An Energy-Efficient and Dynamic Clustering Approach for Wireless Sensor Networks

چکیده مقاله

This paper presents an innovative clustering algorithm designed to address the critical challenge of energy efficiency in Wireless Sensor Networks (WSNs), where limited battery power constrains network lifespan. Recognizing that node mobility is often overlooked in traditional clustering protocols, the authors propose the Dynamic Dual Head Clustering algorithm for WSNs (DDHCWSN). The core objective is to significantly reduce energy consumption and extend network operational life by introducing a novel, multi-faceted approach to cluster management that adapts to dynamic network conditions.

The key innovation of DDHCWSN lies in its four-phase architecture, which fundamentally rethinks cluster organization and communication. Firstly, the network is divided into specific zones during initialization to ensure an even distribution of clusters. Within each zone, the algorithm selects not one, but two distinct Cluster Heads (CHs): an Aggregation Cluster Head (ACH) responsible for gathering and processing data from member nodes, and a Transmission Cluster Head (TCH) dedicated to communicating with other TCHs and the Base Station (BS). This separation of duties prevents the early energy depletion of a single CH, effectively balancing the load. Furthermore, DDHCWSN employs a dynamic update mechanism for clusters based on node energy levels and mobility, avoiding the computationally expensive process of complete re-clustering in each round. Additionally, the TCH intelligently decides between single-hop or multi-hop communication to the BS in real-time, optimizing transmission paths based on distance.

The experimental evaluation demonstrates the clear superiority of DDHCWSN across six diverse scenarios that vary in network size, density, and BS location. Compared to several state-of-the-art algorithms including LEACH variants, CRPD, and DMH-LEACH, DDHCWSN consistently achieved a longer network lifetime, particularly in larger, sparser networks. For instance, in a large-scale 300x300 network with the BS outside the area, DDHCWSN sustained operation for 1,455 rounds, far outperforming other protocols. The algorithm also exhibits a slower, more uniform rate of energy consumption, indicating efficient and balanced energy utilization across all nodes. While its execution time is moderately higher than the fastest algorithms due to its sophisticated decision-making processes, this is a justified trade-off for the substantial gains in network longevity and energy savings.

In conclusion, the DDHCWSN algorithm represents a significant positive contribution to energy-efficient WSN management. Its main achievements are the successful integration of node mobility into the clustering logic, the intelligent load distribution via dual CHs, and the adaptive communication strategy, all of which collectively lead to a markedly extended network lifespan. The algorithm shows robust and favorable performance, especially in scalable and dynamic environments, making it a promising solution for real-world WSN applications where energy conservation is paramount.