HOSNA: Boosting Smart Agriculture Efficiency With The Hybrid Optimization-Based Sensor Node Activation Model

V. Kavitha, Viktor K. Prasanna, S. Lekashri, M. VenkatesanJournal of Machine and Computing 2025 · DOI

Abstract

Smart agriculture leverages Wireless Sensor Networks (WSNs) to monitor environmental parameters such as soil moisture, temperature, and humidity, enabling precision farming and efficient resource utilization. The Hybrid Optimization-Based Sensor Node Activation (HOSNA) model designed to enhance the efficiency and lifespan of Wireless Sensor Networks (WSN) in smart agriculture applications. HOSNA integrates clustering, energy-efficient activation, hybrid optimization algorithms, and machine learning to optimize sensor node operations while ensuring accurate and real-time environmental monitoring. The model employs Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to determine optimal sensor activation schedules, reducing energy consumption and prolonging network lifetime. Additionally, a Long Short-Term Memory (LSTM) neural network predicts environmental changes, allowing proactive sensor activation. Simulation results demonstrate that HOSNA achieves a 94.0% data accuracy after 1000 operational rounds, surpassing LEACH (90.0%), PEGASIS (86.0%), and Random Duty Cycling (RDC) (70.0%). Energy consumption reduced by 24% compared to LEACH, while network lifetime extended by 32% over PEGASIS. These results highlight HOSNA’s ability to provide reliable, energy-efficient, and scalable solutions for precision agriculture. Future improvements could involve adapting the model for heterogeneous sensor networks and integrating solar-powered nodes for sustainable energy.
BibTeX
@article{kavitha2025hosna,
  author = {V. Kavitha and Viktor K. Prasanna and S. Lekashri and M. Venkatesan},
  title = {{HOSNA: Boosting Smart Agriculture Efficiency With The Hybrid Optimization-Based Sensor Node Activation Model}},
  journal = {Journal of Machine and Computing},
  year = {2025},
  doi = {10.53759/7669/jmc202505040},
}
Figure