On MAP-Based Target Tracking Using Range-Only Measurements

We consider the target tracking problem using range-only measurements. Unlike the existing tracking methods, which are based on the minimum mean square errors (MMSE) criterion, we propose a maximum \emph{a posteriori} (MAP) estimation method. Owing to the fact that the posterior density is multi-modal, we use a mixture of Gaussian distributions to approximate it to reduce the approximation errors using single-modal distributions. To obtain the initial distribution, we divide the initial angle estimate range to obtain a series of Gaussian distributions. Due to this, we name the proposed method as "angle-parameterized MAP" (APMAP) method. Simulation results show that the proposed method is a good trade-off between performance and computational complexity.