Application of robust capon beamforming to vector sensor array

An acoustic vector sensor can measure acoustic pressure and the three orthogonal components of the acoustic particle velocity at a single point in space.Unlike pressure-sensor array,the left/right ambiguity problem does not arise when using vector-sensor array.Furthermore,The inter-sensor spacing of a vector-sensor array can be increased beyond a half wavelength to extend the spatial aperture to a certain extent.It is well known that the MVDR beamformer has better resolution and much better interference rejection capability than the classic beamformer,provided that the array steering vector corresponding to the signal of interest(SOI) is accurately known.However,in most applications,knowledge of the SOI steering vector is inaccurate.In some case such as towed arrays,the steering vector is time-varying due to the change in the sensor′s positions.Whenever this happen,the performance of MVDR degrades severely.The robust capon beamforming(RCB) is a recently proposed robust adaptive beamforming algorithm.It estimates the steering vector directly based on a steering vector uncertainty region,thus avoiding the su-ppression of SOI caused by steering vector mismatches.In this paper,we extend RCB to the case of rank-deficient and combine it with the vector-sensor array.Experimental result at sea shows the excellent performance of this method.