Adaptive epsilon greedy reinforcement learning method in securing IoT devices in edge computing

• Published in: Discover Internet of things Vol. 4; no. 1; pp. 27 - 23
• Main Authors: Kumar, Anit, Singh, Dhanpratap
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 20-11-2024; Springer Nature B.V; Springer
• Subjects: Access control; Adaptive; Algorithms; Artificial intelligence; Communication; Communications networks; Computer Science; Cyber-physical systems; Edge computing; Electrical Engineering; Exploitation; Floods; Information Systems Applications (incl.Internet); Internet of Things; IoT; IoT-Edge computing; Machine learning; Malware; Malware attacks; Network security; Sensors; Task scheduling; ℇ-greedy reinforcement learning; Network security; Task scheduling; Malware attacks; Adaptive; ℇ-greedy reinforcement learning; IoT-Edge computing
• ISSN: 2730-7239; 2730-7239
• DOI: 10.1007/s43926-024-00080-7

Attacks on IoT devices are increasing day by day. Since IoT devices nowadays have become an integral part of our daily lives, the data gathered from IoT devices benefits intruders in many ways. Financial and Healthcare institutions also allow their customers to use their services by using handheld IoT devices. Smart homes and autonomous vehicles use many IoT devices to gather data through the built-in sensors and send it to the Edge server for further processing. The computation result on the Edge server determines the decision to fulfill the user-specific needs. As these data are vital in the future cycle of execution of an intelligent algorithm of IoT device software program, hence the data are not just of temporary use, but it is transferred to a Cloud server for permanent storage. The data flows from IoT sensors to the Edge server, then from the Edge server to the Cloud server. Here the riskiest part for data to stay is on the Edge server. To counter such a security risk, we proposed and implemented the Adaptive Epsilon Greedy Reinforcement Learning (AEGRL) method which is the extension of the traditional Epsilon (ℇ) greedy reinforcement learning method. The proposed method works efficiently for both static and dynamic environments. Experimental results show that our proposed security method outperforms the recent similar security approaches in terms of scalability, robustness, and accuracy.
Article Highlights
The research work proposed in this paper is to prevent malicious attacks on the IoT edge server since the edge server continuously gathers data from the surrounding IoT devices. The following are the main highlights of the paper, which bring novelty to our research work.
We have considered the dynamic nature of traffic data concerning volume and the pattern of malicious data. The metadata of the malicious packets keeps changing by the smart hackers who also use intelligent tools to cross the common security barriers.
Considering the dynamic environment over the Edge server we extended the traditional reinforcement learning approach to include Adaptive ability on the dynamic constraints.
We experimented with both simulation-based and real-time environments.