Edge computing has been viewed as a powerful technology to realize the vision of network services. However, due to the limited capabilities and insufficient security systems, edge computing is vulnerable to distributed denial of service (DDoS) attacks which may exhaust the resources of edge servers with excessive requests and degrade their service capabilities. Setting detection thresholds for DDoS detection indicators can effectively prevent DDoS attacks, but existing thresholding methods fail to update detection thresholds in time to guarantee the detection performance whenever the system settings vary. In this paper, we propose a multi-dimensional thresholding method against DDoS attacks in edge computing. We design three detection indicators based on the behavior features of DDoS attackers. By solving a threshold optimization problem, we obtain closed-form solutions and numerical solutions of the optimal detection thresholds, which adapt to dynamic system settings. Simulations show that the proposed thresholding method has a superior detection performance in terms of both the accuracy and robustness.
Multi-Dimensional Security Indicator Design and Optimization for DDoS Detection in Edge Computing
10.10.2023
1071031 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
A Multi-layer Security approach for DDoS detection in Internet of Things
Emerald Group Publishing | 2020
|