.png){kind=icon}
.png){kind=icon}
Return
Malicious Node Identification Strategy With Environmental Parameters
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2020.3013840.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Wireless sensor network (WSN) works in a complex environment where it is difficult for people to reach or work. The openness of nodes leads to security threats vulnerable to various attacks. The trust and reputation model can be applied in WSN to reduce damage caused by malicious nodes. However, there is a high false-positive rate in trust and reputation models because a node with less reputation due to the communication environment is judged as a malicious one directly. This paper presents a trust & reputation-based malicious node identification strategy with environmental parameters (TRS&EP) to interdict the malicious nodes, such as interrupt attack nodes and selective forwarding attack nodes. Using the linear regression of machine learning and combining the energy of nodes, data volume, number of adjacent nodes, the node sparsity and other deterministic parameters can solve environmental parameters. Then TRS&EP estimates benchmark trust according to the environmental parameters. The Gaussian radial basis function is simplified to calculate the similarity between the benchmark trust sequence and cycle reputation sequence. Furthermore, TRS&EP sets three reputation intervals and an adoptive threshold span to identify the malicious nodes by dynamically considering the work environment and states of nodes. The simulated results show that TRS&EP improves the recognition of malicious nodes above 1% compared to comparison algorithms and reduces the false-positive percentage by more than 1%.
Keywords:
Wireless sensor networks
Security
Energy consumption
Bayes methods
Biological system modeling
Maintenance engineering
Machine learning
Wireless sensor network
interrupt attacks
selective forwarding attacks
reputation model
environmental parameter
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W