| 1. |
- Yang, Bin, 1982, et al.
(author)
-
Array Calibration using Array Response Interpolation and Parametric Modeling
- 2015
-
In: Proc. European Signal Processing Conference 2015 (EUSIPCO 2015). - 2076-1465. - 9780992862633 ; , s. 1346-1350
-
Conference paper (peer-reviewed)abstract
- High-performance array applications often require an accurate array response model. A common way to achieve this is by array calibration which involves measuring the response for a finite number of given source directions and employing interpolation. This paper considers the array calibration problem by combing interpolation techniques and parametric modeling. The idea is to model the array response as a product of a mutual coupling matrix, an ideal array response vector (derived from the geometry of antenna array) and an angle-dependent correction vector. Since the major effects are captured by the physical model and the mutual coupling matrix, the correction vector will be a smoother function of angle as compared to direct interpolation of the measured array response. In numerical experiments of a real antenna array, the method is found to improve the performance of the array calibration significantly.
|
|
| 2. |
- Yang, Bin, 1982, et al.
(author)
-
Array response interpolation and DOA estimation with array response decomposition
- 2016
-
In: Signal Processing. - : Elsevier BV. - 0165-1684. ; 125, s. 97-109
-
Journal article (peer-reviewed)abstract
- High-resolution direction-of-arrival (DOA) estimation requires an accurate array response model, which is usually achieved by measuring the response for given directions of the sources and employing interpolation. This paper considers the array response interpolation and DOA estimation by coherently combing parametric modeling and interpolation techniques. The proposed method is to decompose the array response as a product of a mutual coupling matrix, an ideal array response vector (dependent on the geometry of antenna array) and a DOA-dependent correction vector. Our idea is now that the correction vector will be a smoother function of DOA as compared to direct interpolation, since parts of the phase variation in one embedded element are due to the mutual coupling and the ideal array response. In numerical experiments of a real antenna array, the method is found to improve the performance of array response interpolation and DOA estimation significantly when the array response is updated by interpolation of the correction vector. In addition, we show how the DOA estimation performance can be predicted using a simple model of the interpolation error statistics.
|
|
