Chameleon-ACO Hybrid: Optimizing Power-Efficient Clustering and Routing in Wireless Networks

The issue of limited battery power in sensor nodes is a common challenge faced by wireless sensor networks. To tackle this problem, efficient clustering and multi-hop routing techniques can be employed to minimize communication and energy costs by aggregating similar data. This study introduces an innovative hierarchical clustering algorithm named Chameleon, which utilizes a dynamic cluster model and ant colony optimization to enable intelligent routing. To find the clusters that are most similar to one another, the “chameleon” method uses both the “relative interconnectivity and relative closeness” measurements. “Chameleon” outperforms the majority of existing clustering algorithms at identifying natural clusters of various shapes and sizes, according to experimental assessments carried out on numerous datasets with different features. The proposed algorithm employs a two-phase approach to cluster highly variable data. In the initial phase, a graph-partitioning algorithm divides cluster nodes into several sub-clusters of small size. Subsequently, in the second phase, the sub-clusters are iteratively combined using a suitable algorithm to identify authentic clusters. After the clustering process is finished, data gets transmitted from source to destination by using shortest multipath routing and ant colony optimization algorithms. According to simulation results, the proposed method has leveraged higher energy efficiency than the energy efficiency performance improvements of existing ant-based routing algorithm.