Estimation of broadband signal radiated by a single source in the presence of multiple acoustic interference sources via microphone array

The scenario is considered when a mix of signals from multiple acoustic sources is received by a microphone array; the problem is to estimate the waveform of one of these sources. It is supposed that the position of the desired source is a priori known. The considered problem can arise in many applications like video conferencing, acoustic room surveillance and others, where it is necessary to capture human speech against acoustic interferences. To solve this problem, an adaptive algorithm is independently applied to each narrow band of the received signal. The adaptive algorithm employs the interference model consisting of a set of unknown steering vectors. It is known that most adaptive beamforming algorithms suffer from degradation due to imperfections in desired signal spatial model and finite sample size effect. The proposed method is shown to be quite robust to these effects. Such robustness is achieved without extra information about the error in the desired signal spatial model. The results of numerical modeling and experiments confirm the effectiveness of the proposed method.