Suppression of moving interference with SIS-based projections, in shallow water environments

We present an approach for suppressing moving interference in shallow water environments. Adaptive beamforming techniques are based upon applying a covariance matrix inverse or projection operator to the data, in order to suppress noise and interference. Their performance typically degrades in the presence of moving interference, since more adaptive degrees of freedom are needed to suppress it. In this work we propose a method for obtaining a projection operator that dynamically evolves with time. The method uses sequential importance sampling (SIS) to obtain sequences of wavefronts based on hypothesized velocities and depths. The sequences are associated with weights whose magnitude indicates how well they represent the data. We base the projection operator on the dominant sequences at surface, reducing the rank required for nulling of the surface interferer.