Please use this identifier to cite or link to this item:
Title: Sparse array 3-D ISAR imaging based on maximum likelihood estimation and CLEAN technique
Authors: Ma, C. 
Yeo, T.S. 
Tan, C.S.
Tan, H.S.
Keywords: 3-D imaging
inverse synthetic aperture radar (ISAR)
maximum likelihood estimation
sparse antenna array
steepest descent
Issue Date: Aug-2010
Citation: Ma, C., Yeo, T.S., Tan, C.S., Tan, H.S. (2010-08). Sparse array 3-D ISAR imaging based on maximum likelihood estimation and CLEAN technique. IEEE Transactions on Image Processing 19 (8) : 2127-2142. ScholarBank@NUS Repository.
Abstract: Large 2-D sparse array provides high angular resolution microwave images but artifacts are also induced by the high sidelobes of the beam pattern, thus, limiting its dynamic range. CLEAN technique has been used in the literature to extract strong scatterers for use in subsequent signal cancelation (artifacts removal). However, the performance of DFT parameters estimation based CLEAN algorithm for the estimation of the signal amplitudes is known to be poor, and this affects the signal cancelation. In this paper, DFT is used only to provide the initial estimates, and the maximum likelihood parameters estimation method with steepest descent implementation is then used to improve the precision of the calculated scatterers positions and amplitudes. Time domain information is also used to reduce the sidelobe levels. As a result, clear, artifact-free images could be obtained. The effects of multiple reflections and rotation speed estimation error are also discussed. The proposed method has been verified using numerical simulations and it has been shown to be effective. © 2006 IEEE.
Source Title: IEEE Transactions on Image Processing
ISSN: 10577149
DOI: 10.1109/TIP.2010.2045711
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Jun 24, 2022


checked on Jun 24, 2022

Page view(s)

checked on Jun 23, 2022

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.