Automatic Verification and Execution of Cyber Attack on IoT Devices

Area of Science:

• Cybersecurity
• Computer Networks
• Internet of Things (IoT)

Background:

• Internet of Things (IoT) devices are increasingly prevalent in daily life and critical infrastructure.
• The widespread use of IoT devices makes them attractive targets for cyberattacks, as exemplified by the Mirai botnet.
• Existing security concerns are amplified by the infeasibility of manual security testing for billions of IoT devices.

Purpose of the Study:

• To investigate the application of automated attack execution models for autonomous penetration testing of IoT devices.
• To determine if common and severe Wi-Fi vulnerabilities persist in current IoT devices.
• To develop and validate a system for autonomous IoT security assessment.

Main Methods:

• Adapted automated attack execution models from cyber range research for IoT penetration testing.
• Designed and implemented autonomous penetration testing agents.
• Conducted a case study to evaluate the system's effectiveness on deployed IoT products.

Main Results:

• Demonstrated that automated models can effectively model and design autonomous penetration testing agents for IoT devices.
• Confirmed the presence of well-known and severe Wi-Fi vulnerabilities in currently sold IoT products.
• Showcased the capability of the developed system to autonomously detect these vulnerabilities.

Conclusions:

• Autonomous penetration testing is a feasible and effective approach for assessing IoT device security at scale.
• Significant security vulnerabilities, particularly related to Wi-Fi, remain prevalent in commercial IoT devices.
• The developed system provides a valuable tool for identifying and mitigating security risks in the rapidly expanding IoT ecosystem.