A Reduced-Complexity Data-Fusion Algorithm Using Belief Propagation for Location Tracking in Heterogeneous Observations

This paper presents a low-complexity and high-accuracy algorithm to reduce the computational load of the traditional data-fusion algorithm with heterogeneous observations for location tracking. For the location-estimation technique with the data fusion of radio-based ranging measurement and speed-based sensing measurement, the proposed tracking scheme, based on the Bayesian filtering concept, is handled by a state space model. The location tracking problem is divided into many mutual-interaction local constraints with the inherent message- passing features of factor graphs. During each iteration cycle, the messages with reliable information are passed efficiently between the prediction phase and the correction phase to simplify the data-fusion implementation for tracking the location of the mobile terminal. Numerical simulations show that the proposed forward and one-step backward refining tracking approach that combines radio ranging with speed sensing measurements for data fusion not only can achieve an accurate location close to that of the traditional Kalman filtering data-fusion algorithm, but also has much lower computational complexity.