Effect of noise field and array configuration on Matched-Field Processing in underwater acoustics

The performance that can be achieved using Matched-Field Processing (MFP) depends on the nature of the ambient noise and the array configuration used to measure the field. In this paper, computer simulations were used to estimate MFP performance for an underwater source in different ambient noise fields and for different array configurations. Best MFP performance was obtained in a thermal ice cracking noise field and poorest performance in a fishing boat noise field. For arrays with fewer than 40 sensors in a surface noise field, best localization was obtained with a vertical array geometry. With more than 40 sensors, billboard arrays (composed of multiple vertical line arrays) outperformed horizontal line arrays which in turn outperformed vertical line arrays.