International Journal of Numerical Modelling-Electronic Networks Devices and Fields最新文献

This study presents an innovative approach leveraging TCAD simulations and a Convolutional Artificial Neural Network (C-ANN) to address challenges in VLSI design. A statistical sample of 4000 distinct values was simulated to predict drain current (I_ds), achieving a dramatic reduction in runtime from 46 to 48 days (conventional TCAD) to just 100–120 s using the proposed ML-based C-ANN. The proposed gate-stacking SiO₂ + HfO₂ FE-MOSFET device demonstrates significant advancements, including reductions in short-channel effects (SCEs), subthreshold swing (SS) by 3.12%–4.04%, and drain-induced barrier lowering (DIBL) by 10.19%. Enhanced performance metrics include 52.95% higher I_ON, 90% reduced gate leakage, and improved transconductance g_m, transconductance generation function (TGF), early voltage (V_EA), and intrinsic gain (A_v) by 26.18%, 27.12%, 29.35%, and 101.24%, respectively. RF parameters such as gate capacitance (C_gg), unity gain frequency (f_t), and gain frequency product (GFP) improved by 34.53%, 48.74%, and 21.18%, making this device ideal for high-speed switching and RF applications, promoting efficiency in low-power VLSI designs.

In this paper, we investigate a nonlinear Schrödinger equation that includes a combination of Kerr law nonlinearity and weak nonlocality. The model includes linear and nonlinear dispersion and has several applications in nonlinear optics and optical fibers. We retrieve bright, dark, and singular soliton solutions for this system. We implement a variety of powerful schemes to derive this variety of optical soliton solutions. Moreover, we derive more solutions of distinct structures that include periodic and exponential solutions.

The magnetic dipole serves as a fundamental concept in understanding electromagnetic phenomena. It has extensive applications across various fields such as geophysics and indoor navigation, which require accurate determination of its magnetic field. Although the magnetic dipole approximation yields satisfactory results in the far field, its computational accuracy is poor in the near-field region. Here, we propose a method of accurately calculating the magnetic dipole field in both the near and far fields. This method encompasses three steps: first, calculating the magnetic field strength B_T at the position r; second, determining the direction of the magnetic field at r; and third, calculating three components of the magnetic field. Numerical tests show that the calculation error of B_T is < 1% at r > 1.2 R, and is < 0.1% at r > 10 R, where R is the radius of the magnetic dipole. Additionally, the magnetic field direction can be precisely modeled via multi-parameter fitting, yielding angular errors < 0.1° in most regions at r > 1.2 R. Integration of the direction and B_T enables us to accurately calculate three components of the magnetic field with an error of < 1% at r > 1.8 R. These results indicate that our method is able to achieve high accurate calculation of the magnetic dipole field in both the near and far fields. This method can provide an effective computational algorithm for the applications relying on magnetic dipoles.

Dynamic updating technique for initial structure in pixel antenna design and optimization is proposed. The conventional approach to pixel antenna design employs a fixed initial pixel structure set at the start of the entire process, while rarely studying the setting of the initial structure; therefore, the performance potential is not fully exploited. The proposed approach adaptively updates the initial structure to enhance the performance of the pixel antenna design, aiming to find the optimal initial pixel structure that achieves miniaturization and broadband capabilities. In general, the design procedure starts with an initial structure with relatively big element size and small overall size, then gradually reduces the element size and expands the overall size of the pixel area. A two-port pixel antenna is used as a design example to validate the proposed updating technique. The goal was to design a dual-port pixel antenna operating in the band of 2.4–3.2 GHz, using a miniaturized size. After two rounds of updates, the obtained −10 dB impedance bandwidths increased from 0.44 GHz (2.47–2.91GHz) to 0.75 GHz (2.45–3.20 GHz) and to 1.07 GHz (2.35–3.42 GHz), while having isolation better than −15 dB. The statistical results of 10 optimization runs for 3 initial structures also showed the performance enhancement of each updated initial structure. The proposed updated technology can be applied to other types of pixel antenna designs, with different design specifications.

This study is concerned with treating the fractional integral equations with functional kernels and variable delays, introducing a hyperaccurate semi–analytical method based on the Stieltjes–Wigert polynomials, matrix expansions, and the Laplace transform. After analytically converting the terms in the governing equation into the matrix expansions of the Stieltjes–Wigert polynomials type at the collocation points, the method gathers these matrices into a unique matrix equation and then readily solves it by an elimination technique. The residual improvement technique is also introduced to correct the obtained solutions. The residual error bound analysis is theoretically proved via algebraical properties and the mean value theorem for fractional integral calculus, respectively. Six model equations are treated via the method, which runs on a devised computer program. Based on the outcomes, the method is straightforward to treat model equations and to encode its mainframe on a mathematical software.

This article focuses on dynamic parameter identification for industrial robots and proposes a parameter identification method based on an improved excitation trajectory. First, a complex nonlinear friction model is adopted and modified according to joint friction characteristics, with a genetic algorithm utilized to determine its six parameters. Second, a weighted optimal excitation trajectory is designed to address nonlinear friction requirements and smooth operation constraints. Then, a global parameter optimization algorithm based on the least squares method and the modified sparrow search algorithm is proposed. Finally, the proposed method is validated on a self-developed six-axis industrial robot. Experimental results demonstrate that the proposed method achieves higher identification accuracy compared with two representative identification approaches.

Bias temperature instability (BTI) is a time-based degradation mechanism that causes serious damage to the performance of analog and digital integrated circuits. The increasingly probabilistic nature of this phenomenon renders machine learning-based modeling approaches more advantageous, as they can deliver more accurate results in that context compared to analytical methods. In this paper, the Long Short-Term Memory (LSTM) method, a time-series approach, has been adopted to model BTI in 40 nm CMOS p-type metal-oxide-semiconductor field-effect transistors (MOSFETs). The aging model has been established by training the experimental data collected from a dedicated test chip. A bi-directional LSTM structure has been employed in model generation. Mean-square error (MSE) results indicate that the model can be effectively utilized in interpolation exercises where the test data falls within the same interval as the training data, with great accuracy. Moreover, the model has yielded promising outcomes in extrapolation exercises where the test data lies outside the defined training range. This property potentially qualifies the proposed approach for time-to-market and cost-reduction efforts.

In this paper, a gap coupled ring-shape multi-band antenna of quad-band characteristics has been designed. The geometry of the proposed gap coupled ring-shape antenna is obtained by loading two horizontal and six vertical strips of the same width and the same gap. The quad band characteristic of the proposed antenna is obtained at frequencies 2.48, 3.85, 5.14, and 5.61 GHz. The quad band lies within frequency ranges 2.34–2.59 GHz (first band), 3.56–4.06 GHz (second band), 5.09–5.21 GHz (third band) and 5.47–5.96 GHz (fourth band). The second and fourth bands of the antenna arise due to outer vertical strips, while the first and third bands of the antenna are due to middle and inner vertical strips, respectively. The return loss of −17.31, −29.18, −18.40, and − 19.19 dB is obtained at frequencies 2.48, 3.85, 5.14, and 5.61 GHz, respectively. Additionally, a parametric analysis is also conducted to enhance the performance of the proposed gap coupled antenna. To validate the geometry of the proposed antenna, experimental measurement is performed with the prototype antenna. The first band (2.34–2.59 GHz) of the antenna covers WLAN, ISM band, and WiMAX; the second band (3.56–4.06 GHz) of the antenna covers 5G; the third band (5.09–5.21 GHz) of the antenna covers WLAN; and the fourth band (5.47–5.96 GHz) of the antenna covers WiMAX and WLAN applications. The proposed quad band antenna is a cost-effective and efficient solution for multi-band communication devices. It eliminates the need for multiple antennas and lowers hardware costs.