| dc.description.abstract | In this study we are considering an infrasound dataset from the array station IS37. We are using array processing in order to estimate the direction of arrival of incoming microbaroms. The study is an assessment of what is a better window length, when computing power maps using the beamformers, Capon's method and Delay-and-Sum. We started by considering a smaller project of an infrasound stream containing a transient signal from a volcano eruption. From here we build a framework of how to process longer streams on infrasound, and predict the direction of arrival from microbaroms. Then we considered a stream of infrasound recorded from an infrasound array, IS37 at Bardufoss. After numerous computations of power maps, and evaluating various plots, we found that the Capon's method was sensitive to present incoherent noise. Because of this, the method benefits from using longer window lengths, as the variance is lowered. The Delay-and-Sum beamformer was more robust and handled the incoherent noise better. Because of the robustness, we can use shorter window lengths, for less noise in the plots and a shorter processing time. | eng |