Bayesian Joint Localization and Tracking Algorithm Using Multiple-Input Multiple-Output Radar

We present a novel joint localization and tracking algorithm for multiple-input multiple-output active radars. The proposed algorithm, which we dub Bayesian localization and tracking (BLaT), relies on approximate Bayesian inference using the mean field approach and processes all available received data to jointly estimate the target's track and location. This approach makes it possible to take advantage of the inherent synergy between the tracking and localization tasks. BLaT is shown by simulation to outperform a classical sequential processing baseline in terms of its ability to track targets in low signal-to-noise ratio conditions as well as superior tracking of manoeuvring targets.