A comparison of modal decomposition algorithms for matched-mode processing

This paper compares a variety of modal decomposition methods used in matched-mode processing (MMP) for ocean acoustic source localization. MMP consists of decomposing far-field acoustic measurements at an array of sensors to obtain the constituent mode excitations (modal decomposition), and then matching these excitations with modelled replica excitations computed for a grid of possible source locations. Modal decomposition can be ill-posed and unstable if the sensor array does not provide an adequate spatial sampling of the acoustic field, so the results of different approaches can vary substantially. Solutions can be characterized by modal resolution and solution covariance; however the ultimate test of the utility of the various methods is how well they perform as part of a MMP source localization algorithm. In this paper, the resolution and variance of the methods are examined using an ideal ocean environment, and MMP results are compared for a series of realistic synthetic test cases, including a variety of noise levels and sensor array configurations. Zeroth order regularized inversion is found to give the best results.