Acoustic vector-sensor array processing

Abstract

Existing theory yields useful performance criteria and processing techniques for acoustic pressure-sensor arrays. Acoustic vector-sensor arrays, which measure particle velocity and pressure, offer significant potential but require fundamental changes to algorithms and performance assessment. This thesis develops new analysis and processing techniques for acoustic vector-sensor arrays. First, the thesis establishes performance metrics suitable for vector sensor processing. Two novel performance bounds define optimality and explore the limits of vector-sensor capabilities. Second, the thesis designs non-adaptive array weights that perform well when interference is weak. Obtained using convex optimization, these weights substantially improve conventional processing and remain robust to modeling errors. Third, the thesis develops subspace techniques that enable near-optimal adaptive processing. Subspace processing reduces the problem dimension, improving convergence or shortening training time.