Analyzing anonymous activities using Interrupt-aware Anonymous User-System Detection Method (IAU-S-DM) in IoT

Area of Science:

• Cybersecurity
• Network Security
• Internet of Things (IoT)

Background:

• Unauthorized access to Internet of Things (IoT) networks poses significant security risks.
• Existing intrusion detection systems often suffer from high computation times, hindering accurate and timely intruder identification.
• Intermediary access during IoT data transmission introduces vulnerabilities.

Purpose of the Study:

• To develop an efficient intrusion detection method for IoT networks.
• To address the challenge of high computation time in existing systems.
• To improve the accuracy and speed of identifying unauthorized access in IoT environments.

Main Methods:

• Implementation of the Interrupt-aware Anonymous User-System Detection Method (IAU-S-DM).
• Utilizing concealed service sessions to detect anonymous interrupts.
• Training the system with parameters including origin, session access demands, and user legitimacy.
• Employing a deep recurrent learning approach for data processing to identify service failures and breaches.

Main Results:

• The IAU-S-DM method demonstrated a service failure rate of 10.65%.
• Achieved a detection precision of 14.63% and a detection time improvement of 15.54%.
• The classification ratio reached 20.51%, indicating effective intruder activity recognition.

Conclusions:

• The IAU-S-DM method offers a computationally efficient solution for detecting intrusions in IoT networks.
• Deep recurrent learning enhances the identification of service failures and breaches, improving detection rates.
• The method effectively utilizes the TON-IoT dataset for identifying intruder activities and validating system consistency.