HBIA-DNN Based Framework for Efficient Node Placement and Target Tracking in Wireless Sensor Networks

Wireless Sensor Networks (WSNs) represent a rapidly advancing technology with applications in diverse fields, including surveillance, smart environment development, and target tracking. Despite their versatility, WSNs continue to face persistent challenges in optimizing energy consumption and network longevity, particularly for demanding tasks like dynamic target tracking, often due to inefficient node deployment. This study introduces the Efficient Node Placement and Target Tracking Using Machine Learning (ENTML) framework, a novel method designed to address these constraints through the integration of machine learning techniques. The Hybrid Bird-Inspired Algorithm (HBIA) is utilized to compute optimal, energy-efficient node placements for establishing an efficient network topology. Meanwhile, an adaptive Deep Neural Network (DNN) model supports real-time adaptive tracking of targets by processing sensor data and dynamically adjusting parameters in real-time. This combination approach optimizes both network structure and operational responsiveness. Comprehensive simulations were conducted to evaluate ENTML against existing methods in terms of various performance metrics, including energy consumption, network lifetime, end-to-end network delay, packet delivery ratio, and packet loss, for diverse target mobility scenarios. The experimental results demonstrate the superiority of the proposed framework over the existing state-of-the-art approaches, achieving a significant 24% reduction in overall energy consumption, a 31% decrease in end-to-end delay, and a 9% extension of network lifetime. Furthermore, ENTML was also shown to provide better packet delivery ratios along with less packet loss compared to baseline methods. The outcome of these experiments highlights the remarkable advantages of utilizing combined machine learning methods, such as HBIA and DNN, to create more robust, energy-efficient sensor nodes along with adaptive WSNs tailored for challenging practical scenarios in dynamic environments. This research provides a valuable contribution towards the development of intelligent and sustainable WSN solutions.