A novel two-dimensional direction-of-arrival estimation approach with sparse array geometry

Abstract

A novel two-dimensional (2-D) direction-of-arrival (DOA) estimation method exploiting cyclostationarity of the sources is presented. The proposed array of the method is composed of two parallel minimum-redundancy linear arrays (MRLAs). Compared with the two parallel uniform linear arrays (ULAs), this new sparse array geometry requires fewer sensors on condition of identical array aperture. Furthermore, by constructing the pseudo-data Matrices on the basis of the conjugate cyclic correlation functions of sensor outputs, the proposed method also extends the effective array aperture, which consequently improves the DOA estimation performances. In addition, the new method does not require any 2-D searching or pairing procedures. Computer simulations demonstrate the capabilities of our method.