The tremendous computing capacity of GPU offers significant potential in processing hierarchically compressed text data without decompression. However, current GPU techniques offer only traversal-based text data analytics; random access is exceedingly inefficient, limiting their utility significantly. To address this issue, we develop a novel and widely applicable solution that prompts random access to hierarchically compressed text data without decompression in GPU memory. We address three main challenges for enabling efficient random access to compressed text data on GPUs. The first challenge is designing GPU data structures that facilitate random access. The second challenge is efficiently generating data structures on GPU. The CPU is inefficient when generating data structures for random access, and this inefficiency increases considerably when PCIe transmission is incorporated. The third challenge is query processing on compressed text data in GPU memory. Random accesses, such as data updates, cause massive conflicts among countless threads. In order to address the first challenge, we develop several compressed GPU data structures, including indexing within the intricate GPU memory hierarchy. To handle the second challenge, we propose a two-phase process for producing these data structures on GPU. For the third challenge, a double-parsing design is proposed as a solution to avoid conflicts. We evaluate our solution on three platforms, two server-grade GPU platforms and one edge-grade GPU platform, using five real-world datasets. Experimental results show that random access operations on GPU achieve an average speedup of 52.98× compared to the state-of-the-art solution.

Motion errors are inevitably introduced when data is acquired and considerably degrade the image quality in terms of geometric resolution, radiometric accuracy and image contrast, especially in high‐resolution spotlight synthetic aperture radar (SAR) imagery. In this study, the authors present a novel two‐dimensional (2D) autofocus algorithm directly inserted into polar format algorithm, which compensates the envelop error and the phase error sequentially. A coarse error correction is first performed by global positioning system or inertial navigation system in the range‐compressed domain, then a new envelop compensation strategy, stage‐by‐stage approach, is designed, obtaining promising results for removing range cell migration after 2D interpolation. Additionally, a weighed contrast enhancement autofocus algorithm based on spatially variant model is developed to compensate for the residual phase error, which remarkably improves the estimation accuracy. The presented algorithm is very robust to deal with substantial errors over a variety of scenes even in conditions of homogenous areas with no prominent point scatterers and enables the utilisation of fast Fourier transform. The experimental results obtained by the proposed algorithm confirm that the analysis extends well to realistic situations.

An efficient and accurate inverse extended omega-K algorithm (IEOKA) is proposed to simulate synthetic aperture radar (SAR) raw data with trajectory deviations. Different from the traditional inverse omega-K algorithm that assumes an ideal flight trajectory, the IEOKA not only recovers the range cell migration accurately but also considers the motion errors including both range and phase errors due to the use of inverse extended Stolt interpolation. Furthermore, the azimuth dependence of the motion errors is discussed. A beam division method based on frequency division technique is presented to generate the azimuth-dependent phase error more accurately. The accuracy and effectiveness of the proposed algorithm have been verified using the generated SAR raw data consisting of the azimuth-dependent motion error.

For high-squint synthetic aperture radar (SAR) with curved trajectory, the traditional hyperbolic range model (HRM) is inaccurate, and the variation of velocity caused by constant acceleration in azimuth dimension cannot be ignored. Thus, the traditional omega-K imaging algorithms based on HRM are no longer available. For this problem, this letter proposes an improved range model, which can precisely take account for the curved trajectory. Then, a modified omega-K algorithm based on this new range model is proposed for high-squint SAR imaging. Simulation results validate the effectiveness of the improved range model and imaging algorithm.

In this letter, several important aspects are addressed for highly squinted synthetic aperture radar data with subaperture (SA) processing. First, the influences of the reference linear range walk correction and the resulting residual Doppler centroid are addressed. Then, the differences between the full-aperture data and SA data in the azimuth focusing are investigated in detail. Finally, a novel frequency-nonlinear-chirp-scaling algorithm with an addition of highly varying residual Doppler centroid correction over azimuth is proposed to equalize the azimuth-variant Doppler parameters so that the uniform azimuth processing can be realized. Compared with the previous works, this addition greatly improves the focused quality of the final image. Airborne real data processing validates the effectiveness of the proposed algorithm.

This letter proposes a new motion compensation approach for high-squint and high-resolution airborne spotlight synthetic aperture radar (SAR) integrated with polar format algorithm (PFA). The motion error is coarsely compensated by means of measuring systems. In PFA, the expressions of the line-of-sight polar interpolation are first explicitly revealed, and the nonsystematic range cell migration (NsRCM) induced by residual motion error is also corrected, which paves the way for autofocus application. After the NsRCM correction, the enhanced total least square estimator-based phase adjustment by contrast enhancement (PACE) algorithm is considered to account for the residual phase errors extended to the range-dependent case. The processing results of airborne SAR real data are presented to demonstrate the validity of the proposed algorithm.

The imagery of highly squinted synthetic aperture radar mounted on maneuvering platforms with nonlinear trajectory is a challenging task due to the existence of acceleration and the cross-range-dependent range migration and Doppler parameters. In order to accommodate these issues, a frequency-domain imaging algorithm based on tandem two-step nonlinear chirp scaling (TNCS) with small aperture is proposed. For the cross-range-dependent range cell migration (RCM) caused by the linear range walk correction and acceleration, the first-step NCS is introduced to suppress this dependence and realize the unified RCM correction. Based on the differences between full-aperture and small-aperture data in the cross-range processing, the second-step NCS is introduced in frequency domain to equalize the cross-range-dependent Doppler parameters, for cross-range processing is more sensitive to the cross-range dependence than range processing. Furthermore, a novel geometric correction method based on inverse projection is utilized to eliminate the negative effects caused by the imaging processing. Simulation results and real data processing are presented to validate the proposed approach.