Contiguously clustered linear arrays through Bayesian compressive sensing

Abstract

Bayesian Compressive Sensing (BCS) is applied for the synthesis of contiguously clustered linear arrays. The standard sub-array problem is formulated as a probabilistic BCS synthesis problem and the Relevance Vector Machine (RVM) is used to obtain a sparse contiguous non-overlapping subarray configuration which has maximal far-field pattern match with a given reference pattern. Selected numerical experiment results are reported to validate the synthesis technique.