A novel multi-stage attack dataset for smart home intrusion detection.

Area of Science:

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

Background:

• Smart homes utilize interconnected devices, creating vulnerabilities due to weak cybersecurity protocols.
• Evolving cyber-attack patterns, particularly multi-stage attacks, pose significant detection and mitigation challenges.

Purpose of the Study:

• To present a novel dataset of smart home network traffic, including normal operation and seven distinct multi-stage attack scenarios.
• To offer researchers insights into network parameter behavior during complex, sequential cyber-attacks within heterogeneous smart home environments.

Main Methods:

• Dataset creation involved capturing network traffic under normal conditions and simulating seven multi-stage attack scenarios.
• The dataset comprises 178,831 samples, divided into training (148,959) and testing (29,872) sets.
• Common network flow parameters were collected to reflect real-world smart home network behavior.

Main Results:

• The dataset provides a unique resource for studying interleaved attack patterns involving multiple sequential attack steps.
• It enables the development and evaluation of intrusion detection systems specifically for smart home environments.
• The captured network parameters allow for potential extension of research to other network types.

Conclusions:

• The novel dataset is crucial for advancing cybersecurity research in smart home environments.
• It facilitates the creation of more robust intrusion detection systems capable of handling sophisticated, multi-stage cyber threats.
• This resource supports the development of enhanced network security strategies for the growing number of connected devices in homes.