基于稳健Capon波束形成的阵形校正

由于受地形、海流和布放方法等的影响,声阵在布放后通常会偏离原定阵形。若直接利用原定阵形作方位估计,则各种高分辨方位估计算法的性能通常会退化甚至失效。因此,为了在实际中应用这些算法,必须对阵元的位置进行校正。RCB(Robust Capon Beamforming)是最近提出的一种稳健自适应波束形成算法,该算法直接对导向矢量进行估计,并用估计的导向矢量作波束形成,从而有效避免了因阵列流形失配而导致的算法性能下降。借用RCB的思想,提出一种新的阵形校正方法,该方法对导向矢量进行估计,然后用估计的导向矢量推导出阵元的位置。湖试结果表明了该校正方法的有效性。

A new array shape calibration method based on robust capon beamforming

Direction finding is one of the central tasks of sonar. It is well known that most of the high-resolution algorithms have good performance provided that the positions of hydrophones are accurately known. The array shape usually departs from the planned original due to various reasons such as compli-cated terrain at seabed,ocean current,and even the method of deployment. If we still use the planed shape,performance of these high resolution algorithms will degrade severely. Therefore,before applying these al-gorithms to reality,we must have the array shape calibrated. Robust Capon beamforming (RCB) is a rec-ently proposed robust adaptive beamforming algorithm. It directly estimates the steering vector based on a steering vector uncertainty region,thus avoiding suppression of the signal of interest due to steering vector mismatch. In this paper,we propose a new array shape calibration method that borrows the idea from RCB,i.e.,we estimate the steering vector and further deduce the positions of hydrophones. Results of lake ex-periment show effectiveness of the proposed method.