This paper proposes an unsupervised deep learning (DL)-based hybrid analog precoding framework incorporating true time delay (TTD) and phase shifter (PS) elements to mitigate the severe beam squint effect in wideband terahertz (THz) ultramassive multiple-input multiple-output (UM-MIMO) systems. The proposed framework operates directly in the measurement domain, bypassing explicit channel state information (CSI) reconstruction at inference stage while jointly optimizing TTD and PS coefficients to maximize average array gain across wideband frequencies for multiple users. Specifically, we design a modelbased loss function that guides the training process to maximize wideband average array gain without requiring labeled precoding coefficients. Through this design, and unlike conventional approaches that rely on accurate CSI, the proposed framework learns to map pilot-induced measurements to hybrid precoding coefficients in a single step. Simulation results validate that the proposed DL-based precoding framework effectively mitigates the beam squint effect, achieving near-optimal performance with moderate computational complexity.

In this letter, a coded trihedral corner reflector (TCR) capped by a one-dimensional photonic crystal (PhC) resonator is proposed for passive angular sensing in the W-band. The approach exploits the angle-dependent spectral response of the PhC resonator, while the TCR redirects the resulting signatures toward the retro-direction for mono-static interrogation. Due to interference between scattered waves of the TCR and the resonator, the angle–frequency relationship exhibits local non- monotonic behavior. To model this response, a monotonic staircase approximation is introduced for lookup-table-based angle estimation. Two prototypes with electrical sizes of $9.3\lambda _{0}$ and $15.4\lambda _{0}$ (at 92.5 GHz) are fabricated and experimentally characterized. For the smaller prototype, an average frequency fitting error of 141.4 MHz and an average angle error of $0.94^\circ$ are achieved, while the larger prototype improves the performance to 45.3 MHz and $0.34^\circ$, respectively. The lookup-table-based estimation is further evaluated under measurement noise for the larger prototype using Monte–Carlo simulations, yielding mean and maximum angle estimation errors of $0.84^\circ$ and $3.8^\circ$, respectively, for an SNR of 10 dB. The proposed approach is suitable for passive localization and outdoor sensing applications.

This paper investigates the feasibility of 3D self-localization and tracking using chipless radio frequency identification (RFID) tags operating in the terahertz (THz) frequency band. The primary objective is to achieve sub-millimeter (sub-mm) localization and tracking accuracy while minimizing reliance on external infrastructure. To this end, a hybrid localization framework is proposed that jointly exploits round-trip time-of-flight (RToF) and angle-of-arrival (AoA) measurements to enhance localization performance. Although near-field propagation effects are inherently significant in the considered THz operating regime, a simplified far-field approximation is adopted to facilitate tractable system modeling and analytical development. The proposed framework is further extended to dynamic scenarios through an extended Kalman filter (EKF)-based tracking algorithm, which incorporates temporal state evolution to improve estimation robustness under noisy measurements. Furthermore, the Cramér–Rao lower bound (CRLB) for the hybrid RToF-AoA system is derived to establish the fundamental limits of localization accuracy under varying system configurations and measurement conditions. Simulation results demonstrate that the proposed approach is capable of achieving sub-mm localization and tracking accuracy with a highly constrained anchor infrastructure, including operation with a single anchor in the considered scenario. These findings highlight the potential of THz chipless RFID technology as a promising enabling solution for next-generation high-accuracy localization and tracking applications.

In this work we present two packaged indium phosphide (InP) triple-barrier resonant tunneling diode (TB-RTD) zero-bias detectors with integrated on-chip antenna. The detectors are characterized in the WR1.0 (750–1100 GHz) band in a free-space setup using the lock-in technique. The InP diode chip is mounted by flip-chip bonding onto an FR4 substrate and connected to the lock-in amplifier via an SMA connector. For broadband operation, a bow-tie or log-spiral on-chip antenna is monolithically integrated with the RTD. A silicon lens is attached on the chip backside to facilitate radiation coupling into the backside-illuminated device. The overall detector modules achieve a peak responsivity of 40V/W (bow-tie) to 32 V/W (log-spiral) at a frequency of 787 GHz, taking into account the lens/antenna gain. For comparison with the intrinsic diode performance of 2.1 kV/W stated in [1], possible losses of the module and the measurement setup are shown, and values are given for a few of them (estimated by simulations).

The emergence of new radar systems operating in the terahertz frequency range facilitates new applications such as material defect detection, material characterization, hyperaccuracy localization, and scattering analysis of rough surfaces. To locate defects inside the material, the synthetic aperture radar (SAR) principle can be realized using a radar system, and localization of defects can be implemented in SAR images. Since a material typically exhibits a non-unity refractive index, neglecting this factor during SAR image formation can cause smearing and displacement of internal defects in the reconstructed image. These effects become particularly critical for materials with high refractive indices or when defects are located deep within the medium. In this paper, we present a wave propagation model inside a non-unity refractive index material, and then the model is utilized to develop a backprojection algorithm for SAR imaging inside a non-unity refractive index material. The simulation results are provided to show the effects of non-unity refractive indices on SAR images, whereas the experimental results help us to verify the proposed wave propagation model in practice. The experiments are based on an SAR testbed based on a vector network analyzer operating in the frequency range $220-330 \text{GHz}$ and an electrical insulator with internal damages.

This letter investigates hybrid analog precoding for wideband near-field terahertz (THz) ultra-massive multiple-input multiple-output (UM-MIMO) systems under beam squint and practical true time delay (TTD) and phase shifter (PS) hardware limitations. We formulate the joint optimization of TTD and PS coefficients as a non-convex mixed-integer problem subject to TTD range/resolution and PS quantization constraints. To address this problem, we propose a hardware-aware unsupervised deep learning (DL) framework that operates directly on measurements, requiring channel state information (CSI) only during offline training while enabling CSI-free inference. The framework employs a hardware-aware loss function that maximizes near-field array gain while enforcing TTD and PS constraints. Simulation results demonstrate superior performance over recent schemes under practical TTD and PS limitations.

Practical realization of two-dimensional guidedwave devices in layered dielectrics is addressed by simulation. A transformation-optics method directly realizes the required local variation of the propagation speed (modal index) of the fundamental TE slab mode, without reference to a bulk permittivity distribution. A Maxwell's fish-eye index profile is approximated by mirror-symmetric, two-material layer stacks of finite thickness selected from a lookup library. A 6-port waveguide crossing at 60 GHz serves as example. Results show smooth power guidance and low reflection over 55-65 GHz. Conclusions are drawn regarding achievable discretization accuracy, required layer thickness, and suitable material parameters for low-loss millimeter-wave interconnects in antenna arrays and beamforming networks.

Embedding coding elements in corner reflectors has recently emerged as a promising beacon technology for various applications, such as RFID-based indoor localization systems. Typically, coded reflectors are designed to backscatter at specific resonance frequencies (i.e., frequency codes or IDs), generating either high RCS (peaks) or low RCS (notches) in the backscattered spectrum. The same concept of frequency coding can be utilized to enable angle-of-arrival sensing at the beacon side. This reduces the complexity of reader hardware and algorithms while enhancing localization accuracy. In this paper, we propose a trihedral corner reflector (TCR) with a one-dimensional photonic crystal (PhC) resonator attached at its aperture to sense the angle of the interrogation signal from the reader. The 1D PhC resonator provides angle-dependent signatures, where each angle produces a unique resonance frequency transmitted through the resonator to the corner. The corner reflector then retro-reflects these angle-dependent signatures back toward the direction of interrogation. A non-linear relationship between the incidence angle and the frequency position is observed. The results demonstrate a promising solution for achieving high angular resolution in various applications assisted by coded infrastructure.

Terahertz (THz) Synthetic Aperture Radar (SAR) is an emerging technology capable of generating a high-resolution map of the environment. SAR integration with unmanned aerial vehicles (UAVs) or mobile robots enables promising applications such as indoor/outdoor THz environment map, incorporating high-resolution imaging and localization, surface profiling and material characterization. The aforementioned applications must be executed in real-time, making it essential to propose a computing platform that can accelerate these applications effectively. This paper presents a roadmap towards realizing a computing platform for real-time material mapping utilizing THz SAR. In this paper, we present the state-of-the-art of our novel computing architecture and methodology for material map generation.

The emerge of Terahertz (THz) radar systems allows the short-range applications such as material characterization, hyper accuracy localization at cm or even better level, scattering analysis of rough surfaces, and material defect detection. The paper presents the experimental results about material defect detection based on synthetic aperture radar (SAR) operating at THz frequencies. For the experiments, an electrical insulator is damaged causing the defects in the orders of mm and submm. A SAR testbed at THz frequencies built with a vector network analyzer and a frequency extender in the frequency range 325-500 GHz is used to measure the electrical insulator in the form of SAR with a two-dimensional (2D) aperture. The damaged parts inside the electrical insulator can be observed clearly in the three-dimensional (3D) SAR image. This supports the material defect detection that enhances monitoring industrial production processes and controlling product quality.

Terahertz (THz) technology provides precise monitoring capabilities in dynamic environments, offering unique insights into insect habitats. Our study focuses on environmental monitoring of European honey bees (Apis mellifera) through a combination of measurements and simulations. Initially, the dielectric material properties of honey bee body parts are characterized across the spectral range of 1-500 GHz to collect heterogeneous empirical data. To extend the study, honey bee mockups made from polyamide 12 (PA12) and epoxy resin are employed and validated as effective substitutes for real bees through comparative scattering analyses. The research further explores radar cross-section (RCS), imaging, and spectral properties using advanced THz technologies, including resonant tunneling diodes (RTDs) operating at 250 GHz and THz time-domain spectroscopy (THz-TDS) for frequencies exceeding 250 GHz. High-resolution imaging, utilizing a 450 GHz bandwidth, captures intricate anatomical features of both real and 3D-printed bees, showcasing the potential of THz technology for detailed environmental monitoring. Finally, simulations at 300 GHz assess the dosimetry and feasibility of non-invasive, continuous monitoring approaches based on the heterogeneous honey bee model.

This paper investigates the impact of phase noise on range estimation accuracy in harmonic Frequency-Modulated Continuous-Wave (FMCW) radar systems. Harmonic FMCW radars offer advantages in many applications due to their ability to suppress clutter. However, phase noise, particularly in harmonic systems, presents a significant challenge by degrading range accuracy and increasing frequency measurement errors. In this study, a comprehensive theoretical model is developed to quantify the effects of phase noise on range estimation errors, providing a foundation for understanding its implications on system performance. This model is rigorously validated through both extensive simulations and real-world measurements, offering a holistic assessment of phase noise behavior under practical operating conditions. The results demonstrate that phase noise severely impacts range estimation accuracy, with its effects becoming more pronounced at greater target distances. These findings are further substantiated by experimental evaluations using a practical harmonic radar system, where the system’s range accuracy is analyzed under realistic conditions. This study provides valuable insights and design guidelines for mitigating its impact in harmonic FMCWradar architectures. The results highlight the necessity of advanced phase noise suppression techniques, including optimized hardware configurations and adaptive signal processing methods, to enhance performance in high-precision applications such as industrial positioning and biomedical sensing.

Accurate insect classification is crucial for environmental monitoring and pest control. This study explores the use of terahertz (THz) radar technology for non-invasive and non-contact insect identification. A THz circular synthetic aperture radar (CSAR) system is employed to acquire radar cross-section (RCS) data of insects, enabling the capture of finer details and distinctions of small insects. Digital twins simulation is used as a tool for generating synthetic radar datasets, which will be used in the future to train machine learning models. The experimental results demonstrate the ability of THz radar to resolve fine structural features of insects, offering improved identification capabilities compared to microwave radars.

The use of THz frequencies has enabled the opportunity to perform synthetic-aperture radar (SAR) imaging at the sub-mm level. It is of great interest for applications where high-resolution remote sensing in short range is required. However, with the increase of the operating frequencies from microwave to THz, the SAR image formation algorithms work with a larger amount of data and become more sensitive to phase errors that can be caused by insufficient signal sampling rate or physical factors that cause platform deviations. This motivates the use of fast image formation capable to handle phase errors. In this paper, we present the experimental results on the performance of the local backprojection (LBP) algorithm for processing THz SAR signals. The LBP algorithm has been tested with the real data in the frequency range 0.325-0.5 THz. The results demonstrate the efficiency of the LBP algorithm for the SAR scene reconstruction and highlight the necessity of the use of two times higher signal upsampling to achieve reconstruction accuracy similar to the global backprojection algorithm.

Terahertz technology positions itself as a promising alternative to conventional vital signs measurement methods due to its unique characteristics, making it particularly suitable for non-contact measurements in free space. In this work, the influence of different frequency bands and clothing on the accuracy of respiration rate measurements is analyzed, using a motion-based method that utilizes amplitude information. The highest accuracy is obtained in the lowest studied frequency band of 220GHz- 330 GHz, with a mean absolute error of 0.34 rpm. Notably, clothing does not impede the measurements. The results demonstrate the capability of terahertz technology to effectively detect and monitor respiration rate.

In emergency situations involving fire and smoke, optical and infrared sensors offer limited information due to visual obstructions and intense heat, whereas radar sensing seems to be a promising solution. Radar sensing's ability to penetrate smoke and fire is well established at microwave frequencies of sub-24 GHz. However, spatial resolution is limited at these frequencies, prompting the exploration of higher-frequency regions. Therefore, this study investigates radar sensing within the millimeter-wave (mmWave) spectrum, specifically in the 75-110 GHz frequency range. A vector network analyzer (VNA) based testbed in a monostatic configuration is implemented, and test cases involving the standardized generation of various smoke and flame types are examined. Synthetic aperture radar (SAR) technique is employed to generate a 2D map of the investigated environment. The map is explored for estimation of attenuation in signal power due to the fire. Additionally, the imaging results are analyzed for artifacts resulting from phase errors caused by flame and smoke.

In this paper, we propose an ultra-resolution channel estimation scheme for ultra massive multiple input multiple output (UM-MIMO) systems operating in the Terahertz (THz) band. Specifically, we formulate the channel estimation as a non-convex sparse signal recovery problem, characterized by the absence of a global minimum solution and the lack of knowledge about the actual number of paths. To solve this problem, we propose a hierarchical multi-grid orthogonal matching pursuit (HMG-OMP) scheme. The proposed scheme accurately estimates the UM-MIMO channel while concurrently identifying the unknown number of paths. We extend the HMG-OMP scheme to wideband channel estimation using the concept of subcarrier-grouping. Particularly, the subcarriers in the wideband regime are bundled into groups with an instance of HMG-OMP executed per group, aiming at estimating the wideband channels with limited computational complexity. We provide a complexity analysis of the proposed schemes in comparison to recent literature, validating their feasibility and practicality. Furthermore, numerical simulations demonstrate their superiority over existing literature works, where ultra-resolution estimation accuracy is attained.

This paper presents an optimization approach for static antenna array thinning using a genetic algorithm (GA). The goal is to reduce hardware complexity while maintaining a radiation pattern similar to that of a fully populated uniform linear array (ULA). The proposed method statically deactivates a specific percentage of the array elements, achieving the desired hardware complexity reduction. Furthermore, a previously proposed algorithm for compensating the beam squint effect in a hybrid beamforming architecture, initially tested on a ULA, is applied to the thinned array, demonstrating its effectiveness on non-uniformly distributed arrays.

This paper investigates the effectiveness of the extended Kalman filter (EKF) as a tracking algorithm for indoor radio frequency identification (RFID) systems operating at terahertz (THz) band, with a critical objective of achieving sub-millimeter (sub-mm) accuracy and minimizing the infrastructure dependency. We acknowledge the inherent trade-off between the number of anchors employed and the achievable accuracy. This study presents a comparative analysis of localizing and tracking a moving object between the least squares (LS) and the EKF estimators, focusing on the minimum number of anchors required to attain sub-mm accuracy.

Monitoring and farm surveillance, as part of biosecurity measures, involves observing the presence, absence, and population densities of both pollinators and pests. When considering future miniaturized mobile terahertz (THz) environmental monitoring systems for tracking minuscule radar targets such as aerial insects (e.g., honey bees), a key challenge lies in their reduced radar cross section (RCS). This results in lower backscattered power, thereby limiting the maximum detection range and reducing applicability across various scenarios. In this paper, a realistic 3D model of a Western honey bee and a Varroa mite is introduced. While the complex permittivity of the honey bee was measured experimentally in our previous work, the Varroa mite was modeled using estimated homogeneous dielectric parameters. Using a monostatic radar configuration, RCS analysis is performed at a carrier frequency of 300 GHz to identify the presence of a mite on a honey bee and confirm its precise location.