Enhancing data transmission efficiency in wireless sensor networks through machine learning-enabled energy optimization: A grouping model approach

Wireless Sensor Networks (WSNs) which plays a crucial role in data transmission in today’s digital era faces various types of research challenges due to limited energy constraints in sensor nodes. This study introduces a Machine Learning-based Energy Optimization Approach (ML-EOA) to overcome these research challenges. ML-EOA integrates data aggregation, an Artificial Neural Network (ANN) for Cluster Head (CH) selection, a steady-state phase for data transmission, and a Fuzzy Logic-based method for update/sleep cycle calculation, combined with a multi-objective fitness function. This approach enhances energy optimization, network coverage, and reduces latency. Simulation results show ML-EOA outperforming traditional methods by extending network lifespan to 140 min, reducing energy usage by 19.8 %, achieving a data delivery ratio of 92.6 %, ensuring 92.8 % network coverage, and lowering latency to 10.3 ms. These advancements in ML-EOA improve data reliability, monitoring, and scalability in WSNs, marking a significant step towards sustainable data transfer and opening new research possibilities for diverse applications.