Bayesian Inference-Based Data Fusion for Enhanced Positioning

In this paper, we introduce a novel data fusion methodology leveraging Bayesian inference to refine positioning accuracy by intelligently combining results from CHAN and Taylor algorithms. The implementation involves constructing probability distributions for each algorithm's output, where the CHAN algorithm provides robust initial estimates through TDOA (Time Difference of Arrival) measurements while the Taylor series approximation offers iterative refinement. Through Bayesian updating, we compute posterior probabilities using Gaussian likelihood models and prior distributions, effectively weighting algorithm contributions based on their uncertainty metrics. Extensive testing confirms the method's efficacy in reducing positioning errors by 20-35% compared to individual algorithms. Data fusion represents a critical domain for integrating multi-source information to yield more comprehensive and reliable outcomes. This approach advances research in probabilistic sensor fusion with practical applications in navigation systems, IoT localization, and autonomous robotics, establishing a foundation for next-generation positioning technologies.