Research on Distributed Collaborative Positioning Algorithm for Wireless Sensor Networks Assisted by Artificial Intelligence

This study addresses the challenge of distributed collaborative positioning in wireless sensor networks (WSNs) by integrating artificial intelligence (AI), deep learning, and edge computing into the positioning framework. We propose a two-tier distributed positioning protocol, in which anchor nodes broadcast beacon signals while mobile nodes exchange time difference of arrival (TDOA) measurements and gradient updates through lightweight communication. A lightweight deep neural network (DNN), deployed on edge nodes, is introduced to refine initial position estimates derived from traditional Chan/Taylor methods by learning nonlinear signal propagation patterns in dynamic environments. Each node maintains a neighbor table to support consensus-based position refinement, enabling adaptive, and energy-efficient coordination across the network. Extensive experiments demonstrate that the proposed AI-enhanced distributed scheme significantly improves positioning accuracy and robustness, while maintaining low latency and communication overhead, making it a scalable and practical solution for real-world WSN deployments.