Multitarget tracking using binary directional information

In this paper, we present a new methodology for the tracking of multiple targets using binary directional informations provided by a sensor network. While several works have been done on multiple target tracking, this is the first time an algorithm is proposed with this kind of directional observations. Our algorithm, inspired by the Adaptive Importance Sampling approach, uses the binary informations from the sensors and their locations to provide an estimator of the normalized velocities of the targets. Then, these estimated velocities are used to estimate the location of the targets at each time t. The estimation is completed by the construction of a confidence region for the targets' location at each time. We provide several simulation scenarios that demonstrate the efficiency of our algorithm.