Search

PIN diodes, due to their simple structure and variable resistance characteristics under high-frequency high-power excitation, are often used in radar front-end as limiters to filter high power microwaves (HPM) to prevent its power from entering the internal circuit and causing damage. This paper carries out theoretical derivation and research on the HPM effects of PIN diodes, and then uses an optimized neural network algorithm to replace traditional physical modeling to calculate and predict two types of HPM limiting indicators of PIN diode limiters. We proposes a neural network model for each of the following two prediction scenarios: in the scenario of time-junction temperature curves under different HPM irradiation, the weighted mean squared error (MSE) between the predicted values from the test dataset and the simulated values is below 0.004. While in predicting PIN limiter's power limitation threshold, insertion loss, and maximum isolation under different HPM irradiation, the MSE of the test set prediction values and simulation values are all less than 0.03. The method proposed in this research, which applies an optimized neural network algorithm to replace traditional physical modeling algorithms for studying the high-power microwave effects of PIN diode limiters, significantly improves the computational and simulation speed, reduces the calculation cost, and provides a new method for studying the high-power microwave effects of PIN diode limiters. [ABSTRACT FROM AUTHOR]

Single-event effects (SEE) are an important index of radiation resistance for fully depleted silicon on insulator (FDSOI) devices. The research into traditional FDSOI devices is based on simulation software, which is time consuming, requires a large amount of calculation, and has complex operations. In this paper, a prediction method for the SEE of FDSOI devices based on deep learning is proposed. The characterization parameters of SEE can be obtained quickly and accurately by inputting different particle incident conditions. The goodness of fit of the network curve for predicting drain transient current pulses can reach 0.996, and the accuracy of predicting the peak value of drain transient current and total collected charge can reach 94.00% and 96.95%, respectively. Compared with TCAD Sentaurus software, the simulation speed is increased by 5.10 × 102 and 1.38 × 103 times, respectively. This method can significantly reduce the computational cost, improve the simulation speed, and provide a new feasible method for the study of the single-event effect in FDSOI devices. [ABSTRACT FROM AUTHOR]

To develop an advanced deep convolutional neural network (DCNN) architecture to generate synthetic CT (SCT) images from MR images for intensity-modulated proton therapy (IMPT) treatment planning of nasopharyngeal cancer (NPC) patients. T1-weighted MR images and paired CT (PCT) images were obtained from 206 NPC patients. For each patient, deformable image registration was performed between MR and PCT images to create an MR-CT image pair. Thirty pairs were randomly chosen as the independent test set and the remaining 176 pairs (14 for validation and 162 for training) were used to build two conditional generative adversarial networks (GANs): 1) GAN3D: using a 3D U-net enhanced with residual connections and attentional mechanism as the generator and 2) GAN2D: using a 2D U-net as the generator. For each test patient, SCT images were generated using the generators with the MR images as input and were compared with respect to the corresponding PCT image. A clinical IMPT plan was created and optimized on the PCT image. The dose was recalculated on the SCT images and compared with the one calculated on the PCT image. The mean absolute errors (MAEs) between the PCT and SCT, within the body, were (64.89 ± 5.31) HU and (64.31 ± 4.61) HU for the GAN2D and GAN3D. Within the high-density bone (HU > 600), the GAN3D achieved a smaller MAE compared with the GAN2D (p < 0.001). Within the body, the absolute point dose deviation was reduced from (0.58 ± 1.61) Gy for the GAN2D to (0.47 ± 0.94) Gy for the GAN3D. The (3 mm/3%) gamma passing rates were above 97.32% for all SCT images. The SCT images generated using GANs achieved clinical acceptable dosimetric accuracy for IMPT of NPC patients. Using advanced DCNN architecture design, such as residual connections and attention mechanism, SCT image quality was further improved and resulted in a small dosimetric improvement. [ABSTRACT FROM AUTHOR]

Fully depleted Silicon on insulator technology (FDSOI) is proposed to solve the various non-ideal effects when the process size of integrated circuits is reduced to 45 nm. The research of traditional FDSOI devices is mostly based on simulation software, which requires a lot of calculation and takes a long time. In this paper, a deep learning (DL) based electrical characteristic prediction method for FDSOI devices is proposed. DL algorithm is used to train the simulation data and establish the relationship between the physical parameters and electrical characteristics of the device. The network structure used in the experiment has high prediction accuracy. The mean square error of electrical parameters and transfer characteristic curve is only 4.34 × 10â€"4 and 2.44 × 10â€"3 respectively. This method can quickly and accurately predict the electrical characteristics of FDSOI devices without microelectronic expertise. In addition, this method can be extended to study the effects of various physical variables on device performance, which provides a new research method for the field of microelectronics. [ABSTRACT FROM AUTHOR]

The tunnel field‐effect transistor (TFET) is considered to be a suitable substitute for metal oxide semiconductor‐field effect transistors in the post‐"Moore's Law" era owing to its low power consumption. However, Si‐TFETs face the drawbacks of low on‐state currents and significant ambipolar leakage. This study proposes a GeSi/Si heterojunction double‐gate TFET with a T‐channel hetero‐gate dielectric (HJ‐HGD‐DGTFET) structure to overcome these problems. It also presents a novel method of predicting and optimizing the performance of the existing TFETs which use deep learning to accelerate the device design. Furthermore, this study proposes a neural network based on different requirements to perform two functions: prediction of the device performance using the forward design, and the forecast of the device structure using the inverse design. It can thus be used to determine whether the output of the network meets the design objectives and if it is necessary to change the output by adjusting the input, and lastly achieve the TFET performance prediction and device optimization. The proposed method can be used to design TFETs accurately and efficiently even without professional knowledge. This study provides guidance for the design and optimization of TFETs along with other microelectronic devices. [ABSTRACT FROM AUTHOR]

Purpose: Clinical implementation of magnetic resonance imaging (MRI)‐only radiotherapy requires a method to derive synthetic CT image (S‐CT) for dose calculation. This study investigated the feasibility of building a deep convolutional neural network for MRI‐based S‐CT generation and evaluated the dosimetric accuracy on prostate IMRT planning. Methods: A paired CT and T2‐weighted MR images were acquired from each of 51 prostate cancer patients. Fifteen pairs were randomly chosen as tested set and the remaining 36 pairs as training set. The training subjects were augmented by applying artificial deformations and feed to a two‐dimensional U‐net which contains 23 convolutional layers and 25.29 million trainable parameters. The U‐net represents a nonlinear function with input an MR slice and output the corresponding S‐CT slice. The mean absolute error (MAE) of Hounsfield unit (HU) between the true CT and S‐CT images was used to evaluate the HU estimation accuracy. IMRT plans with dose 79.2 Gy prescribed to the PTV were applied using the true CT images. The true CT images then were replaced by the S‐CT images and the dose matrices were recalculated on the same plan and compared to the one obtained from the true CT using gamma index analysis and absolute point dose discrepancy. Results: The U‐net was trained from scratch in 58.67 h using a GP100‐GPU. The computation time for generating a new S‐CT volume image was 3.84–7.65 s. Within body, the (mean ± SD) of MAE was (29.96 ± 4.87) HU. The 1%/1 mm and 2%/2 mm gamma pass rates were over 98.03% and 99.36% respectively. The DVH parameters discrepancy was less than 0.87% and the maximum point dose discrepancy within PTV was less than 1.01% respect to the prescription. Conclusion: The U‐net can generate S‐CT images from conventional MR image within seconds with high dosimetric accuracy for prostate IMRT plan. [ABSTRACT FROM AUTHOR]