ABSTRACT Tunable terahertz (THz) absorbers require broadband response, dynamic control, and simple device architectures. Here, we propose a dual‐state VO 2 –graphene composite metasurface comprising one patterned graphene layer, a SiO 2 dielectric spacer, and a VO 2 substrate. The thermally induced insulator‐to‐metal transition of VO 2 and the electrically tunable graphene Fermi level provide two independent control dimensions for absorption and transmission. At 345 K, the device achieves broadband absorption ( A > 90%) from 3.00 to 7.25 THz, corresponding to a 4.25 THz bandwidth and an average absorptivity of 0.98. At 313 K, it retains moderate absorption and a tunable transmission response, enabling reversible absorption–transmission switching. Impedance matching and overlapping localized surface plasmon resonance (LSPR) modes account for the high‐temperature broadband absorption. The change in VO 2 conductivity regulates the loss distribution and resonance coupling, whereas increasing the graphene Fermi level strengthens the plasmonic response and broadens the operating band. This design provides a feasible solution for multifunctional THz devices and has potential application value in intelligent thermal control, dynamic filtering, THz modulation, and sensing.
ABSTRACT This paper presents a novel stepwise numerical method based on Fibonacci wavelets for solving fractional stochastic delay differential systems. Such systems arise in numerous applications where memory effects, randomness, and time delays coexist, and their accurate long‐time numerical treatment remains challenging. The proposed method combines the fractional derivative operational matrices associated with the Fibonacci wavelet basis with a stepwise discretization strategy, thereby integrating high spectral accuracy with stable computation over extended time intervals. First, the arbitrary‐order fractional derivative operational matrix for the Fibonacci wavelet basis functions is constructed via block pulse functions. Then, the original problem is solved on successive subintervals using a recursive stepwise procedure that efficiently incorporates the delay term. At each step, the solution is approximated by a Fibonacci wavelet expansion; applying collocation conditions together with the operational matrix reduces the problem to a system of algebraic equations. A rigorous convergence analysis and error bounds of the proposed method are provided. Finally, several numerical examples and comparisons with existing methods demonstrate the efficiency and robustness of the algorithm. The results confirm that the method achieves remarkable accuracy while maintaining computational efficiency for problems defined on long time domains.
ABSTRACT The increasing penetration of renewable energy resources, energy storage technologies, and flexible AC transmission system (FACTS) devices has significantly increased the complexity of day‐ahead dynamic optimal power flow (DOPF) problems. The simultaneous consideration of wind power uncertainty, multi‐mode compressed air energy storage (CAES), STATCOM control, and valve‐point loading effects results in a highly nonlinear, nonconvex, and time‐coupled optimization problem. To address this challenge, this paper proposes an oppositional supercell thunderstorm algorithm (OSTA), in which opposition‐based learning is incorporated into the supercell thunderstorm algorithm (STA) to enhance population diversity, improve exploration capability, and mitigate premature convergence. A stochastic multi‐period DOPF framework is developed by integrating wind power uncertainty modeled through the Weibull probability distribution, detailed CAES operational characteristics including charging, discharging, and gas turbine modes, and STATCOM control within a unified optimization model. The objective is to minimize the total operating cost while satisfying power balance, network security, and operational constraints over a 24‐h scheduling horizon. The effectiveness of the proposed methodology is validated through multiple case studies on the IEEE 30‐bus system, while its scalability is further assessed on the IEEE 118‐bus system. Comparative evaluations against STA, marine predators algorithm (MPA), particle swarm optimization (PSO), and grey wolf optimizer (GWO) demonstrate that OSTA consistently provides superior solution quality, improved convergence characteristics, enhanced voltage profile performance, and better voltage stability margins. The obtained results confirm the effectiveness, robustness, and scalability of the proposed OSTA for solving large‐scale stochastic DOPF problems in renewable‐rich power systems.
ABSTRACT The special issue collects 13 papers presenting recent advances in the measurement, modeling, and simulation of microwave and terahertz devices for communication and sensing applications. The contributions discuss electromagnetic material characterization, microwave and THz devices for biomedical and healthcare applications, and component design for IoT, wearable, and biomedical technologies. Novel modeling and measurement techniques are also introduced, with an emphasis on linear and nonlinear approaches for accurate calibration. The application of artificial intelligence and machine learning in model development and validation is also explored, alongside the design, simulation, and performance analysis of high‐frequency circuits and systems. These articles highlight current progress and emerging directions in the modeling, characterization, and application of high‐frequency technologies.
We present a ferroelectric-gated 2D material-based high electron mobility transistor (HEMT) that provides an innovative solution for the challenges associated with the electrostatic control, the carrier density and the power performance of the conventional 2D FETs and III-V HEMTs. The device employs an atomically thin monolayer InSe channel and a ferroelectric HfO2 high-k gate stack that exploits the negative capacitance effect of ferroelectrics to achieve strong charge control and voltage gain for the 2D channel. Unlike the polarizations in conventional polarization-induced HEMTs, the polarization created by the ferroelectric leads to the formation of a high carrier density 2DEG in an atomically thin channel. The electrostatics, transport, and RF performance of the device are modeled using detailed TCAD simulations. The simulated device has a positive threshold voltage (similar to 0.35 V), allowing it to operate in the normally-OFF regime. The simulated 2DEG has a charge density of similar to 9 & times; 10(12) cm(-2) and an electron mobility of similar to 1000 cm(2)/V-s. We show peak transconductance values of similar to 500 mS/mm, low OFF-state leakage current densities below 10(-12) A, improved subthreshold behaviors approaching sub-thermionic switching, and excellent scalability, suppression of short-channel effects, and good electrostatic gate control through the combined effect of the ferroelectric polarization and the 2D confinement mechanisms. Thus, the proposed ferroelectric-assisted 2D HEMT overcomes the issue of conventional 2D FET and III-V HEMT both the carrier density and gate control are improved simultaneously with low-power consumption. This makes the ferroelectric-assisted 2D HEMT a good candidate for next-generation Nanoelectronic, RF and biosensing applications.
A combined computational and experimental approach was used to characterize crack shapes in flawed specimens. The first step involved a three-dimensional finite-element method based on the (A, V-A) formulation to analyze field variations in cracked conductive materials and to evaluate the influence of defect geometry on the eddy-current response. This numerical model enabled the determination of crack shapes. The study also employed a giant magnetoresistance (GMR) sensor to measure signals from different crack forms using a GMR-based eddy-current (EC) probe. The model was validated experimentally through a prototype unit, and measurements were performed on aluminum reference standards containing various crack types. Furthermore, a TinyML model was developed using the Edge Impulse platform to automatically classify crack shapes according to relevant standards. Using the GMR-based EC probe, the system achieved a mean accuracy of 98%, demonstrating the feasibility of the method. A key advantage of this approach is the rapid and efficient development of embedded machine-learning models enabled by the open-source platform. The approach offers a cost-efficient solution for industrial NDT, with future improvements focused on expanding the dataset and validating system performance in real operating environments.
ABSTRACT In this work, authors have deeply investigated a combination of gallium nitride (GaN) and gallium oxide (Ga 2 O 3 ) semiconductors, which are considered strong potential candidates for solar‐blind ultraviolet (UV) photodetectors. Using Silvaco TCAD software, the proposed structure contains platinum as a Schottky contact on the top of the Ga 2 O 3 absorbed layer. An intermediate GaN layer was deposited on the sapphire (Al 2 O 3 ) substrate, and it has a crucial role in increasing the photogeneration rate of the used photodetector. The proposed device shows distinct current–voltage behavior under dark and illumination conditions. A high responsivity in the deep UV region has been demonstrated. Different parameters, such as doping concentration, work function, traps, and temperature effect, have been investigated to highlight the potential of the proposed structure for the next generation of UV‐Schottky photodetectors with excellent spectral selectivity and self‐powered effect.
ABSTRACT Device modeling refers to the representation of electronic devices through mathematical equations, equivalent circuits, or look‐up tables (LUTs), supported by sound physical assumptions and reasoning. To simulate any electronic device in a circuit simulator, an accurate model is essential. The double‐gate (DG) MOSFET is considered a promising candidate for future CMOS technologies. For effective integration into CMOS circuits, models for both n‐channel and p‐channel DG MOSFETs are required. Although extensive studies have been conducted on n‐channel DG MOSFETs, limited attention has been paid to their p‐channel counterparts, particularly in the nanoscale regime. Additionally, accurate device simulation requires a well‐calibrated simulation environment, configured using data from standard reference devices. This work focuses on the nanoscale p‐channel DG MOSFET to enable its use in DG‐CMOS circuits for future integrated systems. A device‐level simulation is performed using the device simulator to extract potential distribution and drain current characteristics, with the ultimate goal of developing a LUT‐based model. The p‐channel structure is simulated on a silicon substrate and then replicated using germanium to compare performance. Guidelines for configuring the simulation environment are also provided. The simulation environment is tuned using drift‐diffusion transport, constant mobility, and optimized source/drain (SD) region and electrode lengths to achieve a high level of accuracy. From the simulation study, it was observed that germanium‐based devices are more susceptible to short‐channel effects (SCEs) at the nanoscale. The simulation environment is configured using minimal‐physics models, simple yet accurate numerical methods, and calibrated reference data. Analytical modeling of the p‐channel DG MOSFET also shows high accuracy when compared with reference results. These findings contribute to the enablement‐balanced modeling for DG–CMOS technology.
ABSTRACT The Seebeck coefficient is an intrinsic thermoelectric material property, and is a key parameter in microelectromechanical systems (MEMS) sensors and energy harvesting based on the thermoelectric principle. However, extracting this coefficient accurately from MEMS test structures demands a careful co‐design to balance measurement fidelity against power consumption. Till now, the trade‐off between accuracy and power consumption is not very well addressed by traditional approaches based on computationally expensive finite element analysis (FEA) or trials and errors methods. To address this issue, we present a physics‐guided deep learning (PGDL) surrogate model that predicts key device level parameters; output voltage, heater resistance and temperature of the hot and cold junctions through model training on 5000 high‐fidelity data‐set generated on COMSOL Multiphysics simulations. Rigorous benchmarking against conventional deep learning, random forest, k‐nearest neighbors, and gradient boosting, confirms that PGDL consistently delivers superior predictive accuracy and generalization. In PGDL, domain‐specific physical knowledge for instance, expected monotonic trends between geometry and thermal response are integrated in a weakly constraint form during training, improving generalization and avoiding non‐physical predictions without imposing a form of a partial differential equation. We subsequently combine this surrogate with the NSGA‐II multi‐objective optimizer and use it to minimize (i) the relative error in recovering a known reference Seebeck coefficient, and (ii) the heater power consumption subjected to constant current operation concurrently elucidating the trade‐offs between conflicting objectives. The achieved Pareto optimal design consequently realizes a power dissipation drop by 48% (1.63 to 0.85 mW ), and minimize the Seebeck prediction error up to 2.1%–2.96%, while allowing an R 2 values between 0.948 and 0.970 across all model predicted outputs. This research demonstrates a scalable, data efficient, and physically consistent computational framework for the automated co‐design and in situ characterization of MEMS thermoelectric devices.
This study proposes a compact Vivaldi-based phased array antenna (PAA) for the non-destructive detection of internal defects in tree trunks at varying heights. The proposed PAA comprises four specially designed Vivaldi elements operating from 0.6 to 1.4 GHz, enabling deep penetration into tree trunks while maintaining high resolution. To achieve a compact design with a narrow inter-element spacing of 0.4 and reduced mutual coupling, the outermost elements are slightly inclined, where is the wavelength at the center frequency. Additionally, a slotted metal reflector is placed behind the elements to improve impedance matching at low frequencies. The array is excited using Wilkinson power dividers to ensure stable amplitude and phase distribution. The measurement results show that the proposed PAA can steer its beam from -30 degrees to 30 degrees while maintaining an active voltage standing wave ratio (VSWR) below 2.5. It achieves a gain of 12-13 dBi while steering the beam. These characteristics, including a wide bandwidth, high gain, compact size, and beam-steering capability, make the proposed PAA highly suitable for scanning tree trunks and detecting defects at different heights. The proposed PAA was further validated on a real tree trunk, confirming its effectiveness for defect detection in practical scenarios.
Arithmetic circuits are the basic building blocks of frequently used computational systems. As the functions used in the computational system increase, the design constraint effects are proportional to the requirement. Approximate computing is a technique with the advantage of controlling design constraints by trading off accuracy. Approximation has major challenges in image processing, speech processing, and machine learning. So, an image-processing blending application, noise removal, or changing the transparency of images are designed using arithmetic circuits like adders and subtractors, respectively. Here, an 8-bit Dadda multiplier and a restoring divider are designed using approximate compressors and subtractors, respectively. The approximate full adders and full subtractors are implemented using tabular manipulation methods and logic-level pruning. The full adders are replaced either in the MSB or the LSB positions for better accuracy and error analysis. Image processing blending application, and background removal are performed using MATLAB, and design constraints are evaluated using Synopsys DC Compiler. The results show the logic-level pruning compressors, multipliers, and restoring dividers get a 40%-70% gain in PDP.
This investigation proposes a mathematical framework for the drain current as well as short-channel effects of an n-type heterojunction gate-all-around (GAA) nanowire tunnel field-effect transistor (TFET) architecture that incorporates a core-insulator within the channel (HJ-CI-GAA-NW-TFET). An analytical model for drain current is formulated employing Kane's interband tunneling model in conjunction with the shortest tunneling length approach. To derive the drain current and SCEs, numerical integration of the tunneling generation rate over the entire device volume is performed. The influence of variations of gate work function, diameter of the core-insulator, gate oxide thickness, and source doping concentration on several electrical parameters is systematically evaluated. Furthermore, the drive current is computed at a scaled channel length of 10 nm and under different gate-source voltages (V GS) spanning from 0.8 to 1 V. The analytical model developed yields results that agree well with Silvaco 3D TCAD simulations, reinforcing the validity of the analysis with just a 7% marginal deviation for different device parameters.
Behavioral models for Power Amplifier (PAs), particularly those based on Neural Networks (NNs), often suffer from high computational complexity when handling wideband signals. In this paper, we present a novel behavior model for wideband GaN PAs based on a Two-Sided Temporal Convolutional Neural Network (TST-CNN). The model uses a specialized convolutional layer to process the input data in order to extract the essential basis functions needed to accurately capture the dynamic memory effects and static nonlinearities of the PA. These enhanced features are then fed into a Fully Connected (FC) layer to establish the predictive model. Due to the efficient feature extraction enabled by the convolutional structure, the proposed architecture successfully handles strong memory effects without a significant increase in model complexity. The experimental results, obtained at a carrier frequency of 2.14 GHz using a 100 MHz wideband signal emulating 5th Generation (5G) wireless technologies, demonstrate that the TST-CNN outperforms current state-of-the-art models by achieving an Adjacent Channel Power Ratio (ACPR) of -52.1 dB. Furthermore, the TST-CNN requires only 237 coefficients, resulting in a high reduction in computational complexity compared with existing approaches, which makes it a very effective solution for future wideband communication systems.
This paper presents a developed iterative algorithm based on a Time-Domain Electric Field Integral Equation (TD-EFIE) formulation to analyze electromagnetic coupling in antenna systems. The primary objective is to determine the optimal immunity distance d that ensures mutual electromagnetic compatibility (EMC) between two antennas when one is subjected to atransient electromagnetic field. The proposed methodology employs the Method of Moments (MoM) in the time domain, utilizing Laguerre polynomials as temporal basis functions for stable recursive solution through a Marching-On-in-Degree (MOD) scheme. Space hybrid meshing techniques are applied to accurately model complex geometries. The core functionality of the algorithm consist in iteratively solving a coupled matrix system to compute transient induced currents, and comparing them with threshold values to determine the minimal safe separation distance. Validation through case studies-coupling between parabolic antennas and between a parabolic and a wire antenna demonstrates strong agreement with commercial simulator FEKO, confirming the algorithm's accuracy and computational efficiency. An evaluation of the proposed algorithms is conducted with respect to their computational complexity, CPU requirements, efficiency, and residual error. The resulting matrix system exhibits a computational complexity of where N and M denote spatial and temporal degrees of freedom, respectively. Compared with simulations performed using FEKO, the proposed algorithm achieves up to 40 & times; reduction in CPU time while maintaining strong accuracy agreement.
This paper presents two CMOS-based memtranstor emulator architectures capable of accurately replicating the flux-charge (phi-q) relationship and pinched hysteresis loop (PHL) characteristics of an ideal memtranstor. Both designs employ a voltage differencing inverting buffered amplifier (VDIBA) along with two second generation current conveyors (CCIIs), an analog multiplier, and a few passive components to realize nonlinear charge-flux coupling. The first topology offers a compact, hardware-efficient design with tunable butterfly-shaped PHLs, whereas the second design introduces an external control voltage that enables quadrant-dependent operation and adjustable hysteresis slopes. The proposed emulators operate at a low supply voltage of +/- 0.9 V, exhibit an operating frequency range of 35-50 kHz, with an average power consumption of 12.65 mW. Further, the emulator operates reliably for low frequencies up to 3 kHz, by appropriately setting the values of passive components. The tunability of the proposed memtranstor design is demonstrated by adjusting the bias voltage, excitation frequency, and capacitance. To validate the practical feasibility of the proposed architecture, a full-custom layout of the memtranstor emulator has been designed, occupying a silicon area of 12882.93 mu m2 (165.06 & times; 78.05 mu m). The post-layout results show close agreement with the pre-layout simulations while preserving the distinct memtranstance states of the emulator. Furthermore, the robustness of both designs is verified through extensive process-voltage-temperature analyses, confirming stable and reliable performance under device mismatch and environmental variations. The proposed memtranstor emulator circuits have been designed and simulated using Cadence Virtuoso 180 nm gpdk CMOS technology framework. The practical utility of the proposed emulators is further validated through two applications namely a chaotic oscillator-exploiting their nonlinear dynamics and an artificial synapse circuit-demonstrating long-term potentiation and depression for neuromorphic computing. In addition, the hardware implementation of the proposed memtranstor emulator and chaotic oscillator has been carried out, to experimentally verify the functionality of the design. All these results highlight the proposed VDIBA-CCII-based memtranstor emulators as compact, tunable, and energy-efficient circuits for nonlinear, chaotic, and neuromorphic hardware systems.
In this article, ultra-wideband (UWB) single-element and two-port multiple-input-multiple-output (MIMO) antenna structures with circular polarization (CP) characteristics are designed to support numerous modern wireless systems. The basic CP element is designed to obtain an ultra-wide operating bandwidth. In the design process, a circular stub is integrated into the ground plane, and a circular slot is structured into the patch to achieve CP. On the other hand, a rectangular stub is used to improve the axial ratio bandwidth (ARBW). The elemental antenna achieves an impedance bandwidth (IBW) of 97.96% (4.13-12.06 GHz), an ARBW of 49.62% (5-8.3 GHz), a peak gain of 5.75 dB, and a minimum efficiency of 80%, maintaining small dimensions of 20 mm 20 mm 1.6 mm (0.275 lambda( 0) & times; 0.275 lambda( 0) & times; 0.022 lambda( 0)). Meanwhile, a two-element UWB MIMO radiator of 20 mm 44 mm 1.6 mm (0.275 lambda( 0)& times; 0.606 lambda( 0) & times; 0.022 lambda( 0)) is designed and experimentally validated. It addresses the challenges of achieving wide impedance and axial-ratio bandwidths while maintaining a compact size and high isolation for MIMO operation with attractive diversity performance. It achieves an impressive ultrawide IBW of 100% (4-12 GHz), ARBW of 42.73% (4.6-7.1 GHz), and 10.90% (7.8-8.7 GHz) while offering improved isolation. The MIMO geometry exhibits outstanding diversity performance, with an envelope correlation coefficient (ECC) < 0.01, diversity gain (DG) > 9.92 dB, total active reflection coefficient (TARC) < 10 dB, and channel capacity loss (CCL) < 0.2 bits/s/Hz. These features render the proposed UWB MIMO antenna exceptionally appropriate for various wireless communication uses, such as microwave C-band (4-8 GHz), WiMAX (5.725-5.850), WLAN (5.150-5.825 GHz; 5.925-7.125 GHz), and satellite communication in X-band (downlink: 7.25-7.745 GHz and uplink: 7.9-8.4 GHz).
To overcome the limitations of narrow bandwidth, structural complexity, and limited tunability in existing graphene-based terahertz absorbers, structural complexity, or inconvenient tunability in existing graphene-based terahertz absorbers, this paper proposes a tunable broadband terahertz absorber based on a graphene proportional metasurface. Here, "proportional" refers to the fixed ratio between the graphene pattern feature size () and the unit cell period (), that is, , as well as the consistent size ratios among different resonant regions. The design employs a classic three-layer structure, consisting of a gold reflective layer, a silicon dioxide dielectric layer, and a graphene resonant absorption layer. Through systematic optimization, the absorber achieves a high-efficiency absorption bandwidth exceeding 90% across a range of 3.003 THz, with a peak absorption rate of 98.8% at 2.61 THz and an average absorption rate of 94.6% over the entire bandwidth. It also exhibits excellent impedance matching characteristics and angular adaptability. Research indicates that the broadband absorption property of the device originates from the electric field coupling effect between different resonant units. This coupling effect causes multiple absorption peaks to overlap, thereby broadening the bandwidth. Furthermore, by applying an external voltage to adjust the Fermi level of graphene, the absorption bandwidth can be continuously tuned. The device offers flexible dynamic control capabilities and demonstrates good tolerance to the incident angle of electromagnetic waves. This study provides a feasible approach for designing structurally simple and tunable high-efficiency terahertz absorbers. The proposed design holds potential application value in fields such as terahertz detection, communication, and stealth technology.
Small-signal model of a MOSFET is an equivalent circuit that represents its electrical components and specifies the device's electrical characteristics. The non-quasi-static (NQS) model is among the most precise small-signal models utilized in the design of analog and RF circuits. This work introduces an innovative device design known as Gate Stack Silicon on Insulator Schottky Barrier (GS-SOISB CGAA) cylindrical MOSFET and analog/RF characteristics are extracted utilizing the Silvaco 3D device simulator. An analytical model of GS-SOISB CGAA is also introduced and validated against the simulation findings. The NQS model of the GS-SOISB cylindrical MOSFET is formulated to ascertain the extrinsic and intrinsic parasitic components utilizing Y-parameters in both the on and off states, respectively. The analog/RF characteristics and NQS model parameters of the GS-SOISB cylindrical MOSFET have been compared with those of the SOISB cylindrical MOSFET and the SB cylindrical MOSFET. The revolutionary device has a lower gate-source capacitance (C GS) of 7.95% and 8.79%, a gate-drain capacitance (C GD) of 7.22% and 4.57%, and an overall gate capacitance (C GG) of 7.91% and 8% than SB and SOISB cylindrical MOSFETs. Compared to conventional SB cylindrical and SOISB cylindrical MOSFET structures, the proposed device shows reduced extrinsic parameters of the NQS model by 7.93% and 3.99% in C gde and C gse, respectively, and 33.4% and 23.9% in R D and R S. The revolutionary device's intrinsic gate-drain capacitance (C gd) reduces by 7.13% and 4.29%, and its intrinsic gate-source capacitance (C gs) reduces by 7.93% and 8.79% as compared to SB and SOISB cylindrical MOSFETs.
The photovoltaic (PV), wind turbine (WT), and battery energy storage (BES) based hybrid system design and optimal placement using chaotic quasi-oppositional crayfish optimization algorithm (CQOCOA) in a radial distribution network (RDN) under load uncertainty is the main objective of this study. Here, crayfish optimization algorithm (COA) is modified and improved by adding quasi-oppositional behavior to it. Then chaos theory is added to speed up the convergence pace and avoid the local optimality. For optimal placement of hybrid PV/WT/BES system, simultaneous active power loss and annual operation costs minimization is taken as the objective to enhance the efficacy of the RDN. The uncertainty modeling of PV and WT distributed generation (DG) is considered for power generation as solar irradiance and wind speed can change. This algorithm is validated on 69-bus and 94-bus to establish the potency of the suggested CQOCOA algorithm. The active power loss cost is also evaluated after the installation of hybrid PV/WT/BES system. Adjusting the growing load demand, 25% increased load and 10% decreased load is considered for load uncertainty modeling. In both (69-bus and 94-bus) systems, the placement of hybrid PV/WT/BES system using the CQOCOA method reduces the active power loss by 58.93%, 60.53%, 53.53%, and 62.19%, 65%, 62.99% for normal, 25% increased, and 10% decreased loading conditions, respectively. In yearly running cost of hybrid system design by CQOCOA method for 69-bus at normal and 10% decreased load gives yearly savings of 25 364$, 31 951$, 88 951$ and 16 511$, 1527$, 25 608$ than COA, DAOA, and AOA methods. The comparative study of results revealed that the CQOCOA algorithm is better than several optimization algorithms.
