. First of all, the comprehensive reputation is calculated according to the direct reputation and the recommended indirect reputation. The environmental parameters matrix is based on nodes’ running state, taking into account nodes’ energy, data volume, number of adjacent nodes, and node sparsity. Besides, according to the environmental parameters matrix, and the recorded comprehensive reputation matrix, the next cycle’s trust can be predicted. Finally, a similarity of the actual reputation and predicted trust matrix is proposed to compare with an adaptive threshold to identify malicious nodes. The experimental results demonstrate that the proposed strategy improves sensor nodes’ security and reliability in a complex environment. Moreover, compared to comparison algorithms, the TRM-EPO improves the recognition rate above 1% and reduces the falsepositive rate by more than 1%. Keywords Wireless sensor network · Interrupt attacks · Selective forwarding attacks · Reputation model · Environmental parameters
1 Introduction Sensor nodes are used in many applications, such as military, industrial, medical, commercial, and other fields. With the continuous development of wireless sensor networks (WSN), users have higher requirements for the network’s security performance [1, 2]. However, the sensors’ deployment environment is complex and uncontrollable. Nodes are vulnerable to physical attacks and eavesdropping attacks by the enemy. Due to its openness and information transmission through multicast, broadcast, and other * Baohe Pang [email protected] Extended author information available on the last page of the article
13
Vol.:(0123456789)
Z. Teng et al.
communication methods, WSN is vulnerable to replay attacks, denial of service attacks, blocking attacks, and other forms of damage [3]. These external attacks will result in data transmission error, delay, or interruption. Sensor nodes captured by the attacker will be decrypted and tampered with the network’s program by the enemy. When these tampered nodes are put back into the network again, they become internal malicious nodes to break the WSN and launch almost any attack [4]. They will delete critical packets, interrupt or selectively forward
Data Loading...
