• DocumentCode
    178285
  • Title

    DOA estimation performances of multi-parametric music in presence of modeling errors — Case of coherent multi-paths

  • Author

    Ferreol, Anne ; Delestre, Cyrile ; Larzabal, Pascal

  • Author_Institution
    Commun. et Security, Thales, Gennevilliers, France
  • fYear
    2014
  • fDate
    4-9 May 2014
  • Firstpage
    2247
  • Lastpage
    2251
  • Abstract
    The purpose of this paper is to give a closed form expression of the RMS (Root Mean Square) error of the estimated DOA (Direction Of Arrival) for multi-parametric MUSIC in presence of a modeling error. The multi-parametric MUSIC approach, [1] firstly introduced by [2] in polarization diversity context, estimates with a subspace approach the sources DOAs jointly to the nuisance parameters such as the polarization vector. The results are based on a second order approximation of the multi-parametric criterion with respect to modeling errors. DOA estimation errors is then an Hermitian form of multi-variate complex random variables. Theoretical results are validated by simulations in the context of coherent multi-paths in polarizations diversity.
  • Keywords
    Hermitian matrices; direction-of-arrival estimation; mean square error methods; multipath channels; signal classification; DOA estimation; Hermitian form; RMS error; closed form expression; coherent multipaths; direction of arrival estimation; modeling errors; multiparametric MUSIC approach; multivariate complex random variables; nuisance parameters; polarization diversity context; polarization vector; polarizations diversity; root mean square error; second order approximation; subspace approach; Context; Direction-of-arrival estimation; Estimation; Multiple signal classification; Noise; Signal processing algorithms; Vectors; DOA estimation; coherent paths; modeling error; performances; polarization;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
  • Conference_Location
    Florence
  • Type

    conf

  • DOI
    10.1109/ICASSP.2014.6853999
  • Filename
    6853999