This article reviews the main semiconductor-based modulators (SBMs) as a key technology enabling pulsed power systems in industrial and commercial applications (ICAs), spanning from food processing, medical use, accelerator research, manufacturing processes, agricultural practices, to environmental issues. This review covers modulator topologies ranging from generators with capacitive energy to inductive energy storage (IES) employing closing and opening switches, using switches only for pulse shaping, or including passive elements. The type of pulses produced, the switching technologies, the switch materials from Si, SiC to GaN, and the trigger techniques to switch many devices synchronously at different potentials. SBM offers a multiplicity of advantages, such as electrical efficiency, precise pulse control, fast switching, and high repetition rates, while its modular and scalable design supports flexible system configurations. Nevertheless, ICA has intrinsic characteristics and imposes several challenges, from load dynamics, current and voltage peaks, average power, pulse rise and fall times, to regulatory issues and operational issues. The strategies used to reach the best results in specific environments are addressed. Practical examples of SBM operation in ICA are presented. Also, future trends for new applications, materials, scaling up, thermal management, circuit drives, and component evolution are discussed.
Pulsed-field tumor ablation is evolving from fixed microsecond irreversible electroporation (IRE) toward high-frequency bipolar, nanosecond, burst, multilevel, and combined-waveform modalities. These treatments impose distinct requirements on voltage, current, pulsewidth, edge speed, polarity, timing accuracy, repetition rate, and adaptation to the dynamically changing electrode–tissue load. This review examines all-solid-state high-voltage pulse generators from a treatment-oriented perspective. Representative electrical ranges for IRE, high-frequency IRE (H-FIRE), nanosecond pulsed electric field (nsPEF), and combined waveforms are first linked to membrane response, neuromuscular stimulation, electrode polarization, thermal effects, and generator specifications. Solid-state Marx generators and multilevel generators, linear transformer drivers (LTDs), pulse-forming networks (PFNs) and pulse-forming lines (PFLs), Blumlein structures, inductive-energy-storage circuits, and DSRD/SOS opening-switch generators are then compared with emphasis on load-end waveform fidelity, voltage droop, fault current, arcing, and controllability under rapid impedance variation. Representative clinical systems and research prototypes are summarized, and the practical implications of silicon carbide (SiC) and gallium nitride (GaN) devices, isolated driving, electromagnetic interference (EMI)/electromagnetic compatibility (EMC), patient-side insulation, protection, and testing are discussed. The analysis indicates that medical readiness depends not only on maximum voltage or minimum pulse width, but also on reproducible load-end dose delivery, dynamic-load protection, measurement traceability, and coordinated safety control. Future development is likely to favor modular and programmable platforms integrating waveform synthesis, electrical feedback, fault protection, and treatment-data recording.
LuTan-1 (LT-1) was an interferometric SAR constellation of China, which was launched about four years ago. It was designed for 1:50 000 topographic mapping, geohazard monitoring, and image classification. In this article, we discuss the full lifecycle of LT-1, including requirements, satellite project systems, applications, and perspectives. First, the requirements of LT-1 are the inputs of the satellite project and provide the development parameters for the satellite system. Second, key technologies of the platform system, payload system, ground system, and application system have been proposed to ensure baseline and phase accuracy. After that, applications for China’s national products have been provided. The accuracy of the deformation products has been validated and demonstrated using images covering Shaanxi Province, showing the LT-1’s capability to ensure quantitative interferometric applications. Finally, the next-generation LT-1 has been proposed to achieve higher revisit time and deformation sensitivity.
Remote sensing (RS) for Earth observation (EO) is undergoing an unprecedented data explosion driven by high-resolution satellite missions, new trends such as the development of digital twins (DTs), open-access policies, and cloud-native data ecosystems. At the same time, the rapid evolution of deep learning (DL) and geospatial foundation models (GFMs), a class of large-scale DL models, is transforming the way EO data are represented, transferred, and exploited across tasks and domains. These developments impose demanding computational requirements across the full EO processing workflow, from data acquisition and transmission to storage, analysis, model training, and deployment, all of which increasingly rely on high-performance computing (HPC) capabilities. This survey reviews the intersection of HPC and EO, providing a structured overview of computing paradigms, infrastructures, and hardware accelerators that support modern EO workflows. Rather than proposing a prescriptive rule for selecting a specific computational solution, the survey relates the main stages of a typical EO workflow to the infrastructures commonly used to support them, including edge computing, ground-segment computing, cloud computing, and supercomputing environments. This workflow-oriented perspective highlights how the suitability of each infrastructure depends not only on computational performance but also on data locality, latency, scalability, interoperability, energy constraints, economic cost, and deployment requirements. The survey also analyzes the role of hardware accelerators, including graphics processing units (GPUs), field-programmable gate arrays (FPGAs), tensor processing units (TPUs), and other artificial intelligence (AI) accelerators, as well as emerging paradigms such as quantum and neuromorphic computing. Special attention is devoted to distributed training frameworks, model-sharding libraries, cloud-optimized data formats, workflow orchestration tools, and hardware-aware optimization strategies needed to support large-scale EO processing and GFM development. Finally, the survey identifies open challenges and future research directions toward scalable, sustainable, interoperable, and reproducible EO processing at the convergence of RS, DL, and HPC.
