Optimized clustering-based discovery framework on Internet of Things

With the proliferation of technology, a system of connected and interconnected devices, henceforth referred to as Internet of Things, is emerging as a viable method for automated interactions between users and environment in day-to-day life. However, such proliferation leads to an impractical task with respect to interactions among humans and devices. The major reason behind this impractical task is that domain of human’s eye for interaction is limited and devices have their own obligations and prohibitions in context. Motivated by this observation, the paper has proposed four-layered framework, namely, Optimized Clustering-based Discovery Framework on Internet of Things (OCDF-IoT), that (1) automatically discovers resources and their associated services using ontology, (2) governs resources using knowledge formation and representation, (3) provides efficient procedures to index resources on the basis of maximum similarity match, and (4) delegates the selection of the near optimal resource among indexed resources. The framework’s efficiency is evaluated using toll datasets that are gathered from Shambhu Toll Plaza, Panipat–Jalandhar section, Haryana, India. The obtained results support the framework’s efficacy providing more accurate similarity searches, consuming less search time. It is found that framework is stable in providing accurate erred parametric resources and helps in finding the rightful resource with computation of maximum resources. The framework takes minimum CPU throughput for processing queries and increases CPU’s efficiency with less load on server.