DOA Estimation of Multiple Convolutively Mixed Sources Based on Principle Component Analysis

  • Authors:

  • Weidong Jiao;Shixi Yang;Yongping Chang

  • Affiliations:

  • Dept. of Mechanical Engineering, Jiaxing University, Jiaxing, China 314001 and Dept. of Mechanical Engineering and Automation, ZheJiang University, Hangzhou, China 310027;Dept. of Mechanical Engineering and Automation, ZheJiang University, Hangzhou, China 310027;Dept. of Mechanical Engineering, Jiaxing University, Jiaxing, China 314001

  • Venue:
  • ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part I
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Direction of arrival (DOA) estimation is a basic task in array signal processing. A method based on principal component analysis (PCA) is presented for estimating DOA of multiple sources mixed convolutively. Convolutive mixtures of multiple sources in the spatio-temporal domain are firstly reduced to instantaneous mixtures by using the well-known short-time Fourier transformation (STFT) technique. From the time-frequency mixture in each frequency bin, one frequency respond matrix of the mixing system from sources to sensors is estimated by the PCA based whitening. Furthermore, the DOAs of multiple sources are probed by using a whole estimating strategy. Consequently, all mixtures in total frequency bins contribute to a final estimation set, in which the source directions are shown as several direction clusters and/or local maxima. Experimental results indicate that the PCA based method has advantages over the well-known MUSIC (MUltiple SIgnal Classification) method, especially under such conditions as the same number of sensors as sources, and closely placed sensors.