Automated IoT Device Identification Based on Full Packet Information Using Real-Time Network Traffic.

Area of Science:

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

Background:

• Internet of Things (IoT) environments generate vast amounts of sensitive data, necessitating robust security measures.
• Traditional identification metrics like IP and MAC addresses are inadequate for securing IoT devices due to their volatility and ease of compromise.
• Malicious activities, including data spoofing, pose significant risks to critical IoT applications.

Purpose of the Study:

• To develop and evaluate an enhanced device identification framework for Internet of Things (IoT) systems.
• To improve the security and authenticity of data originating from IoT devices.
• To address the limitations of conventional IoT device identification methods.

Main Methods:

• A classification-based device identification framework was proposed, integrating sensor measurements, statistical features, and header information.
• Various machine learning algorithms were employed to analyze different combinations of these feature sets.
• Real-time data from IoT devices was collected under both normal and attack conditions for evaluation.

Main Results:

• The proposed framework demonstrated effectiveness in identifying IoT devices using a combination of sensor and header features.
• Machine learning algorithms showed varying degrees of success in identifying devices based on feature set combinations.
• The system proved robust in distinguishing legitimate devices from malicious activities in a controlled lab environment.

Conclusions:

• Combining sensor measurements, statistical features, and packet header information offers a more secure approach to IoT device identification.
• Machine learning-based classification provides a flexible and powerful tool for enhancing IoT security.
• The proposed method contributes to building more resilient and trustworthy Internet of Things ecosystems.