Edge computing-based semantic modeling and identifier analysis for intelligent IoT

Le résumé fourni par la source

This study addresses the semantic ambiguity and low semantic tag caching efficiency caused by heterogeneous Internet of Things devices in smart cities, power grids, and industries. It proposes a semantic modeling framework based on network ontology language and combines edge computing and particle swarm optimization algorithms to optimize semantic tag caching. Furthermore, a multi-objective Harris Eagle optimization algorithm is introduced at the tag parsing layer for efficient parsing. Test results show that the model achieves an average end-to-end latency of 18.3 ms and a word inference energy consumption of 21.7 mJ on the T-Drive-Shenzhen dataset. The semantic tag buffering efficiency reaches 90.2%, and the parsing accuracy reaches 98.6%. In real-world scenario tests, the parsing accuracy remains above 95% under steady-state, peak load, and fault disturbance conditions. These results demonstrate that the proposed smart Internet of Things semantic modeling method can improve semantic tag caching efficiency and semantic parsing accuracy, while reducing end-to-end device latency, providing an efficient and low-energy-consumption solution for Internet of Things semantic interaction.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.