Decentralised Global Service Discovery for the Internet of Things

Area of Science:

• Computer Science
• Network Engineering
• Distributed Systems

Background:

• The Internet of Things (IoT) involves millions of interconnected devices across diverse networks, presenting challenges for global service discovery and data utilization.
• Effective service discovery is crucial for unlocking the full potential of IoT, enabling data sharing and interaction beyond initial deployment contexts.

Purpose of the Study:

• To introduce Decentralised Service Registries (DSRs), a novel, trustworthy, and decentralized system for global IoT service discovery and interaction.
• To enable users to create and share public/private service registries, register/query services, and access historical data, addressing limitations of existing discovery methods.

Main Methods:

• DSRs utilize signed objects, cryptographically linked into a signature chain, for service registration and discovery.
• A novel verifiable Distributed Hash Table (DHT) overlay is employed for efficient storage and querying of registry information.
• The system decouples registries from supporting infrastructure, ensuring privacy, multi-tenancy, and data/service verification.

Main Results:

• DSRs mitigate risks of service impersonation and data alteration through verifiable registry entries and historical data.
• A functional DSR was implemented and tested for registering and searching real-world IoT devices and their data.
• Performance was evaluated using a scalable cluster-based testbench for high-fidelity emulation of peer-to-peer applications.

Conclusions:

• DSRs present a novel and useful solution to key IoT challenges, particularly in global service sharing and discovery.
• The decentralized approach enhances trust, security, and accessibility for IoT services and data compared to centralized or other decentralized methods.