Enhanced Signal Processing in IoT-Enabled Wireless Sensor Networks and VANETs Using the Butterfly Optimization Algorithm and K-Means++ Clustering

Abstract

IoT-enabled Wireless Sensor Networks (WSNs) and Vehicular Ad Hoc Networks (VANETs) utilize the Butterfly Optimization Algorithm (BOA) with K-Means++ clustering to enhance data transmission, energy management, and real-time communication. Signal processing in WSNs and VANETs faces challenges such as uneven energy distribution, suboptimal clustering, high latency, and reduced network lifetime, which are further complicated by scalability and dynamic topology in IoT environments. The methodology begins with initializing sensor and vehicular nodes, followed by K-Means++ clustering to form energy-efficient clusters, minimizing intra-cluster distances and optimizing data aggregation. Cluster Heads (CHs) are selected based on residual energy, mobility, and proximity to ensure efficient data relay. BOA optimizes signal processing by mimicking butterfly behaviors through global and local searches, iteratively refining configurations to balance energy efficiency, latency, and signal quality. This hybrid approach enhances network performance by minimizing energy consumption, extending network lifetime, and improving real-time data transmission. By leveraging BOA's optimization and K-Means++'s effective cluster formation, the proposed model outperforms existing methods. Results indicate improved energy efficiency, reduced latency, superior signal quality, and enhanced vehicular communication stability in dynamic environments.