Year: 2025
Type: preprint
Abstract: —Efficient data transmission in resource-constrainedInternet of Things (IoT) systems requires semantics-aware man-agement that maximizes the delivery of timely and informativedata. This paper investigates the optimization of the semanticmetric Version Age of Information (VAoI) in a status updatesystem comprising an energy-harvesting (EH) sensor and a des-tination monitoring node. We consider three levels of knowledgeabout the system model—fully known, partially known, andunknown—and propose corresponding optimization strategies:model-based, estimation-based, and model-free methods. By em-ploying Markov Decision Process (MDP) and ReinforcementLearning (RL) frameworks, we analyze performance trade-offs under varying degrees of model information. Our findingsprovide guidance for designing efficient and adaptive semantics-aware policies in both known and unknown IoT environments.
Source: arXiv (Cornell University)
Authors: Delfani, Erfan, Pappas, Nikolaos
Institution: Linköping University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4414809506