An adaptive likelihood fusion method using dynamic Gaussian model for indoor target tracking

It is hard to obtain a general error model for range-based wireless indoor target tracking system due to the complicated hybrid LOS/NLOS environment. In this paper, we employ a dynamic Gaussian model (DGM) to describe the indoor ranging error. A general Gaussian distribution is constructed firstly. The instantaneous LOS or NLOS error at a typical time is considered as the drift from this general distribution dynamically. Based on this modeling method, we propose an adaptive likelihood method of particle filter. Our method is adaptable for dynamic environment and achieves accurate estimation. The simulation and real indoor experiment demonstrate that the estimation accuracy of our algorithm is greatly improved without imposing computational complexity.