Title :
Matrix pencil methods for ISAR image reconstruction
Author :
Baqai, Farhan A. ; Hua, Yingbo
Author_Institution :
Dept. of Electr. & Electron. Eng., Melbourne Univ., Parkville, Vic., Australia
Abstract :
The application of matrix pencil (MP) methods to inverse synthetic aperture radar (ISAR) imaging is considered. The noise sensitivity of the reported MP methods is studied. Numerical results that provide insights into the perturbation variances of the estimated 2-D frequencies are given. It is shown that the variant method (see Y. Hua, Asia-Pacific Microwave Conference, pp.691-694, 1992) can significantly reduce the computational burden and improve the estimation accuracy. Simulation results show that the variant method can provide estimates that are closer to the Cramer-Rao bound than those obtained by the original method (see Y. Hua, IEEE Trans, SP, vol.40, no.9, pp.2267-2280, Sept. 1992). Results of the original method for both synthesized and real ISAR data sets are given to demonstrate the robustness of MP methods to model errors.<>
Keywords :
computational complexity; image reconstruction; inverse problems; matrix algebra; perturbation techniques; sensitivity; synthetic aperture radar; variational techniques; ISAR image reconstruction; computational burden; estimation accuracy; inverse synthetic aperture radar; matrix pencil; noise sensitivity; perturbation variances;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1993. ICASSP-93., 1993 IEEE International Conference on
Conference_Location :
Minneapolis, MN, USA
Print_ISBN :
0-7803-7402-9
DOI :
10.1109/ICASSP.1993.319850