S-ID: Implementing Intelligent Interoperability for Heterogeneous IoT Identifiers Through a Semantic-Based Framework

In the Internet of Things (IoT), each object (i.e., an IoT instance) requires a unique identifier to serve as its digital fingerprint, enabling users to access detailed information about the object. However, the diversity of IoT identifier frameworks leads to a heterogeneous set of identifiers. Moreover, these identifiers, often represented as character sequences, lack embedded semantic information. This absence of semantic information hampers the ability of objects with different identifiers to autonomously recognize and collaborate with each other. To address this challenge, we introduce a Semantic Identifier (S-ID) framework designed to bridge the gap between heterogeneous IoT identifiers. The S-ID provides each legacy identifier with a semantic prefix and maps the object to a common IoT ontology. To ensure compatibility with legacy identifiers across multiple frameworks, we develop an algorithm called RHINo (Recognize the frameworks of Heterogeneous Identification Numbers). RHINo employs three novel Identifier-to-Polygon (I2P) transformation strategies and embeds framework-specific information into the S-ID. Based on RHINo, we present a working mechanism to autonomously generate S-IDs and to facilitate intelligent interactions among IoT entities. Our evaluation, conducted on two custom-built datasets, demonstrates that RHINo outperforms existing state-of-the-art methods in classification accuracy. A real-world self-service store use-case further substantiates the efficacy of the S-ID as a working mechanism to ensure interoperability among heterogeneous IoT identifiers.