• DocumentCode
    290465
  • Title

    Further study of the SDMP method for parameter estimation of multiple transients

  • Author

    Evans, James B. ; Hua, Yingbo

  • Author_Institution
    Dept. of Electr. & Electron. Eng., Melbourne Univ., Parkville, Vic., Australia
  • Volume
    iv
  • fYear
    1994
  • fDate
    19-22 Apr 1994
  • Abstract
    A variant of the subspace decomposition matrix pencil (SDMP) method for estimating the angles of arrival of multiple transient signals is presented. The transient signals we consider here are modeled as the sum of complex exponentials. The SDMP-variant uses the same approach of the SDMP method to find the temporal poles. Then the SDMP-variant carries out a 1-D search for finding the spatial parameters (d.o.r´s). The advantages of the SDMP-variant are (a) that there is no pairing routine required to match the estimated d.o.r´s to the signal poles, (b) the noise sensitivity performance is better, and (c) the computation is more efficient. A sensitivity analysis is used to determine the performance, which is supported by simulation
  • Keywords
    direction-of-arrival estimation; exponential distribution; matrix decomposition; noise; pole assignment; transient analysis; SDMP method; SDMP-variant; angles of arrival; complex exponentials; computational efficiency; multiple transients; noise sensitivity performance; pairing routine; parameter estimation; simulation; spatial parameters; subspace decomposition matrix pencil method; temporal poles; transient signals; Amplitude estimation; Analytical models; Computational modeling; Delay effects; Delay estimation; Direction of arrival estimation; Matrix decomposition; Parameter estimation; Sensitivity analysis; Sensor arrays;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on
  • Conference_Location
    Adelaide, SA
  • ISSN
    1520-6149
  • Print_ISBN
    0-7803-1775-0
  • Type

    conf

  • DOI
    10.1109/ICASSP.1994.389833
  • Filename
    389833