Robust Receivers
Abstract
We considered two SAR applications: robust detection in the presence of motion through resolution cells (MTRC); and clutter classification. To overcome the loss of detection performance due to MTRC, we account for the MTRC blurring in the reconstructed image domain rather than in the image formation stage. This leads to a much simpler detector than in current approaches. Our geometric approach avoids the costly step of undoing the MTRC blurring, a step that is as costly as phase corrections in traditional SAR methods. Performance analysis and experimental studies with synthetic data show that the wavelet-based detector is robust and exhibits better detection performance than other common detectors, e.g., the correlator or the energy detector. Regarding clutter classification, we designed an algorithm that best matches (in the sense of the Bhattacharyya coefficient) wavelet packet bases to a family of random processes. This family accounts for the variability in the clutter. We tested our robust classifier with polarimetric SAR data with good results.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 07, 2001
- Accession Number
- ADA391458
Entities
People
- Jose M. Moura
Organizations
- Carnegie Mellon University