Loading...

An improved GraDe method for blind separation of graph signals

Sadeghi, M ; Sharif University of Technology | 2023

0 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/TSP.2023.3331264
  3. Publisher: Institute of Electrical and Electronics Engineers Inc , 2023
  4. Abstract:
  5. For blind source separation (BSS) of Gaussian graph signals, an algorithm called GraDe (graph decorrelation) has been previously introduced. In the current paper, it is shown that GraDe does not achieve a good performance for some types of graphs. This is attributed to the estimation of covariance/autocovariance matrices using signal samples, which may not be reliable. To address this weakness, an improvement based on the spectral representation of the signals is proposed, focusing on removing the impact of the outlier eigenvalues. Numerical simulations show that the proposed method outperforms the original GraDe algorithm. © 1991-2012 IEEE
  6. Keywords:
  7. Blind source separation ; Graph decorrelation ; Graph signal processing
  8. Source: IEEE Transactions on Signal Processing ; Volume 71 , 2023 , Pages 4382-4391 ; 1053587X (ISSN)
  9. URL: https://ieeexplore.ieee.org/document/10325653