Abstract
Computer networks are usually randomly deployed in locations where there is no underlying network infrastructure. When a computer node is exposed to a malicious attack environment, the nodes are vulnerable to unknown attacks, resulting in misplaced errors. Traditional node detection methods are difficult to operate safely without human participation, and the detection speed is slow. In this paper, a method of boundary malicious node detection based on multivariate classification is proposed. First of all, the background and the objectives are described. Then, the detection method is studied. The experimental results show that this method can quickly identify unknown attack problems, and through the effective detection method to complete the detection of boundary malicious nodes, with fast speed and low false detection rate.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call