Journal of Computational Electronics最新文献

The implementation of a neural network on very large-scale integrated (VLSI) circuits provides flexibility in programmable systems. However, conventional field-programmable gate array (FPGA) neural chips suffer from longer computation times, higher costs, and greater energy consumption. On the other hand, multilayer perceptron (MLP) network implementation over VLSI exhibits increased speed with a smaller chip size and reduced cost. This work aims to implement an MLP neural network using double-gate metal oxide semiconductor field effect transistors (DGMOSFETs) functioning as neurons. The suggested network architecture is offered as a package utilizing very high-speed integrated circuit hardware description language (VHDL). The weights of the MLP are obtained by training a neural network with electrocardiogram (ECG) signals taken from the PhysioNet database. The ECG input signals, obtained weights and bias, are given to the designed MLP for testing. The classification accuracy of this trained neural network is 94.48%. A power analysis is also conducted for the hardware-designed MLP to validate the power reduction performance. In terms of speed, the required number of components and power, the performance of this design employing DGMOSFET outperforms its single-gate MOSFET (SGMOSFET) counterpart.

The modern electronic devices’ development heavily relies on the miniaturization of MOSFET transistors. On the other hand, reduction in transistor sizes will face significant challenges, like short-channel effects. To enhance transistor performance, it is essential to explore and utilize new materials. Black phosphorene has emerged as a promising material for constructing transistors and other electronic components. Accurate modeling is crucial for predicting the behavior of future nanoscale transistors. One of proposed simulation methods is the top-of-barrier model. This study analyzes transistors based on black phosphorene nanoribbons. The electronic structure of these nanoribbons is calculated using the tight-binding method with up to five nearest neighbors. The top-of-barrier computational approach within the Landauer framework is employed to determine device characteristics. Initial evaluations of a structure without antidots show that creating an off-center antidot increases the on current to 4.98 mA. The threshold voltage also rises by 0.2 V, indicating an increase in the energy band gap, which reduces the off current significantly. The on/off current ratio can be improved by up to 2500 times with an optimal antidot design. Non-central antidots do not significantly affect the threshold voltage or off current.

This manuscript investigates how magnetic and non-magnetic effects influence the firing patterns, oscillations, and synchronization properties of the Hindmarsh–Rose neuron model under different magnetic conditions. The development of a fractal–fractional Hindmarsh–Rose neuron model is proposed for investigating self-similarity across different scales to analyze and understand the complexities when extreme magnetic flux varies and reaches its critical value. The mathematical modeling of the Hindmarsh–Rose neuron model is established under an application of the Caputo–Fabrizio and Atangana–Baleanu fractional differential operators. For the sake of numerical simulations via the Adams–Bashforth–Moulton method, the discretization of spatial and time domains on fractal–fractional derivatives is employed to generate numerically powerful schemes within approximate accuracy. For understanding the brain function and neural oscillations, the magnetized and demagnetized Hindmarsh–Rose neuron model revealed suppressed neuronal activity and the effects of transcranial magnetic stimulation. Our results suggested two aspects: one is trapping of neurons, striking phenomena and firing patterns under demagnetization, while the other is neurological disorders, spiking and bursting in neurons based on neural interfaces under demagnetization.

The performance of 2D material-based field-effect transistors (2D FETs) is significantly influenced by the vertical extension, or depth, of electrostatically doped side Schottky contacts, which is determined through etching. This study employs TCAD modeling to compare back-gated FETs with varying source/drain contact depths and channel lengths. Results indicate that deeper side contacts hinder electric field crowding at the metal/channel interface, resulting in wider Schottky barriers, diminished carrier tunneling, and reduced on-state current. In contrast, introducing a low-k dielectric beneath the source and drain yields the opposite effect. Therefore, in the development of industry-compatible 2D FETs, the depth and design of side contacts must be carefully optimized, as they are critical factors in achieving low-contact resistance devices.

In this work, we present the study of AlGaN/GaN metal–oxide–semiconductor high-electron-mobility transistor (MOS-HEMT) biosensors for protein detection. We study the effects of technological parameters including the gate width, gate length, AlGaN layer thickness, oxide thickness layer, and oxide type including HfO₂, Al₂O₃, and SiO₂ on the output characteristics, sensitivity of the MOS-HEMT biosensors, and C–V characteristics. The model developed is compared with experimental data to verify its validity. The AlGaN/GaN bio-MOS-HEMTs show the greatest change in drain current of 208.08 mA with W_g = 100 µm, L_g= 0.3 µm, d_AlGaN=15 nm, and SiO₂ oxide thickness of 25 nm at protein permittivity of 2.5.

This paper introduces a novel physics-informed learning approach that combines principles from physics with deep learning techniques to optimize the simulation process of microstrip antennas. These deep learning-based approaches are preferable because traditional full-wave models used in antenna design are computationally intensive and require significant memory due to their reliance on iterative algorithms, leading to exponential increases in resource demands as input parameters grow. In contrast, the proposed deep learning method requires significant computational resources only during training, with a constant time complexity of O(1) during deployment. This results in much faster modeling, allowing a broader range of antenna configurations to be processed more quickly, thereby improving the efficiency of the design workflow. Unlike conventional deep learning methods that rely solely on data, our approach leverages the underlying physical laws governing antenna behavior, particularly beneficial when labeled data is scarce or difficult to obtain. We propose a bias observational physics-informed learning technique by integrating physical laws into the loss function, which includes two components: Neuron Loss, the standard MSE measuring prediction accuracy against actual data, and Physics Loss, which penalizes deviations from physical laws as represented by a cavity model. The total loss combines these two, with higher physics loss indicating poorer alignment with physical principles and lower physics loss suggesting better adherence to them. This approach refines predictions by balancing data fidelity with physical constraint, wherein the dataset is sourced from simulations and real-world measurements. This strategy ensures model uncertainty and broad generalization capabilities. Computational efficiency is a key consideration, with our approach implemented on low-specification hardware, emphasizing optimal resource and power consumption. The H-shaped microstrip antennas (HMAs), known for its wide and dual-band properties, serves as the target antenna for our study. We employ three sequential models’ recurrent neural networks (RNN), long short-term memory (LSTM), and gated recurrent unit (GRU)—integrated with a cavity model-driven resonance frequency representation to maintain the resonance mode TM¹⁰ at prediction. Comparative analysis of these models encompasses execution time, prediction convergence, loss reduction, prediction score (R²), as well as memory and CPU usage. This research contributes four main sections elucidating the methodology, experimental setup, and results analysis, underscoring the efficacy of our deep learning approach in antenna optimization.

In this paper, we investigated theoretically the hydrogenic donor impurity states in strained wurtzite (In,Ga)N-GaN coupled quantum wells (CQWs). The variational approach is employed to obtain the dependence on built-in electric field (BEF), hydrostatic pressure, indium composition, and structure size of the binding energy of hydrogenic donor impurity (BEHDI). The results reveal that hydrostatic pressure and structure size of the CQWs have a great influence on BEF which affects strongly the BEHDI. With the increment in hydrostatic pressure, the BEF strength of well and barrier layers enhances monotonously. However, by increasing the well width (barrier width), the BEF strength of well layer reduces (enhances) gradually, and that of barrier layers enhances (reduces). Meantime, it reveals that the binding energy (1) enhances linearly as the hydrostatic pressure is increased, (2) is more sensitive to geometrical parameters (width of well and/or barrier), and (3) demonstrates a maximum value as an impurity ion is shifted from one side of the CQWs to the other.

Perovskite solar cells (PSCs) are improving in efficiency, but their stability remains a challenge compared to other solar technologies due to the use of hybrid organic–inorganic materials. To overcome this, researchers have shifted focus from methylammonium-based PSCs to more stable cesium (Cs)-based PSCs. By optimizing multi-layer structures to enhance solar spectrum absorption, substantial performance improvements are possible. In this study, we explored single (CsPbIBr₂), dual (CsPbIBr₂/KSnI₃), and triple (CsPbIBr₂/KSnI₃/MASnBr₃) absorber layer designs. The optimization of bilayer and triple-layer PSCs takes into account various factors, such as absorber layer thickness, defect density, and interface defect density for each PSC type. Finally, using the optimal triple-absorber layer combination, we optimized the electron transport layer, hole transport layer, series resistance, and shunt resistance. In this research, we attained impressive efficiencies of 34.22% for the triple-layer solar cell, 20.41% for the bilayer solar cell, and 7.32% for the single-junction PSC. This design approach led to an optimal configuration that showed substantial improvements over the experimental benchmark, including a 7.08% increase in open circuit voltage, a 256.9% increase in short circuit current, a 22.32% increase in fill factor, and a 367.5% increase in efficiency. By meticulously aligning multiple absorber layers in perovskite solar cells, we can unlock new pathways to developing highly efficient solar cells for the future.

A photonic crystal fiber (PCF)-based multi-analyte refractive index sensor is introduced in this study for the detection of four waterborne pathogens: Vibrio cholerae, Bacillus anthracis, Escherichia coli, and Enterococcus faecalis. The sensor comprises a tri-core structure with hexagonal rings encased in a silica substrate. Two selective holes are infused with water samples, enabling concurrent detection of two analytes. The sensor integrates liquid-silica mode coupling as its sensing mechanism. The couplings are precisely estimated and numerically evaluated using a finite-element method (FEM)-based simulation tool. The optimization of the sensor’s structural characteristics resulted in wavelength sensitivity of 6386 nm/RIU, 7104 nm/RIU, 8510 nm/RIU, and 3409 nm/RIU for sample pairs of V. cholerae–pure water, V. cholerae–V. cholerae, V. cholerae–B. anthracis, and E. coli–V. cholerae, respectively. Furthermore, the sensor exhibits the highest wavelength resolution of (text {1.59} times text {10}^{-5}) RIU and figure of merit of 142 (text {RIU}^{-1}) and is also assessed for detection limit, detection accuracy, and signal-to-noise ratio. Featuring a straightforward design and remarkable sensing capabilities, the proposed sensor is anticipated to be exceptionally effective at detecting waterborne pathogens, with potential to excel in identifying chemicals, biomedical substances, and other diverse analytes.

Formulations to determine the electric field and the electrostatic potentials in a thermoelectric couple through solving the Poisson’s equation are introduced in this work. Analytical approximations of the auxiliary energies introduced in the author’s earlier work in the relaxation time approximation of the Boltzmann transport equation are developed based on the coupled equations of heat and electric current. These auxiliary energies are used in the Poisson’s equation at each temperature node along the thermoelectric leg to obtain a set of algebraic equations with the electric field and the electrostatic potentials as unknowns. The algebraic equations are then solved using the derived algorithm and the boundary conditions determined by the continuity and the carrier concentration equations.