Conference:
EuroCyberSec 2024, 23. October 2024, Krakow, Poland
Authors:
Nasereddin M, Nakip M, Gelenbe E.
Abstract:
Internet of Things (IoT) networks are highly vulnerable to network attacks, the most common examples being DoS and DDoS attacks. Those attacks flood the limited system resources of IoT devices and overwhelm networks with large numbers of attack packets, causing severe performance impacts. In order to mitigate those attacks in IoT networks, this paper develops a lightweight yet effective Intrusion Detection and Prevention System (IDPS). This IDPS sequentially detects and mitigates the attack via a Deep Random Neural Network (DRNN) and the proposed Drop-Idle-Repeat (DIR) process within milliseconds after the attack begins. The developed IDPS is evaluated for DoS, in particular UDP Flood attacks of different lengths on a real-life experimental test-bed built within this paper. Experimental results first revealed that a 60-second long UDP Flood attack is sufficient to paralyse the Server used in the testbed. On the other hand, the proposed IDPS can successfully mitigate this attack; therefore, the Server utilizing the IDPS can continue its routine operation without being affected by the attack and resume communication with other devices immediately after the attack ends.
