Loading...

Signal Detection in Clutter Using Fractals in High Resolution Radars

Mohammad Pournejatian, Nima | 2012

690 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 44062 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Nayebi, Mohammad Mehdi
  7. Abstract:
  8. Accurate modeling of clutter and detection of low observable targets within clutter are the major problems in radar signal processing applications. Recently, fractal geometry is applied to the analysis of high range resolution radar sea clutters. The intrinsic difference in the fractal dimension between sea clutter and the echoes from targets is extracted and used to detect targets. The box-counting method is widely used to estimate fractal dimension but it has some drawbacks never or rarely considered in literature. We explain the drawbacks and propose a new fractal dimension based detector to increase detection performance in comparison with traditional detectors. Both statistically generated and real data samples are used to compare the performances of detectors. Irregular structures often carry essential information in a signal. In radar signals, interesting information also lies in sharp transitions. A wavelet transform focuses on localized signal structures with a zooming procedure that progressively reduces the scale parameter. Using fractal geometery and multiresolution analysis concepts together in designing a new detector, reveals considerable improvement in performance of target detection within clutter. In support of this argument, simulation results using real radar data samples are presented in this dissertation
  9. Keywords:
  10. Sea Clutter ; Wavelet Transform ; Fractal Geometry ; Multiresolution Analysis ; Target Detection ; Fractal Dimension

 Digital Object List

 Bookmark