Low-sample performance of reduced-rank power minimization based jammer suppression for GPS

In the GPS receiver environment, narrowband jammers may have the ability to change frequencies dynamically. In addition, the rapid dynamics of the aircraft during maneuvering can cause arrival angles of wideband jammers to vary rapidly. Thus, an interference suppression algorithm will only be effective if it can rapidly converge with a small sample size. We investigate the performance of the multistage nested Wiener filter (MSNWF) algorithm under conditions of low sample support. It is argued that the MSNWF provides rapid convergence relative to standard LMS based adaptations, and a significant reduction in computational complexity relative to standard RLS based adaptations, especially in the case of large dimension adaptive weight vectors. It is shown that the MSNWF is a key enabling "technology" facilitating real-time implementation of the space-time adaptive algorithms for GPS interference suppression.