Font Size: a A A

Study On HRWS SAR Imaging And The Algorithm Performance Improvement

Posted on:2016-08-27Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y X HouFull Text:PDF
GTID:1108330464462887Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
As an active spaceborne and airborne remote sensor, Synthetic Aperture Radar(SAR) is able to work in all-day and all-weather conditions. SAR becomes one of the most important tools in high resolution earth observation and global resource management, which has been widely used in civil and military fields. After near 40 years development, SAR imaging technology becomes more and more mature benefitting from the development of the digital signal processing technology, which acts as to be efficient, practical, robust and accurate.The wide swath and high resolution are the most important two goals of SAR imaging. Wide swath can cover a large observed scene to reduce the visiting times of the sensitive areas. High resolution is the most important indicator of radar to represent the characteristics and features of the observed target accurately. In general, the range resolution is determined by the bandwidth of the transmitted signal while the azimuth resolution depends on the length of antenna aperture in the azimuth direction. During these years, more and more practical applications require radar to have high resolution and wide swath(HRWS) abilities. The HRWS imaging becomes one of the hottest topics, especially in spaceborne SAR system, because the flight track of the Spaceborne SAR system is commonly a few hundred kilometers high away from the ground, a small beam can irradiate to a wide scene. However, due to the well-known "Minimum Antenna Constraint", Spaceborne SAR usually adopts a high PRF to ensure high azimuth resolution, which inevitably results in range ambiguity. Therefore, the study on effectively resolving the range ambiguity is crucial. In the case of HRWS imaging, the whole width of the swath equals to the sum of all the sub-swath width. The width of each sub-swath is generally wider than the normal scene to realize a swath as wide as possible when the total swath number is determined. However, the imaging performance using the conventional algorithms tends to degrade in this case. In essence, SAR imaging processing can be considered as a decoupling operation in the domain of the two-dimensional spectrum; the coupling term will exhibit a serious spatial-variance in HRWS imaging, which degrades the focusing performance. In this dissertation, we propose two HRWS imaging algorithms respectively for single channel and multi-channel SAR systems. We also present three scaling algorithms based on Chebyshev polynomial to improve the performance of the sub-swath imaging, i.e. Chirp Scaling Algorithm(CSA), Frequency Scaling Algorithm(FSA) and Nonlinear Chirp Scaling Algorithm(NCSA).This dissertation studies the performance improvement of HRWS imaging and SAR algorithms. The main works are summarized as follows:1. In single-channel SAR system, a novel phase encoding algorithm is proposed for HRWS imaging. By introducing an initially modulated phase, the proposed algorithm is executed by solving linear equations, which is able to resolve the range ambiguity to achieve good HRWS imaging results. Compared with the multi-antenna method, the proposed algorithm has the characteristics of low complexity and easy realization.2. A HRWS imaging algorithm based on Compressive Sensing(CS) technology is proposed. CS technology can recover a sparse signal from fewer samples at a high probability,and the imaging result has the characteristics of low sidelobes and high resolution. The proposed algorithm is designed for the MIMO-SAR system, where the stepped frequency LFM signal is used to ensure the orthogonally of the transmitted signal. The CS technique is employed to solve the range ambiguity problem. Meanwhile, the range compression and migration correction can be accurately realized. The phase history can be well preserved to realize azimuth compression. If the SAR system works in low PRF, it should do a samples reconstitution in the azimuth direction before resolving the range ambiguity problem.3. New versions of Chirp Scaling Algorithm and Frequency Scaling Algorithm are proposed. For wide swath imaging, the performance of traditional Chirp Scaling and Frequency Scaling imaging algorithms is poor when the targets are far away from the reference center point. Accordingly, we propose novel algorithms based on Chebyshev polynomial instead of Taylor series. The new algorithms can obtain a more accurate two-dimensional spectrum so that a better focusing performance can be achieved. The scaling functions of the proposed algorithms are both abstractly exhibited from the optic system. In the CSA, the range cell migration correction is done by the scaling operation, which is more compact than traditional algorithms. Most importantly, the approximation error of Chebyshev polynomial is bounded to restrict the error to be not particularly large in the edge-located points, namely increasing the focusing depth of the scene.4. In highly squinted angle imaging, the focusing depth is degraded due to the serious coupling term between range and azimuth dimensions. The original Nonlinear Chirp Scaling Algorithm improves the focusing quality in a certainextent using a cubic phase filter. However, the higher the squint angle and the wider the scene, the worse the focusing quality. In this chapter, NCSA is combined with the Chebyshev polynomial, which can improve the focusing quality of the edge-located points. The suitable squint angle range of the new algorithm is increased almost 10 degrees.
Keywords/Search Tags:high resolution wide swath(HRWS), range ambiguity, compressed sensing(CS), multiple-input multiple-output(MIMO) radar, Chebyshev polynomial
PDF Full Text Request
Related items