Realizing IoT service's policy privacy over publish/subscribe-based middleware

Area of Science:

• Computer Science
• Information Security

Background:

• The publish/subscribe paradigm enhances IoT service scalability and flexibility by decoupling event producers and consumers.
• Existing middleware lacks robust mechanisms for preserving IoT service policy privacy, leaving publishers with limited control and subscribers unable to filter events.

Purpose of the Study:

• To present a novel access control framework designed to preserve IoT service policy privacy within the publish/subscribe communication infrastructure.
• To address the challenges of policy privacy in large-scale IoT systems.

Main Methods:

• A two-layer cooperating method is employed, with a data layer for protecting IoT events and an application layer for preserving service policy privacy.
• The anonymous-set-based principle is utilized for policy embedding, encoding, and matching.
• The framework was implemented by extending the open-source Apache ActiveMQ broker with a policy-based authorization mechanism.

Main Results:

• The proposed policy privacy framework is proven to be Chosen-Plaintext Attack secure.
• Performance evaluation demonstrates that the approach is scalable and introduces only reasonable overheads.

Conclusions:

• The developed framework effectively preserves IoT service policy privacy in publish/subscribe systems.
• The integration with Apache ActiveMQ provides a practical and efficient solution for enhancing IoT security.