Node energy consumption minimization strategy in wireless sensor networks based on CICRL
TL;DR
A collaborative information coverage reinforcement learning algorithm that enhances state representation with multi-source coverage and neighborhood energy interaction, and optimizes policies via an energy consumption differential update mechanism improves policy convergence and energy balancing in high-dimensional scenarios.
Abstract
To address challenges such as energy consumption control, coverage stability, and complex interference in wireless sensor networks, this study proposes a collaborative information coverage reinforcement learning algorithm. It enhances state representation with multi-source coverage and neighborhood energy interaction, and optimizes policies via an energy consumption differential update mechanism. The method improves policy convergence and energy balancing in high-dimensional scenarios. Tests on public datasets achieved up to 89.7% coverage, outperforming benchmarks by 5%–12%. Ablation studies confirmed the contributions of collaborative information and energy differential terms, boosting coverage by 8%–15%. Under various interference levels, the energy per packet remained stable (1.18–1.89 J/packet), lower than that of comparison methods. Network lifetime increased by about 20% in continuous and discrete transmission scenarios, demonstrating advantages in coverage, energy efficiency, and robustness for large-scale sensor and IoT applications.