A BP Method for Track-Before-Detect

Tracking an unknown number of low-observable objects is notoriously challenging. This letter proposes a sequential Bayesian estimation method based on the track-before-detect (TBD) approach. In TBD, raw sensor measurements are directly used by tracking algorithm without any preprocessing. Our proposed new statistical model that introduces object hypothesis for each data cell measurements. It allows to interact and contribute more than one cell. Based factor graph representing our model, we derive message passing equations belief propagation (BP) TBD. Approximations applied certain BP messages reduce computational complexity improve scalability. simulation experiment, BP-based TBD outperforms two other state-of-the-art methods.