LEACH-RLC: Enhancing IoT Data Transmission with Optimized Clustering and Reinforcement Learning

TL;DR

LEACH-RLC tackles energy efficiency in IoT WSNs by coupling a MILP-based CH selection and node-to-cluster assignment with a reinforcement-learning controller that optimizes clustering timing to minimize control overhead. A neural surrogate accelerates MILP decisions, and a DQN-based RL agent governs when to re-cluster, using states that include $E_{net}$, $E_n$, CH sets, and clustering history. Empirical results show LEACH-RLC outperforms LEACH, LEACH-C, EE-LEACH, LEACH-D, and LEACH-CM in network lifetime, energy consumption, and overhead, while maintaining balanced CH distribution. The work demonstrates practical feasibility for real IoT deployments via offline RL training, centralized clustering with scalable overhead management, and integration with standard IoT MAC/PHY stacks, while outlining future work on scalability and more realistic network conditions.

Abstract

Wireless Sensor Networks (WSNs) play a pivotal role in enabling Internet of Things (IoT) devices with sensing and actuation capabilities. Operating in remote and resource-constrained environments, these IoT devices face challenges related to energy consumption, crucial for network longevity. Existing clustering protocols often suffer from high control overhead, inefficient cluster formation, and poor adaptability to dynamic network conditions, leading to suboptimal data transmission and reduced network lifetime. This paper introduces Low-Energy Adaptive Clustering Hierarchy with Reinforcement Learning-based Controller (LEACH-RLC), a novel clustering protocol designed to address these limitations by employing a Mixed Integer Linear Programming (MILP) approach for strategic selection of Cluster Heads (CHs) and node-to-cluster assignments. Additionally, it integrates a Reinforcement Learning (RL) agent to minimize control overhead by learning optimal timings for generating new clusters. LEACH-RLC aims to balance control overhead reduction without compromising overall network performance. Through extensive simulations, this paper investigates the frequency and opportune moments for generating new clustering solutions. Results demonstrate the superior performance of LEACH-RLC over state-of-the-art protocols, showcasing enhanced network lifetime, reduced average energy consumption, and minimized control overhead. The proposed protocol contributes to advancing the efficiency and adaptability of WSNs, addressing critical challenges in IoT deployments.

Figures (15)

• Figure 1: The number of publications on clustering protocols based on leach from 2000 to 2024. The data was obtained from the Scopus database using the keywords 'LEACH AND clustering AND protocol' elsevierScopusDatabase2025.
• Figure 2: System model of the iot network.
• Figure 3: Heatmap of the fnd metric for different values of $\alpha$, $\beta$, and $\gamma$.
• Figure 4: Proposed solution overview integrating the rl agent and the MILP-based clustering solution.
• Figure 5: Confusion matrix for the neural network-based cluster member assignment.