Parameter estimation of acoustic signals using linear microphone arrays

A novel method is developed to sequentially analyze power spectrums, estimate the number of signals and their frequencies, direction of arrival (DOA), and signal-to-noise ratio (SNR). The relationship between the estimation bias and the array gain is investigated and an empirical model is built to improve the parameter estimations. An eight-element linear microphone array is made and used in testing the developed techniques. Five maximum likelihood algorithms and the MUSIC algorithm were used for DOA estimation and the results were compared. The test results show that the proposed methods were efficient and satisfactory estimation accuracy was achieved.