TDOA-Based Mobile Localization Using Particle Filter With Multiple Motion and Channel Models

Mobile terminal tracking is an important topic in wireless communications. A TDOA-based positioning method combining the interacting multiple models and the particle filter is proposed in this paper for non-cooperative target tracking in two dimensions. Three motion models, constant velocity, constant acceleration, and constant turning, and two-channel models, line of sight (LOS) and non-line of sight (NLOS), are considered to describe the trajectory of a mobile terminal in mixed LOS/NLOS environments. The particle filtering technique is employed for state estimation from a set of nonlinear TDOA measurements, and the interacting multiple models are used to mix the multiple motions and channel models to improve the accuracy of positioning. The simulation results are given to demonstrate that the proposed algorithm is close to the derived posterior Craḿer-Rao lower bound and outperforms the three-motion-model scheme with only an LOS channel model and the mixed LOS/NLOS schemes with single or two motion models. The average improvement over the whole time is larger than 40%.