Autopolicy: Automated Traffic Policing for Improved IoT Network Security

Area of Science:

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

Background:

• Massive Distributed Denial of Service (DDoS) attacks, reaching record 2.3Tbps, increasingly leverage hijacked Internet of Things (IoT) devices.
• The widespread reliance on the internet makes these attacks a significant threat to global connectivity and daily life.
• Hijacked IoT devices are a primary source of these large-scale cyberattacks, necessitating novel mitigation strategies.

Purpose of the Study:

• To address the threat posed by hijacked IoT devices in Distributed Denial of Service (DDoS) attacks.
• To propose and validate a novel system, Autopolicy, for automatically mitigating DDoS attacks originating from IoT devices.
• To protect the internet and local networks by controlling resource allocation for IoT devices.

Main Methods:

• Autopolicy automatically limits IP traffic bandwidth and other network resources for IoT devices within a network.
• The system leverages the observation that most IoT devices do not require high-speed network interfaces for normal operation.
• A Software Defined Network (SDN) architecture is utilized, detailing functional blocks, message sequences, and operational flow.

Main Results:

• Experimental validation demonstrates the effectiveness of the Autopolicy system.
• The proposed system successfully mitigates threats originating from compromised IoT devices.
• A prototype open-source implementation of Autopolicy is released.

Conclusions:

• Autopolicy offers an effective, automated solution for mitigating DDoS attacks fueled by IoT devices.
• The system's resource limitation strategy is practical, exploiting the typical low bandwidth needs of IoT devices.
• The open-source release promotes further research and adoption in network security.