A conflicts' classification for IoT-based services: a comparative survey

Area of Science:

• Computer Science
• Software Engineering
• Cybersecurity

Background:

• Internet of Things (IoT) systems are integral to modern life, enabling intelligent environments and device control.
• End-user developed IoT apps interact, leading to critical safety, security, and correctness issues.
• Existing solutions lack comprehensive error and conflict detection for complex IoT app interactions.

Purpose of the Study:

• To classify and detect conflicts in Internet of Things (IoT) systems.
• To investigate and analyze existing research techniques for IoT conflict classification and detection (2014-2020).
• To propose a novel classification of IoT app interaction conflicts and evaluate current detection methods against it.

Main Methods:

• Literature review of IoT conflict classification and detection approaches published between 2014 and 2020.
• Development of a new classification for IoT-based application interaction conflicts.
• Utilizing Satisfiability Modulo Theories (SMT) model checking and formal notations for a priori conflict detection.
• Comparative analysis of existing detection approaches against the proposed classification.

Main Results:

• A new classification for IoT app interaction conflicts is proposed.
• Existing detection approaches exhibit gaps in covering various conflict types and levels.
• The study identifies limitations in current methods, impacting IoT system safety and correctness.

Conclusions:

• Current conflict detection methods for IoT systems are inadequate, posing risks to safety and security.
• There is a critical need for specialized safety and security compilers or tools for IoT applications.
• Future IoT verification frameworks should adopt hybrid approaches combining model checking with semantic technologies for comprehensive conflict coverage.