Failure detection in large arrays through Bayesian compressive sensing

A method for an efficient and reliable diagnosis of large phased arrays based on a Bayesian compressive-sensing (BCS) strategy is presented in this paper. The approach allows to determine the elements which have been damaged also proving an estimation of the degree of reliability of the solution. The far-field measured data are processed by means of an efficient algorithm based on a relevance vector machine (RVM). Representative numerical examples are reported in order to validate the method dealing with failure detection in large linear arrays.