Bistatic Track Association Based on Singular Spectrum Analysis

The association stage of bistatic tracks is an unavoidable part of every Multi-Static Primary Surveillance tracker when there is no identification of a target. Association process creates hypotheses of a common source in cartesian space for bistatic tracks from different receiver-transmitter pairs. However, when tracking multiple objects at the same time, this process can lead to the creation of false cartesian tracks. There are a few methods of these tracks elimination such as the ellipsoids of occurrence proximity test or probabilistic approaches. In this paper, we try to bring another solution to this problem based on Singular Spectrum Analysis of bistatic tracks.