Large-array signal processing in long-range underwater channels: coherence effects and adaptive approach to the multimode signal control

The signal coherence degradation in spatial domain is an intrinsic scenario for large-array signal processing if the array size is greater as compared with the coherence length. For underwater sound, study of this issue is physically motivated by the fact that the discrete spectrum signal-carrying modes lose the mutual covariance at long ranges, so the signal coherence loss at the array inputs can be considerable. This is a case of comparative study of the large-array beamformers, optimal one included, aimed at realistic estimations of the coherence-induced effects on their performance. In this paper, we address the other aspect of the problem, namely, the source array synthesis in a random-inhomogeneous multimode channel under the criteria specified for the signal processor in remote receiving array. The criteria of the output SNR and the array SNR gain are considered to be a reasonable choice. An adaptive approach to the multimode signal control by the use of vertical source array is then formulated as an iterative algorithm for the source excitations and the corresponding modal spectrum of the signal wavefield. Computer simulation is performed for the vertical source and receiving arrays operating in the shallow-water channel from the Barents Sea.