A Method for 3D Direction of Arrival Estimation for General Arrays Using Multiple Frequencies

Fredrik Andersson (1) , Marcus Carlsson (1) , Jean-Yves Tourneret (2, 3) , Herwig Wendt (4, 5)

1 Skane University Hospital [Lund]
2 IRIT-MINDS - CoMputational imagINg anD viSion
3 Toulouse INP - Institut National Polytechnique (Toulouse)
4 IRIT-SC - Signal et Communications
5 CNRS - Centre National de la Recherche Scientifique

We develop a novel high-resolution method for the estimation of the direction of incidence of energy emitted by sources in 3D space from wideband measurements collected by a planar array of sensors. We make use of recent generalizations of Kronecker's theorem and formulate the direction of arrival estimation problem as an optimization problem in the space of sequences generating so called “general domain Hankel matrices” of fixed rank. The unequal sampling at different wavelengths is handled by using appropriate interpolation operators. The algorithm is operational for general array geometries (i.e., not restricted to, e.g., rectangular arrays) and for equidistantly as well as unequally spaced receivers. Numerical simulations for different sensor arrays and various signal-to-noise ratios are provided and demonstrate its excellent performance.

Mots clés: General domain Hankel matrix; Kronecker's theorem; Irregular sampling; ADMM; 3D DOA estimation

Domaines: Synthèse d'image et réalité virtuelle [cs.GR]; Traitement du signal et de l'image [eess.SP]; Traitement des images [eess.IV]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]

Fredrik Andersson, Marcus Carlsson, Jean-Yves Tourneret, Herwig Wendt. A Method for 3D Direction of Arrival Estimation for General Arrays Using Multiple Frequencies. IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP 2015), Dec 2015, Cancun, Mexico. pp. 1-4. ⟨hal-01511890⟩