MST-ID Algorithm for Angular Parameter Estimation of Incoherently Distributed Sources

This paper presents an effective angular parameter estimation method based on the manifold separation technique (MST) for incoherently distributed sources, named as MST-ID algorithm. In the proposed method, at first, a mathematical model is established through the first-order Taylor expansion of the steering vector, in which the nominal direction of arrival (DOA) can be decoupled from the angular spread. Then, the decoupled steering vector is divided into two sub-steering vectors with equal dimensions, and further the nominal DOA is estimated by the shift invariant structure between the sub-steering vectors. Finally, the MST is used to separate the antenna array structure from the nominal DOA in the array steering vector, such that the signal covariance matrix can be easily obtained according to the antenna array structure and the estimated nominal DOA. On this basis, a spectrum search function is given to estimate the angular spread. Compared with the previous works, the presented method can not only improve the angular parameter estimation accuracy but also be suitable for arbitrary line array structures. Theoretical analysis and simulation results confirm the effectiveness of the proposed method.

Automated summary

This paper presents an effective angular parameter estimation method named MST-ID algorithm based on the manifold separation technique (MST) for incoherently distributed sources. The proposed method establishes a mathematical model using the first-order Taylor expansion of the steering vector to decouple the nominal direction of arrival (DOA) from the angular spread. Then, the decoupled steering vector is divided into two sub-steering vectors, and the nominal DOA is estimated using the shift invariant structure between them. The MST is used to separate the antenna array structure from the nominal DOA, facilitating the acquisition of the signal covariance matrix and the estimation of the angular spread using a spectrum search function. The presented method improves the accuracy of angular parameter estimation and is suitable for arbitrary line array structures, as confirmed by theoretical analysis and simulation results.