Node Selection in Edge Computing (Target Tracking, Fisher information)

This research paper examines node selection in Edge Computing and Wireless Sensor Networks (WSNs), with a particular focus on target tracking, energy efficiency, and Fisher Information Matrix (FIM)-based decision-making. The study addresses the limitations of traditional positioning methods, where collecting sufficient measurements from anchor nodes can be difficult, costly, or impractical, particularly in indoor and complex environments. The research explores cooperative positioning, where mobile and neighboring sensor nodes share information to improve positioning accuracy. It highlights the challenge of selecting only the most useful nodes because involving unnecessary nodes can increase communication overhead, computational complexity, and energy consumption. The paper therefore reviews node selection approaches based on distance, information utility, remaining energy, and Fisher Information Matrix techniques. The paper also presents an Efficient and Adaptive Node Selection (EANS) approach for target tracking. The approach dynamically selects appropriate sensor nodes while allowing unnecessary nodes to remain in sleep mode, thereby reducing energy consumption while maintaining tracking performance. It combines target prediction using a particle filter with information-utility-based node selection. Overall, the research focuses on developing more efficient node-selection strategies for wireless and edge-enabled environments by balancing positioning accuracy, computational complexity, communication overhead, and energy consumption. The reported simulation findings indicate improvements in energy cost, computational complexity, and target-tracking accuracy compared with the referenced approaches.