I. Introduction
Noise reduction algorithms are crucial in hearing aids to improve the intelligibility in background noise. Multimicrophone systems exploit spatial information in addition to temporal and spectral information of the desired signal and noise signal, and are thus preferred to single microphone procedures (such as spectral subtraction). In hearing aids, the array length is limited because of aesthetical and practical reasons. In [1], [2], it is shown that noise can be reduced even at low , i.e., (with the frequency [Hz], the microphone interspacing [m] and the velocity of sound ()), be it at the expense of an increased noise sensitivity. The noise sensitivity is defined as the ratio of the spatially white noise gain to the gain of the desired signal [2], [3]. It is used to quantify the sensitivity to errors in the assumed signal model and processing errors (such as microphone mismatch, round off errors due to finite processing accuracy). A noise reduction algorithm should however be robust, i.e., insensitive to small signal model errors. The increased noise sensitivity shows that the robustness of multimicrophone noise reduction algorithms may be crucial for practical use in hearing aids.