A parameter centric service discovery framework for social digital twins in smart City

In the contemporary digital era, the Internet of Things (IoT) and its applications have proliferated extensively, particularly within smart city environments, resulting in increased network traffic and raising the significance of efficient service discovery (SD) mechanisms. The social Internet of Things (SIoT) is an emerging paradigm that enables IoT devices to autonomously establish social relationships based on rules defined by their owners, thereby enhancing services through social relations. Things can interact with others; thus, the huge volume of traffic is increased. Each node or device could select an appropriate peer for the discovery of services, which is thus helpful for human beings. Although numerous service discovery and query processing models have been proposed in the recent literature. However, the existing state-of-the-art approaches often lack a comprehensive analysis of the parameters. Most traditional state-of-the-art models primarily focus on relationships or device similarity. Also neglecting the vital factors, for instance, query processing, efficiency, spatial-temporal dynamics, and service provisioning, etc. Thus, to solve this issue, this research proposes an exhaustive analysis of the main parameters needed to implement service discovery mechanisms for Social IoT and studies their relative importance based on a dataset of real objects. Based on the advanced parameters' selection, an efficient service discovery algorithm is proposed. The proposed model emphasizes efficiency by optimizing the service discovery through reduced social graph traversal (i.e., fewer hops), consideration of the service types, and integration of caching mechanisms. We have conducted a comprehensive analysis of key parameters essential for implementing an effective service discovery mechanism in SIoT, prioritizing based on their importance. Experimental validation demonstrates the superiority of the proposed over state-of-the-art models, confirming its efficacy, scalability.