A Fast Sparse Covariance-Based Fitting Method for DOA Estimation via Non-Negative Least Squares

Chundi Zheng; Huihui Chen; Aiguo Wang

ICASSP 2020accepted0 citations

A Fast Sparse Covariance-Based Fitting Method for DOA Estimation via Non-Negative Least Squares

Abstract

A fast sparse covariance-based fitting algorithm with the non-negative least squares (NNLS) form is proposed for the direction of arrival (DOA) estimation. The Khatri-Rao product of the array manifold of the uniform linear arrays is utilized to achieve the dimension reducing transformation after vectorizing the array covariance matrix. Furthermore, the DOA estimation problem is derived as a NNLS problem by using the non-negative property of the spatial spectrum, which can be solved by some efficient solvers. Numerical experiments show that the proposed method can obtain high resolution with a competitive computational complexity, as well as works in the presence of coherent sources.

BibTeX

@inproceedings{icassp2020_afastsparsecovar,
  title = {A Fast Sparse Covariance-Based Fitting Method for DOA Estimation via Non-Negative Least Squares},
  author = {Chundi Zheng and Huihui Chen and Aiguo Wang},
  booktitle = {ICASSP 2020},
  year = {2020}
}