As the world’s first geosynchronous (GEO) orbit synthetic aperture radar (SAR) satellite, LuTan-4 was successfully launched on 13 August 2023. It was developed by the China Academy of Space Technology, with which the authors are affiliated. This study presents a comprehensive review of the recent advancements in GEO SAR technology. The review first begins by summarizing key considerations in GEO SAR system design, including orbital parameters and synthetic aperture time, transmit power and antenna aperture, two-dimensional beam-steering, imaging parameters and non-ideal factors. In terms of GEO SAR signal processing, the article focuses on two fundamental models, i.e., the high-order slant-range model and the coupled space-variant signal model. It also introduces the current GEO SAR imaging algorithms. Furthermore, the study presents an analysis between GEO SAR and low-Earth orbit SAR systems, highlighting the superior capability of GEO SAR for large-scale surface dynamic processes. Finally, the paper outlines the future development directions and potential applications of GEO SAR technology.

In recent years, Geosynchronous synthetic aperture radar (GEO SAR) has attracted many scientists to carry out relevant research. Its research is in-depth and gradually maturing. However, conventional GEO SAR has the problems of great influence by the earth's rotation, complex timing design, uneven flight speed in the whole orbit, and non orthogonal range and azimuth resolution, which limits it's application. A new concept of sub retrograde geosynchronous orbit SAR is proposed in this paper. Compared with traditional GEO SAR, the satellite has the advantages of less influence by the earth's rotation, simple timing, uniform motion speed in the whole orbit, shorter revisit time, better resolution and small space-variant in azimuth and range. The radar parameters of sub retrograde GEO SAR are designed. Through the research, it is found that the system sensitivity and resolution no longer change with the orbit position, and its timing design does not need to consider the range migration variation. In addition, the imaging characteristics of sub retrograde GEOSAR are investigated. The results show that, the range model can meet the imaging requirements as long as it is expanded to the fourth order. After that, the influence of orbit perturbation error in sub retrograde GEO SAR is given for the first time. Last, two ways to solve observation blind area of sub retrograde GEO SAR are proposed, one is to change the orbit inclination, another way is to use the bistatic mode.

Millimeter-wave radars are widely used in advanced driver assistance systems (ADAS) and autonomous driving. The suppression of mutual interference is challenging for automotive radars to address. The traditional sparse-based methods are effective in interference suppression but have a high computational overhead. However, automotive radars are sensitive to computational overhead. To tackle the problem, this paper proposes a fast convergence method based on the fused Lasso. Taking advantage of the pulse-like shape of the mutual interference in the time domain, a fused Lasso based sparse model is proposed to extract and eliminate the mutual interference. Experiments on simulated and real radar data are conducted to verify the effectiveness of the proposed method, and experiments in dynamic scenarios are conducted. In the two-dimensional range-Doppler (RD) map, the increased signal to interference plus noise ratio (SINR) of the proposed method is about 11.95 dB. In addition, the time cost of the proposed method is much lower than the traditional methods. The robustness of the proposed method is also verified via experiments with different chirp rates and pulse durations.

Two long-standing problems exist in the high-frequency electromagnetic scattering analysis: 1) Full-wave methods suffer a geometric increase in computational costs, along with the rising frequencies. 2) Ray tracing methods struggle to acquire sufficient phase information, when dealing with complex targets. To address these two issues, we propose a phase-informed hybridizable discontinuous Galerkin (PI-HDG) method for extremely high-frequency modeling. Similar to the physics-informed machine learning algorithms, we integrate the priori knowledge of phase information to the basis-function construction in the PI-HDG algorithm, thus combing both strengths from HDG method and geometric optics ansatz. Numerical experiments show that the PI-HDG algorithm can accurately compute high-frequency scattering characteristics of complex targets. Specifically, in a multiple scattering scenario, we achieve only 1/10 temporal consumption while 1/16 for the memory, when compared with the reference finite element software.

Automotive radar is crucial in advanced driver assistance systems (ADAS) and autonomous driving. However, radars encounter considerable challenges when other radars operate within a similar frequency band and generate mutual interference. The radar signal imputation network (RadarImputationNet) is proposed to suppress interference signals by accurately reconstructing missing or corrupted radar data. The long-term dependency enhancement module (LTDEM) is proposed to improve the precision of signal imputation. The proposed RadarImputationNet is used after interfering samples are removed from the pulse. The recent diffusion models have poor signal imputation accuracy and performance in recovering signal peaks. The proposed RadarImputation- Net model solves the dilemma by improving the long-term dependency of the diffusion model. A real-data experiment is conducted in the dynamic scene to verify the proposed RadarImputationNet. Metrics including root-mean-square error (RMSE), mean absolute error (MAE), and continuous ranked probability score (CRPS) are compared, and the proposed RadarImputationNet outperforms existing methods in RMSE, MAE, and CRPS. RMSE decreased by 19.30, representing a 50.7% reduction; MAE decreased by 13.91, representing a 48.6% reduction; and CRPS decreased by 0.17, representing a 48.3% reduction, indicating a significant improvement in the performance of the diffusion model. Furthermore, the proposed RadarImputationNet demonstrates exceptional proficiency in restoring the peak characteristics of the missing samples.