An Improved Range-Doppler (R-D) Algorithm for Squint Synthetic Aperture Sonar (SAS) Imaging

Xuyan Wang,Chao Sun,Bing Liu
DOI: https://doi.org/10.3969/j.issn.1000-2758.2007.01.017
2007-01-01
Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University
Abstract:Aim. In our opinion, the R-D algorithm of Bao et al[3] for squint SAS imaging gives images that are not quite satisfactory and therefore needs to be improved if possible. We now offer an improved R-D algorithm. We found that the Fresnel assumption adopted by Ref. 3 is the cause of image quality not quite satisfactory; so, in our improved R-D algorithm, we abandon this assumption. In the full paper, we explain our improved R-D algorithm in detail; in this abstract, we just add some pertinent remarks to listing the two topics of explanation; (1) the geometric and mathematical model for squint SAS imaging and (2) improved R-D algorithm; in topic 2, we derive eqs.(3) through (15) in the full paper, which are needed for programming our improved R-D algorithm; also, in topic 2, the improved algorithm corrects range walk in azimuth time domain firstly and then corrects range curve in frequency domain; still, in topic 2, by virtue of geometric deformation revision, the image is almost indistinguishable from the original scene. Finally we performed simulations with both Ref. 3's and our improved algorithms, whose results are given in Fig.4(a) (our improved R-D algorithm) and Fig.4(b) (Ref. 3's algorithm). These results show preliminarily that the improved algorithm does enhance the quality of squint SAS imaging.
What problem does this paper attempt to address?