Abstract
In this paper, a full ESPRIT-like method called high-order spatial-temporal algorithm (HOSTA) is presented for classification and localization of fourth-order stationary sources in a scenario where both far-field and near-field sources are present. By constructing one special spatial cumulant matrix, without range information, and controlling its rank in a novel manner, the candidate DOA vector (CDOAV) is extracted. By constructing several spatial-temporal cumulants, regarding the fourth-order stationary property, the parameters estimation matrix is achieved. Consequently, invalid DOAs are eliminated to get a DOA estimation vector. By introducing a rank matching procedure, a reasonable near and far classification, and DOA and range estimations are realized based on a new solution for extracting electric angles in parallel. The proposed method (HOSTA) can detect near and far field sources possessing the same DOAs. Furthermore, the method does not require any spectral search, which reduces the computational complexity to a significant extent. Furthermore, HOSTA does not state any spurious DOA, as well as prevents array aperture loss without any pairing parameters. The simulation results validate its high performance in terms of estimation accuracy and resolution, and also its robustness against partially correlated sources.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call