Sonar array characterization, experimental results

Measurements in the Levantine Sea with a seismic-type array [i.e., the high-frequency array (27 wavelengths at 348 Hz), the mid-frequency array (27 wavelengths at 175 Hz), and the low-frequency array (21 wavelengths at 58 Hz)] were found to have on average results within 1 dB of the theoretical signal gain. Observed signal gain degradations for peak-tracked and short integration times (1 min) had standard deviations from 2 to 3 dB and were caused by the combination of coherent multipaths, array shape, and array motion. The relative motion of source and receiver (5-8 kn) was an important cause of the average degradation at longer integration times (5 min). Equivalent plane wave beam noise levels were measured as a function of frequency, time, bearing, and aperture length. The beam noise level results show contributions from distant surface-ship-generated noise and natural environmental background noise. These results showed resolved distant shipping with median beam noise levels consistent with array noise gain 1-2 dB greater than the theoretical value for incoherent isotropic noise. The beam noise cumulative probability distribution function versus equivalent plane wave levels differed significantly from log-normality. Beam noise surfaces (beam noise levels versus time and bearing) show a higher density of ships for the high-frequency array when compared to the low-frequency array. Beam-to-beam cross correlations were found be sharply peaked and beam autocorrelation functions versus time showed zero crossing times on the order of 9-10 min. Significant space-time noise fade durations were observed at lower frequencies.