Bernoulli filter for track-before-detect using MIMO radar

Abstract

In this letter, we apply a Bernoulli Filter for moving target detection and tracking using real Multiple Input Multiple Output radar data in a challenging environment. The interest in MIMO radars has been growing since they can provide high performance at lower cost than conventional antenna arrays. The Bernoulli Filter performs tracking using the Track-Before-Detect approach, which is required to achieve satisfactory tracking performance in low signal-to-noise scenarios. The algorithm is validated with real MIMO data of a walking person.