Loading...

Statistical Modeling of Spatial and Temporal Characteristics of Target Range Profiles for Radar Target Recognition

Ajorloo, Abdollah | 2013

1034 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45175 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Bastani, Mohammad Hassan
  7. Abstract:
  8. Range Profile (RP) is known as one of the most important tools for radar target recognition. The main problem with range profile for radar target recognition is its sensitivity to aspect angle. To overcome this problem, range profiles are assumed to have the same statistical characteristics in small frames of aspect angles or context dependent models such as HMM may be used. All methods presented, seems to have some shortcomings to offer a model based on physical circumstances of target maneuver. For example in such models, consecutive samples of RPs in an aspect frame are assumed to be statistically independent with the same distribution (IID). Here we propose dynamic system (DS) for modeling the spatiotemporal characteristics of range profiles. Spatial characteristics refer to statistical characteristics of RPs in specific aspect angles, and temporal ones refer to variations of RPs along the maneuver due to aspect changes. PCA is used for feature extraction and based on it, we have presented PCA+DS model which outperforms previous models according to simulation results. For evaluating the quality of modeling, Akaike Information Criterion (AIC) is used. We have also applied DS for spectral features extracted from the range profiles and favorable results were achieved, too. For recognition, three fighters including F-15, MIG-21, and Tornado are considered and we have proposed a method based on PCA+DS and also a method based on the model we called it FA+DS and compared them with the conventional FA method. The total recognition rates for PCA+DS, FA+DS, and FA are 94.9, 91.16, and 90.47 percent respectively
  9. Keywords:
  10. Akaieke Information Criterion ; Dynamical Systems ; Principal Component Analysis (PCA) ; Radar Target Recognition ; Range Profile Function

 Digital Object List

 Bookmark

...see more