We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Off-Grid Direction-of-Arrival Estimation Based on Steering Vector Approximation.
- Authors
Wang, Ben; Gu, Yujie; Wang, Wei
- Abstract
In this paper, a novel off-grid direction-of-arrival (DOA) estimation algorithm is proposed based on steering vector approximation. The conventional multiple signal classification (MUSIC) technique estimates the DOA with a uniformly discretized grid by assuming that the true DOAs lie on the grids. In practice, this assumption is rarely satisfied. In order to improve the estimation accuracy, a dense sampling grid is required for the MUSIC technique. Besides the higher computational complexity, it is still not guaranteed that the dense sampling grids will cover the true DOAs. In order to address the off-grid problem with a lower computational complexity, we use the Taylor series expansion to approximate the steering vectors. We first construct a two-dimensional spectrum searching method to simultaneously estimate the nearest on-grid DOA and the corresponding off-grid error. To further improve the estimation accuracy, we then solve an optimization problem to obtain the accurate DOA estimation, which is formulated by the fact that the signal and noise subspaces are orthogonal to each other. Simulation results demonstrate the effectiveness of the proposed off-grid DOA estimation algorithm.
- Subjects
DIRECTION of arrival estimation; APPROXIMATION theory; PARAMETER estimation; COMPUTER algorithms; SIGNAL classification
- Publication
Circuits, Systems & Signal Processing, 2019, Vol 38, Issue 3, p1287
- ISSN
0278-081X
- Publication type
Article
- DOI
10.1007/s00034-018-0914-5