Online Distributed IoT Security Monitoring with Multidimensional Streaming Big Data

Area of Science:

• Computer Science
• Cybersecurity
• Data Science

Background:

• Internet of Things (IoT) systems generate massive data streams, leading to 'big data' challenges like large volumes and scalability issues.
• Monitoring IoT systems and detecting sophisticated cyberattacks are significant challenges due to data complexity and system scale.

Purpose of the Study:

• To propose an online distributed IoT security monitoring algorithm (ODIS) for enhanced threat detection and system oversight.
• To address the scalability limitations inherent in traditional IoT monitoring approaches.

Main Methods:

• Utilized an advanced influential point selection operation to extract key information from multidimensional time-series data.
• Developed an accurate data structure model to represent IoT system behaviors and employed hypothesis testing for uncertainty quantification.
• Implemented a distributed system architecture to ensure scalability.

Main Results:

• The proposed ODIS algorithm demonstrated effective extraction of critical information from distributed sensor data.
• Hypothesis testing provided a quantifiable measure of uncertainty in monitoring tasks.
• The distributed architecture successfully addressed scalability concerns in the IoT environment.
• Experimental validation on a real sensor network testbed confirmed the algorithm's promising detection and monitoring capabilities against various cyberattacks.

Conclusions:

• The ODIS algorithm offers a robust and scalable solution for real-time security monitoring in IoT environments.
• The integration of spatial-temporal data dependence and hypothesis testing enhances the accuracy and reliability of cyberattack detection.
• This research contributes a novel approach to managing big data challenges in IoT security.