The D-band, providing large available bandwidth in the range from 110 to 170 GHz, can potentially enable wireless communication systems satisfying the ever-growing demand for high data rates, as well as high-resolution sensing applications. The radio channel significantly influences the performance of such systems and must therefore be properly accounted for, making measurement campaigns indispensable for obtaining direct insight into target environments, deriving appropriate channel models, and validating simulation environments. In this work, an extension to an existing channel sounder is presented that enables measurements around 155 GHz, along with the first measurement campaign conducted with this extension in a medical environment, representing a potential use-case environment. Based on the azimuth-sampled double-directional measurements, the environment is analyzed at the path level, statistical channel parameters are derived, and the results reveal multipath-rich propagation conditions. The obtained insights support the development and evaluation of future high-data-rate communication and high-resolution sensing systems.
Terahertz (THz) detection using probe lasers around 1 µm wavelength is becoming increasingly relevant due to the growing adoption of high-power Yb-based femtosecond (fs) laser systems. However, systematic benchmarking of detection architectures in this wavelength regime remains limited. In this study, we benchmark THz detection using CdTe electro-optic sampling (EOS) and LT-InGaAs bowtie photoconductive antenna (PCA) under 1.06 µm Yb-laser excitation. PCA detection is governed by the antenna geometry and the equivalent electronic RC circuit, yielding a resonant peak at 0.35 THz and a relatively narrow bandwidth of 1 THz. In contrast, CdTe EOS captures faster field oscillations, providing broader spectral coverage (0.3 to 3 THz), with a peak at 0.45 THz. While the bowtie PCA exhibits enhanced low-frequency sensitivity and, when combined with a silicon lens, yields a large dynamic range of 65 dB, the performance of CdTe EOS is primarily limited by phase-matching constraints, finite crystal thickness, and strong absorption in the crystal, resulting in a measured dynamic range of 45 dB. Furthermore, the experimentally measured PCA response is interpreted using a photocurrent impulse response model incorporating the intrinsic photocurrent temporal response, the resonant PCA transfer function, and a parasitic 3-dB RC attenuation factor. These results establish practical detector-selection benchmarks for Yb-based THz systems while highlighting, under the specific conditions of this study, the trade-off between the broader bandwidth of CdTe EO sampling and the low-frequency, high-sensitivity, and compact detection enabled by the LT-InGaAs bowtie PCAs.
An E-band two-stage power amplifier (PA) in 22-nm bulk CMOS technology is presented. The proposed PA employs a gain-boosting unit to enhance small-signal gain without additional amplifier stages and uses a two-way transformer-based power combiner to improve output power and efficiency. To ensure stability under high-frequency operation, the neutralized common-source topology and stability factor verification are adopted. The measured gain of PA is above 13.7 dB within 70–80 GHz frequency range, and reaches a peak value of 17.5 dB at 71 GHz. And with a 5-dBm input power, the PA exhibits output power (Pout) higher than 14.4 dBm and power-added efficiency (PAE) above 11.9
Very long baseline interferometry (VLBI) is a powerful technique for detailed observations of the central engine of active galactic nuclei, galactic dynamics, and massive star-forming regions. Converting from circular to linear polarization is necessary for VLBI observation since several receivers use the latter. However, the existing methods for achieving this remain limited. This study addressed this limitation by designing, fabricating, and testing a low-insertion loss, low-reflection coefficient, and low-cross-polarization broadband circular polarizer (CP) for the 70–116 GHz band. The proposed CP is designed to achieve low-insertion loss, a low-reflection coefficient, and low cross-polarization. It consists of a 90^∘ differential phase shifter (PS), which combines three phase-shifting sections to provide the required differential phase over a wide frequency range, and a double-ridged orthomode transducer (OMT), which offers low-loss characteristics and ease of fabrication. By precisely controlling the phase difference and amplitude error of the PS and implementing structural improvements to suppress air gaps at the contact interfaces, low-loss and low cross-polarization performance was achieved over a wide frequency range. A prototype CP was fabricated and tested, showing good agreement with the simulation results. Back-to-back CP measurements demonstrated an insertion loss less than 0.8 dB, a reflection coefficient below -20 dB, and a cross-polarization below -21 dB. These results demonstrate the broadband performance of the proposed CP over the 70–116 GHz band.
In the paper for the first time a high-power vortex Bessel beam with a topological charge |l|= 5 have been formed using the phase reflective element and a gyrotron radiation at a frequency of 0.263 THz (wavelength 1.14 mm). The reflective element was the spiral phase plate manufactured by micromilling on an aluminum plate. In the experiment, the energy efficiency of converting the beam's Gaussian mode to a Bessel mode was approximately 80
In this manuscript, a four-port MIMO antenna is designed for millimeter wave (mmWave) operation at 24/38 GHz. The proposed MIMO antenna has small footprint of 16.0 mm × 16.0 mm × 0.25 mm (1.28 λ_0 × 1.28 λ_0 × 0.02 λ_0 , where λ_0 is wavelength and calculated at lower resonance frequency 24 GHz) using the flexible substrate. The dual-band characteristic is achieved using H-shaped slot in the triangular radiating patch. The orthogonal placement of the patch and defected ground plane using rectangular slot is implemented to achieve the improved isolation > 22 dB across both the operating bands. The antenna demonstrates the gain more than 5.47 dBi and efficiency > 75
Joining and welding techniques of similar and dissimilar materials have reached their limits due to the inflexibility of adapting dielectric materials, e.g., glasses and ceramics, reproducibility for transparent materials, and cycle process time. The ability to join pieces of glasses using laser irradiation allows rapid processing of vacuum insulating glass (VIG) for a widening range of commercial energy-efficient applications. We have used a 1064-nm laser wavelength with 15 picoseconds pulse width and a 155-kHz repetition rate to weld commercial plate glass pieces. The welded samples were characterized via x-ray fluorescence (XRF), time-of-flight secondary ion mass spectrometry (TOF–SIMS), polarimetry, and terahertz time-domain spectroscopy (THz-TDS) to determine the integrity of the weld and chemical changes in the welded areas. Our data shows no significant changes in glass chemistry and structure on average over nano- to micro-scales occurred from the welding process within the limits of the characterization tools. Our results demonstrate successful welding of glass plates without significant changes in the glass chemistry, structure, or properties in the laser-modified region or strains in the bulk glass.
Terahertz time-domain spectroscopy (THz-TDS) can underestimate the Q factor of high-Q metasurfaces fabricated on optically thick substrates, because time-windowing suppresses Fabry–Pérot (FP) echoes at the cost of spectral resolution. We present a quartz-calibrated THz-TDS workflow for echo-consistent comparison between experiment and full-wave simulation. Instead of removing substrate echoes from the measured waveform, we incorporate a calibrated finite-thickness quartz substrate into the simulation, so that ungated experimental spectra and simulated spectra are evaluated under matched physical conditions. The workflow is validated using bare quartz substrates, a MoS2-on-quartz sample, and multiple metasurfaces. For the thin-film case, the extracted parameters are fed back into full-wave simulations, and the resulting spectra reproduce both the transmission magnitude and the FP interference features. These results support the internal consistency of the analytical thin-film treatment and indicate that the extracted parameters capture the main spectral features under the present measurement conditions. For the metasurface case, a systematic gate-end-time analysis shows that the extracted Q factor is window-dependent for narrow q-BIC resonances, and the ungated echo-consistent spectra yield Q factors that agree more closely with the simulations.
We present a detection method and measurements of low-friction formations on road surfaces using a monostatic polarimetric radar installed on a vehicle in motion. Polarimetric parameters (PP) such as entropy, depolarization, and the proportion of surface-scattering are used to identify the surface and classify it as dry, wet, or icy. The polarimetric parameters are calculated from the eigenvalues (EV) of the measured covariance/coherence matrix. Due to the low contrast in the polarimetric parameters between icy and dry surfaces, the classification of the surface becomes challenging in the presence of irregularities on the surface that can be interpreted as ice. To minimize the false alarms, in addition to the PP, we use the sum of the non-normalized eigenvalues, which is a measure of the total scattered power in all polarization components. The sum of the EV has better contrast compared to the PP and helps in reducing the negative effect of natural surface irregularities. To classify the surface, we calculate the adaptive mean value and variance of the PP and non-normalized EV and compare each measurement to a threshold. The surface is classified after a certain combination of parameters exceeds their threshold value. We demonstrate how each of the parameters is affected by an icy/wet surface compared to the same surface in dry conditions. In our measurements, we are able to detect ice patches with a false alarm below 2
This work introduces a compact four-element Trident-Type Slotted Monopole (TTSM) MIMO antenna developed for millimeter-wave super ultra-wideband (SUWB) communication systems. The antenna is realized on a Rogers RT/Duroid 5880 substrate and utilizes an orthogonal arrangement of radiating elements combined with a shared ground plane, where the interelement spacing is optimized to 8.5 mm to maintain compact dimensions of 24 × 24 × 0.787 mm3, while controlling coupling between ports. To reduce mutual coupling further, decoupling structures in the form of SRRs are placed between each element. The optimized SRR-2 configuration integrated with grounded stubs forms an effective isolation network that restrains surface current propagation and reduces coupling while preserving stable radiation behavior. The developed antenna exhibits a broad operating bandwidth from 20.10 to 49.55 GHz (|S₁₁|< -10 dB), along with inter-port isolation greater than 25 dB and peak isolation close to 30 dB. Radiation efficiency is greater than 94
This work presents an upgraded U-shaped beam tunnel sawtooth waveguide (USTW) slow-wave structure (SWS), a novel low-voltage high-phase delay SWS operating scheme, and a microfabrication strategy for high-efficiency microscale high-frequency circuits for 1.03 THz traveling-wave tubes (TWTs). Compared with the base STW, the USTW provides stronger electric field focusing and achieves a better match between the electric field distribution and the pencil beam (PB), resulting in a 15.2
It has become more vital than ever to have reliable environmental monitoring and forecasting systems able to operate in real-time. This paper presents an enhanced framework for the transmission and processing of real-time climatic data using Terahertz (THz) communication in combination with a hybrid deep learning model comprising Convolutional Neural Networks and Long Short-Term Memory (CNN-LSTM) networks. It is constructed specifically to support early warning systems and disaster management systems, the system's basic goal is to greatly increase the speed, accuracy, and dependability of the interpretation of climate data. The system that has been proposed for deployment uses THz communication to facilitate the flow of data at high rates between processing units and sensor nodes. Data pertinent to climate including temperature, humidity, wind speed, and air pressure is gathered by means of a network of dispersed sensors. While the LSTM network models temporal connections and trends, the CNN component extracts spatial information from raw data. Extensive simulations and tests based on data acquired from the actual world have shown that, in comparison to more conventional approaches to machine learning, the CNN-LSTM model greatly increases prediction accuracy. The model is completely capable of attaining a high degree of accuracy in the early identification of possible disasters like floods, cyclones, and wildfires. The technology is suitable for very important real-time applications as THz communication ensures very low latency and quite data loss. The results of this study underline, in the framework of the evolution of resilient and smart disaster response systems, the synergy that exists between intelligent data analytics and wireless communication capabilities. The proposed method gradually improves the data transmission rate by 98.75, prediction accuracy by 94
This paper presents the design of an insulated collector for a high-power W-band extended interaction oscillator (WEIO), integrated into the water-cooling system. Beryllium oxide (BeO) ceramic brazing is employed to electrically isolate the depressed collector from the heat sink, enabling the two components to share a single high-velocity cooling circuit with purified water as the cooling medium, thereby simplifying the overall thermal management system. Under steady-state operation with a 43.5 kV, 2.3 A sheet beam, the full-state spent-beam collection method was adopted, and electrons from one representative period after beam-wave interaction were selected for particle-tracking simulation. Based on this analysis, a slope-wall depressed collector was designed. Simulation results indicate that the overall efficiency increases from 18.22 to 48.24
Effective climate modelling and sustainable energy management depend on linking technology for smart energy grids with high-speed communication networks. The aim of this work is to provide a novel framework improving the gathering and processing of meteorological data in real-time by means of integration of smart energy grid optimization and communication at THz. THz communication allows for the transmission of large amounts of environmental data in a short time between central climate modelling platforms and distributed sensors due to its very high bandwidth and low-latency. The proposed study is the Quantum-Inspired Multi-objective Dragonfly Algorithm (QMDA) that integrates swarm intelligence search behaviour of the Dragonfly Algorithm with quantum-inspired operators that include superposition-based representation and probabilistic position updating to augment exploration potential, convergence rate, and flexibility in the complex optimization problems. This can assist us to effectively control this complex system. QMDA, which has been developed to handle several different objectives vying for attention, is intended to assist in controlling many competing objectives, such as minimizing energy loss, lowering latency, and maximizing data throughput, while maintaining efficient load balancing across the energy grid. In terms of convergence speed, solution variety, and adaptation to dynamic conditions, the simulation findings show that QMDA outperforms considerably traditional single-objectual methods and heuristic procedures. The suggested methodology ensures lowest energy waste and more rapid completion of climate simulation feedback loops by means of simultaneous optimization of energy distribution and THz communication channels. This method sets a basis for the future generation of sustainable infrastructure, characterized by the convergence of high-frequency transmission and sophisticated algorithms to enable precise climate forecasting. The proposed method achieves the energy loss by 13
Antenna-coupled field-effect transistors (TeraFETs) have emerged as a class of room-temperature THz detectors capable of competing with Schottky-barrier diodes in sensitivity and response speed. A key advantage of FET-based detectors is their compatibility with mature semiconductor foundry processes, enabling scalable, high-yield fabrication (e.g., 65-nm Si CMOS). Quasi-optical detectors employing planar, ground-plane-free antennas achieve maximum responsivity under substrate lens illumination. However, resonant patch-antenna-coupled FETs cannot employ this configuration due to the presence of a buried metallic ground plane. In [1], we introduced superstrate-coupling strategies for patch antenna-coupled FET resonant at 580 GHz. Here, we investigate eleven front-side-illuminated, superstrate lens-coupled detectors with resonance frequencies between 0.5 and 2.5 THz. A key result is that the minimum optical noise-equivalent power (NEP)—defined relative to the total incident beam power—around the resonance frequencies increases significantly less with frequency than in waveguide-coupled Schottky-barrier diode technology. While the investigated detectors exhibit slightly higher NEP below 1 THz, they outperform commercial Schottky-barrier diodes above 1.5 THz. Compared with TeraFETs employing broadband antennas, patch-coupled FETs provide substantially improved performance at their resonance frequencies. Minimum optical NEP values between 16 pW/√(Hz) at 0.52 THz and 43 pW/√(Hz) at 2.45 THz were achieved and confirmed by broadband thermal radiation measurements. The results are further compared with substrate-lens-coupled detectors employing log-spiral and other broadband antennas reported in the literature.
Terahertz (THz) metalenses have attracted considerable attention as compact and efficient alternatives to conventional bulky optics. In particular, all-silicon metalenses leveraging the high refractive index of silicon (n ≈ 3.4) have been extensively studied in the THz regime. However, a systematic quantitative comparison of such metalenses with commercial THz optics has been lacking. Here, the design and experimental demonstration of monolithic all-silicon THz metalenses, fabricated on high-resistivity silicon (> 10,000 Ω∙cm) using standard photolithography and deep reactive ion etching, are reported. Optimized cylindrical meta-atoms provided nearly 2π phase control and > 90
Based on millimeter-wave (MMW) images, concealed object detection technology has become an ideal choice for high-throughput security scenarios such as large-scale venues, railway stations, and airports due to its non-contact, real-time detection characteristics. Current model designs primarily focus on whether objects are detected, while neglecting the categories of the objects, which makes it difficult to meet the practical security needs for accurately identifying objects of different threat levels. To address this issue, we propose a Dual Receptive Field Enhancement YOLO Network (DRFE-YOLO). By integrating the spatial adaptive receptive field modulation mechanism of Switchable Atrous Convolution and the channel-wise global receptive field aggregation mechanism of the Squeeze-Aggregated Excitation module, a dual receptive field enhancement system is constructed, which synergizes spatial and channel dimensions to improve the categorical discrimination capability for concealed objects of varying sizes and shapes. To validate the effectiveness and feasibility of the proposed method, we constructed a MMW image dataset containing multiple types of concealed objects and evaluated the model’s performance through comprehensive experiments. Experimental results show that the proposed DRFE-YOLO model achieves significant improvements in mean Average Precision (mAP50 and mAP50-95) compared to the YOLOv8n baseline model, with increases of 8.7
The hyperspectral sounders aboard Low Earth Orbit (LEO) satellites produce 515 Gbits of brightness temperature data in each orbital pass, which is many times more than a typical S-band/X-band downlink window can accept. Current compression pipelines, ranging from CCSDS 123.0-B to retrieval-agnostic deep codecs, reduce data volume without preserving the spectral fingerprints required for climate gas retrieval; channel-equal PSNR/SAM loss functions implicitly treat every spectral channel as equally valuable, so high-sensitivity humidity-sounding channels lose the bit budget needed to maintain physical retrieval fidelity, introducing temperature-profile errors of 1.737 K at compression ratios above 10:1. This paper proposes a physics-aware end-to-end edge AI pipeline, validated on three Chinese satellite collections: FengYun-3 MWHS-2 (15-channel THz sounder, 118–183 GHz), FengYun-3E HIRAS-II (2,275-channel hyperspectral infrared sounder), and GaoFen-5 AHSI (330-band VNIR/SWIR imager). The pipeline comprises three components. First, a Radiative-Transfer-Loss (RTL) Codec—an INT8-quantised 1D + 2D convolutional neural network trained with a Jacobian-weighted RTTOV retrieval loss—achieves a compression ratio of 7.5:1 and limits temperature retrieval error to 0.90 K (MWHS-2 synthetic, PSNR = 50.49 dB) and 0.93/0.94 K on FY-3 real data, satisfying the sub-1 K NWP assimilation requirement. Second, a Cloud Filtering and Band Selection (CFBS) module reduces data volume by 38
The terahertz (THz) frequency range holds enormous potential in widespread applications from condensed matter physics to security systems and quantum communications. Despite the progress revealed, the issue related to highly efficient and compatible THz emitters is still of particular importance. For decades, the photoconductive antennas (PCAs) have proved their efficiency in pulse spectroscopic and imaging setups thanks to their simplicity and versatility. Recently emerging plasmonic technology has significantly increased the performance of A3B5-semiconductor based PCAs. On the contrary, topological insulators (TIs) are very promising quantum materials with unique properties of charge carriers. We applied plasmonic technology to Bi2-xSbxTe3-ySey TI PCA to concentrate optical pump power at the PCA’s photoconductor surface that allowed us to achieve a 7.8-fold increase in amplitude compared to conventional TI-based PCA without plasmonic electrodes.
