[Back]


Contributions to Proceedings:

D. Löschenbrand, C. Mecklenbräuker:
"Fast Antenna Characterization via a Sparse Spherical Multipole Expansion";
in: "2016 4th International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing (CoSeRa)", IEEE Xplore, 2016, 212 - 216.



English abstract:
Reliable antenna characterization at high frequencies requires a considerable amount of time and often leads to the inefficient use of measurement facilities. The proposed method cuts the number of necessary samples for spherical antenna scans by 75% as compared to standard techniques, thus enabling significantly faster scan times without notable loss of accuracy. As a first step, a multipole expansion for spherical antenna measurements is introduced and its coefficient domain is identified as a sparsity transform suitable for compressed sensing approaches. Secondly, a numerical evaluation of the transform's Restricted Isometry Property using the maximum directivity antenna as a worst case scenario provides empirical bounds on the number of necessary samples for successful reconstruction. A novel method for phase center determination based on a 1-norm minimization of the multipole expansion coefficients is proposed which outperforms the standard Least Squares method by a factor of 100 in terms of accuracy. A similar 1-norm minimization of the coefficients is used to find the axis of maximal rotational symmetry of an antenna. The knowledge of the phase center and the axis of rotational symmetry in turn give rise to optimized translation and rotation operations which, in a third step, decrease sparsity of the spherical multipole expansion. Combining estimation of the phase center, estimation of the rotational symmetry axis and reconstruction of the radiated field into an iterative algorithm leads to a reliable procedure for spherical antenna characterization in an optimized coordinate system which requires only a fraction of the sampling points. Applying the proposed strategy to synthetic as well as measured antenna data demonstrates its powerful capabilities in terms of decreasing sparsity in the multipole coefficients domain and therefore increasing the quality of the reconstruction, all while maintaining accuracy.

Keywords:
antenna measurements, sparsity


"Official" electronic version of the publication (accessed through its Digital Object Identifier - DOI)
http://dx.doi.org/10.1109/CoSeRa.2016.7745731


Created from the Publication Database of the Vienna University of Technology.