Optimum array processing in presence of randomly distorded wavefronts

Optimum array processing is generally studied and used in the case of perfectly known wavefronts. In sonar and in presence of non-rigid arrays or in shallow water, the wavefronts may be randomly distorted. This article intends to study optimum array processings in the case of randomly distorded wavefronts. One begins with different temporal or spectral modelizations of the phenomena. Then, one describes optimum array processings for detection or estimation by the use of criteria like signal to noise ratio, maximum likelihood, likelihood ratio, minimum variance, least squares. Finally one shows some results obtained by the first experiments.