SAR Imaging via Efficient Implementations of Sparse ML Approaches

George-Othan Glentis, Kexin Zhao, Andreas Jakobsson, Habti Abeida, Jian Li

Research output: Contribution to journalArticlepeer-review

18 Citations (SciVal)
631 Downloads (Pure)


High-resolution spectral estimation techniques are of notable interest for synthetic aperture radar (SAR) imaging. Several sparse estimation techniques have been shown to provide significant performance gains as compared to conventional approaches. We consider efficient implementation of the recent iterative sparse maximum likelihood-based approaches (SMLAs). Furthermore, we present approximative fast SMLA formulation using the Quasi-Newton approach, as well as consider hybrid SMLA-MAP algorithms. The effectiveness of the discussed techniques is illustrated using numerical and experimental examples.
Original languageEnglish
Pages (from-to)15-26
JournalSignal Processing
Issue numberFebruary
Publication statusPublished - 2014

Subject classification (UKÄ)

  • Probability Theory and Statistics


  • Synthetic aperture radar imaging
  • Non-parametric high resolution spectral analysis
  • Sparse estimators
  • Efficient algorithms


Dive into the research topics of 'SAR Imaging via Efficient Implementations of Sparse ML Approaches'. Together they form a unique fingerprint.

Cite this