Přístupnostní navigace
E-application
Search Search Close
Publication detail
GROLL, H. ZÖCHMANN, E. GERSTOFT, P. MECKLENBRÄUKER, C.
Original Title
Global Robustness Characterization of DOA Estimators by Breakdown Detection
Type
conference paper
Language
English
Original Abstract
We characterize the global robustness of direction of arrival(DOA) estimators against additive outliers in the sensor ar-ray data by detection of the estimator’s breakdown. The pro-posed method contaminates a given proportion of the sensorarray data and observes the corresponding DOA estimate. Anoracle compares the DOA estimation error magnitude with achosen tolerance. The oracle detects that the DOA estimatehas broken down when the magnitude of error exceeds thetolerance. The estimator’s robustness is thus globally charac-terized in a robustness operating characteristic (ROC) curvewhich shows the true positive rate of detected breakdownsversus the false positive rate. The ROC curve is evaluatedfor several DOA estimators: conventional, phase-only, andweighted phase-only beamformer, as well as least absolutedeviation based DOA estimator.
Keywords
robust array processing; DOA estimation; least absolute deviation; phase-only processing; breakdown
Authors
GROLL, H.; ZÖCHMANN, E.; GERSTOFT, P.; MECKLENBRÄUKER, C.
Released
15. 12. 2019
ISBN
978-1-7281-5549-4
Book
Proceedings of 8th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)
Pages from
36
Pages to
40
Pages count
5
URL
https://ieeexplore.ieee.org/document/9022468
BibTex
@inproceedings{BUT170010, author="Herbert {Groll} and Erich {Zöchmann} and Peter {Gerstoft} and Christoph {Mecklenbräuker}", title="Global Robustness Characterization of DOA Estimators by Breakdown Detection", booktitle="Proceedings of 8th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)", year="2019", pages="36--40", doi="10.1109/CAMSAP45676.2019.9022468", isbn="978-1-7281-5549-4", url="https://ieeexplore.ieee.org/document/9022468" }