Acoustic source localization using LS-SVMs without calibration of microphone arrays

Common assumptions of the conventional approaches to acoustic source localization are usually that the microphones used are ideal and that the locations of the microphones are also known a priori, which usually may not hold in practice. Therefore, the microphone arrays need to be calibrated carefully before use. However, it is not an easy task to calibrate microphone arrays perfectly. In this paper, we proposed an algorithm for acoustic source localization based on the least-squares support vector machines (LS-SVMs). The advantage of the proposed algorithm is that it requires no calibration of microphone arrays. The performance and effectiveness of the proposed method is demonstrated by simulation results and the real-data experiments.