Abstract
This paper presents a tree-structured subspace-based algorithm for joint estimation of the directions of arrival (DOAs) and frequencies of arrival (FOAs) in wireless communication systems. The proposed approach is a hybrid of one-dimensional (1-D) subspace-based algorithms and spatial/temporal filtering processes, both of which are invoked alternatively to enhance the estimation accuracy. Two temporal and one spatial 1-D subspace-based algorithms are employed alternatively to estimate the FOAs and the DOAs, respectively. Between these subspace-based algorithms, a constrained temporal filtering process and a constrained spatial beamforming process are addressed, which minimize the filtered noise power under a set of linear constraints. These filtering processes aim to effectively partition the incoming rays and to be robust against the propagation errors in the tree-structured estimation scheme so that the overall performance can be enhanced. Furthermore, the estimated FOAs and DOAs are automatically paired without extra computational overhead. Furnished simulations show that the new approach can provide comparable performance with reduced complexity compared with previous works.