The LuTan-1 satellite is the first Chinese, L-band, distributed, spaceborne interferometric synthetic aperture radar (InSAR) mission. However, the presence of radio frequency interference (RFI) in the L-band poses a significant threat to obtaining a high-quality digital elevation model (DEM) and deformation monitoring. This paper provides a first investigation and assessment of the RFI issues in the operational LuTan-1 InSAR system. The RFI environments are analyzed from the status of frequency allocation. The mathematical model of interference in InSAR image pairs is derived and discussed the variation of interferometry coherence under different imaging modes. Furthermore, this paper proposes an automatic processing pipeline of RFI detection and mitigation for the LuTan-1 ground processing system, which is efficient for dealing with massive images without tuning hyperparameters. Extensive experimental results on diverse scenes in LuTan-1 real measured data with different RFI cases are provided, including the single-pass, repeat-pass, and full polarization modes. Experimental results verify that the proposed detection and mitigation scheme could effectively eliminate the RFI artifacts, enhance the image quality, and improve the interferometric coherence. The proposed RFI detection and mitigation scheme has been successfully incorporated into the LuTan-1 ground processing pipeline.

The inherent unknown deformations of inverse synthetic aperture radar (ISAR) images, such as translation, scaling, and rotation, pose great challenges to space target classification. To achieve high-precision classification for ISAR images, a deformation-robust ISAR image classification network using contrastive learning (CL), i.e., CLISAR-Net, is proposed for deformation ISAR image classification. Unlike traditional supervised learning methods, CLISAR-Net develops a new unsupervised pretraining phase, which means that the method uses a two-phase training strategy to achieve classification. In the unsupervised pretraining phase, combined with data augmentation, positive and negative sample pairs are constructed using unlabeled ISAR images, and then the encoder is trained to learn discriminative deep representations of deformation ISAR images by means of CL. In the fine-tuning phase, based on the deep representations obtained from pretraining, a classifier is fine-tuned using a small number of labeled ISAR images, and finally, the deformation ISAR image classification is realized. In the experimental analysis, CLISAR-Net achieves higher classification accuracy than supervised learning methods for unknown scaled, rotated, and combined deformations. It implies that CLISAR-Net learned more robust deep features of deformation ISAR images through CL, which ensures the performance of the subsequent classification.

Several glaciers in the Bukatage Massif are surge-type. However, previous studies in this region focused on glacier area and length changes, and more information is needed to support the deep analysis of glacier surge. We determined changes in glacier thickness, motion, and surface features in this region based on TanDEM-X, ALOS/PRISM, Sentinel-1A, and Landsat images. Our results indicated that the recent surge of the Monomah Glacier, the largest glacier in the Bukatage Massif, started in early 2009 and ceased in late 2016. From 2009 to 2016, its area and length respectively increased by 6.27 km² and 1.45 km, and its ice tongue experienced three periods of changes: side broadening (2009–2010), rapid advancing (2010–2013), and slow expansion (2013–2016). During 2000–2012, its accumulation zone was thinned by 50 m, while its ice tongue was thickened by 90 m. During 2015–2017, its flow velocity reduced from 1.2 to 0.25 m/d, and the summer velocities were much higher than winter velocities. We conclude that the recent Monomah Glacier surge is thermal-controlled. The subglacial temperature rose to the pressure-melting point because of substantial mass accumulation, and then the increased basal meltwater caused the surge.

As a wideband radar system, synthetic aperture radar (SAR) may conflict with several electromagnetic systems. These signals may severely interfere SAR image quality. Numerous previous researches focused on the interference suppression problem, among which semi-parametric methods, such as low-rank recovery methods, have been verified to have the state-of-the-art performance. However, semi-parametric methods are restricted by extremely strong interferences when the signal-to-interference-and-noise ratio (SINR) exceeds the ability upper bound of semi-parametric methods. In recent years, multi-channel SAR (MC-SAR) systems are widely used for more applications, where multiple antennas are mounted along the azimuth or in elevation. Adaptive digital beamforming (DBF) is a classic spatial filtering method to focus energy at the expected direction and suppress unexpected interferences. Its performance is determined by the array manifold and the interference impinging angle. In this paper, we propose a novel space-time-combined method which takes the advantages of both low-rank recovery methods in two-dimensional (2-D) time domain and adaptive DBF method in spatial domain. Specifically, we construct a single optimization problem to unify both kinds of methods. The alternating direction of multiple multiplier (ADMM) framework is leveraged with closed-form solution for each step. Multiple experiments are provided to demonstrate the effectiveness of the proposed method.