Polarimetric weather radar measurements vary nonlinearly with changes in radar frequency and scanning elevation. In comparative observation experiments between Ku-band weather radar and CINRAD/SA radar, there were systematic errors in the polarimetric data of the two radars, the reason was the difference of frequency and elevation angles of them. Because the observation data of the ground-based Ku-band weather radar is small, it cannot cover most of the precipitation. So the T-matrix method was used to simulate the differential reflectivity factors of the Ku-band and S-band radars at different elevation angles, and the BP neural network was trained based on the simulation data to realize the conversion of the differential reflectivity factor from S-band to Ku-band to correct the systematic errors. Using the measured data, it was verified that the BP neural network can correct the systematic errors between the differential reflectivity factors of the two radars, improve the data consistency of them, and provide possibility for data fusion of S-band and Ku-band weather radars.

Computerized Ionospheric Tomography (CIT) serves as a crucial method for ionospheric monitoring, playing a significant role in space environment surveillance and earthquake prediction. Existing CIT techniques are mostly based on the Global Navigation Satellite System (GNSS) system, constrained by the distribution of receivers. CIT based on geosynchronous SAR (GEO SAR) presents a solution by leveraging Persistent Scatterer (PS) points within the scene. However, current CIT techniques using GEO SAR typically utilize PS points from a single SAR image. This paper introduces a novel approach – a GEO SAR-based long-term joint multi-image CIT method. This method enhances the exploitation of satellite data, offers a broader range of observation angles, and improves tomography accuracy. Finally, through a CIT experiment involving three GEO SAR images, the electron density distribution is derived with a time resolution of 10 minutes. The results align with the International Reference Ionosphere (IRI) data, validating the feasibility and advantages of the proposed method. It is noteworthy that the proposed method caters to the tomography observation mode of a single satellite and is adaptable to each satellite within a satellite formation.

Designing repeat ground track (RGT) synthetic aperture radar (SAR) constellations for achieving rapid revisits over key areas is essential to employ spaceborne differential interferometric synthetic aperture radar (D-InSAR) technology in Earth observation missions such as geological disaster monitoring and prediction. In this paper, the features of average revisit time (ART) maps are first introduced and investigated, and then an efficient and resource-friendly approach to calculate the ART of constellations is proposed. On this basis, a systematic method for designing an RGT constellation is provided, incorporating lookup-table-based optimization. Once the requirements of the expected RGT constellation, the incident angle of sensors on the constellation, and the orbital elements of the seed satellite in the constellation are given, the range of the optimal inclination and longitude of the ascending node (LAN) of the seed satellite can be found and then the entire constellation is determined. The proposed method enhances the efficiency of revisit time analysis and avoids the repeated modeling when the observation requirements change. Therefore, it is applicable not only prior to launch but also guides orbital maneuvering to adjust constellation configuration for an effective response to sudden disasters, etc. Finally, multiple RGT constellation design tasks are presented to demonstrate the proposed method.

Recently, frequency-modulated continuous-wave (FMCW) radar-based hand gesture recognition (HGR) using deep learning has achieved favorable performance. However, many existing methods use extracted features separately, i.e., using one of the range, Doppler, azimuth, or elevation angle information, or a combination of any two, to train convolutional neural networks (CNNs), which ignore the interrelation among the 5-D time-varying-range-Doppler-azimuth-elevation feature space. Although there have been methods using the 5-D information, their mining of the interrelation among the 5-D feature space is not sufficient, and there is still room for improvements. This article proposes a new processing scheme of HGR based on 5-D feature cubes that are jointly encoded by a 3-D fast Fourier transform (3-D-FFT)-based method. Then, a CNN is proposed by building two novel blocks, i.e., the spatiotemporal deformable convolution (STDC) block and the adaptive spatiotemporal context-aware convolution (ASTCAC) block. Concretely, STDC is designed to cope with hand gestures’ large spatiotemporal geometric transformations in the 5-D feature space. Moreover, ASTCAC is designed for modeling long-distance global relationships, e.g., relationships between pixels of the feature at the upper left corner and lower right corner, and exploring the global spatiotemporal context, in order to enhance the target feature representation and suppress interference. Finally, our presented method is verified on a large radar dataset, including 19 760 sets of 16 common hand gestures, collected by 19 subjects. Our method obtains a recognition rate of 99.53% on the validation dataset and that of 97.22% on the test dataset, which is significantly better than state-of-the-art methods.

Geosynchronous Synthetic Aperture Radar (GEO SAR) has advantages of a short revisit time and large coverage for the scene of interest, so lots of theories and analysis toward the GEO SAR have been developed. However, GEO SAR systems may generate radio frequency interference (RFI) to a low earth orbit SAR (LEO SAR), causing a decrease in Signal-to-Interference-plus-Noise Ratio (SINR) of SAR images. In order to evaluate the GEO-to-LEO RFI effect on imaging, we deduce the formulas of the RFI power and image SINR, and verify them by comparing them with numerically evaluated results from simulated images. Based on the formulas, we evaluate SINRs of LEO SAR images for different bistatic scattering coefficients. The results show that when the target forms a specular bistatic scattering geometric relationship with a GEO SAR and a LEO SAR, LEO SAR image quality is poor, with a SINR worse than 5 dB, but the RFI effects can be neglected in other cases.