Answers are grounded in this paper and cite their sources.
Energy-efficient (EE) optimization schemes have been widely used in massive Internet of Things (IoT) deployments to extend device lifetimes under strict quality-of-service (QoS)-assured requirements. EE-aware resource allocation over low-power connecting networks has been challenging beyond conventional requirements. Increasing energy demand, which remains constant during the paused allocation interval, worsens the performance of these networks. In this regard, the proposed work develops a Semi-Bayesian Learning Optimization (SBLO) framework to address the trade-off between energy demand and EE. The low-power connecting devices are evaluated for energy efficiency, with the required consumption and demand determined based on previous sessions. The key element used to distinguish data points is the residual energy taken as input when predicting for adjacent state transitions. The QoS is mostly long-lasting in the IoT area, as forecasted for admissible requests, quantified by connectivity and repair time. The states before and after allocation are EE, demanding from the state before allocation to the state after allocation. Then, a probabilistic demand model defines when to consider the next pause interval, replacement, or recharging. Intermediate-state learning is used to project new EE or demand for connecting requests in the next request. We evaluate the proposed optimization for energy utilization, energy demand, connectivity and service time while comparing to representative baselines through experiments.
No takes yet. Share an insight, caveat, or question.
Alghuried et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: