As the geometries of transistors continue to scale down, single-event transient (SET) are becoming a major source of soft errors in circuit design. This paper proposes a machine learning method for modeling SET in cir...
As the geometries of transistors continue to scale down, single-event transient (SET) are becoming a major source of soft errors in circuit design. This paper proposes a machine learning method for modeling SET in circuit-level simulations with the high efficiency, since traditional models such as the double exponential model are not coupled with circuit responses. The proposed artificial neural network (ANN) model shows a great prediction accuracy for single-event transient currents in 12nm FinFET technology under different time, linear energy transfer (LET), bias voltage, and strike positions. The trained model can be implemented into SPICE with Verilog-A for circuit-level simulation, and its validity in a 7-stage inverter chain and a 6T SRAM are demonstrated to provide the rapid guidance for radiation reliability.
Reliability is one of the major concerns for SiC devices. A proper edge termination structure would enable SiC devices to operate at designed voltages because it mitigates the electric field aggregation caused by the ...
详细信息
ISBN:
(数字)9798331528850
ISBN:
(纸本)9798331528867
Reliability is one of the major concerns for SiC devices. A proper edge termination structure would enable SiC devices to operate at designed voltages because it mitigates the electric field aggregation caused by the cylindrical or spherical junctions. However, a great number of research has been reported on studying the optimization of edge terminations to achieve higher breakdown voltage (BV). In contrast, research on the impact of edge termination design on devices' reliability remains limited. In this work, floating field ring (FFR), single-zone junction termination extension (SZ-JTE), ring-assisted JTE (RA-JTE), multi-floating zone JTE (MFZ-JTE), and Hybrid-JTE are studied. Their reliability performance is evaluated by comparing the BV stability of these termination designs under different implantation doses, critical dimension (CD) deviations, and fixed SiC/SiO 2 interface charge through technology computer-aided design (TCAD) simulations. Among all the edge terminations listed, Hybrid-JTE has exhibited a champion reliability.
This paper proposes a simplified channel modeling method for the multilayer PCB with nonhomogeneous substrate filling using neural network. The neural network model is used to obtain a simplified relationship between ...
详细信息
Chip-to-wafer interconnect approach is to produce denser and smaller dies, while improving inter-chip bandwidth and power dissipation. In this work a vertical system integration is done by copper pillar and solder int...
Chip-to-wafer interconnect approach is to produce denser and smaller dies, while improving inter-chip bandwidth and power dissipation. In this work a vertical system integration is done by copper pillar and solder interconnects. Three-dimensional (3D) integration technology uses chip-to-wafer bonding to achieve effective chip integration. By structural simulation and modeling a reliable package selection is made from different package designs, epoxy mold compounds (EMC), polymer dielectric (PD) materials and different other package parameters. In this simulation and modeling an approach has been taken to model two levels of solder interconnects namely copper pillar solder micro-bump and BGA (ball grid array) solder in the same FEA (finite element analysis) model. FEA mesh densities in 1 st and 2 nd interconnects are joined by contact pair definition. Parametric studies are done for three different EMC and three different PD materials. 1 st level solder interconnect or micro-bump temperature cycling (TC) reliability life is significantly higher than 2 nd level solder interconnect due to presence of EMC around the 1 st level interconnect. High CTE (coefficient of thermal expansion) of EMC shows very poor solder life for 1 st level solder interconnect. PD material CTE shows significant impact on component side solder life for 2 nd level interconnect. Also, higher PD material CTE shows worsening 1 st level interconnect life. Stress analysis shows pad diameter smaller than UBM (under bump metallization) opening is a better design.
New second order high-pass filter of the Sallen-Key family circuit is considered, where there is unrelated tune of generic parameters by a digitally controlled resistor - the pole frequency, Q-factor pole and the scal...
详细信息
This paper studies the radial basis function (RBF) neural network algorithm, applying big data analysis to the latest developments in wireless communication and channel modeling. A channel model based on RBF neural ne...
详细信息
The W-band frequency range has garnered considerable attention among researchers, necessitating the development of efficient frequency doubling techniques. This paper introduces a novel W-band broadband frequency doub...
The W-band frequency range has garnered considerable attention among researchers, necessitating the development of efficient frequency doubling techniques. This paper introduces a novel W-band broadband frequency doubler employing a Schottky diode-based approach. A varactor diode, specifically the M4AE1310 model, serves as the crucial nonlinear element in the frequency doubler design. To streamline the model structure, the biased DC ground structure is eliminated. The frequency doubler comprises a waveguide-microstrip probe, a ground-matched direct, a CMRC (compact microstrip resonant cell) low-pass filter, and an input-output matching section. The modelingsimulation is conducted using a combined ADS and HFSS. The simulation results demonstrate exceptional frequency doubling efficiency, yielding an output power of approximately 10 dBm when driven with a power of 20 dBm within the W-band frequency range.
This article delves into the application and efficacy of planar loop sensors for detecting partial discharge (PD) phenomena. The preference for loop sensors in PD detection stems from their simple design, non-destruct...
详细信息
ISBN:
(数字)9798350308976
ISBN:
(纸本)9798350308983
This article delves into the application and efficacy of planar loop sensors for detecting partial discharge (PD) phenomena. The preference for loop sensors in PD detection stems from their simple design, non-destructive testing feature, and contactless characteristics. The paper explores both experimental and finite element modeling (FEM) approaches to evaluate the performance of distinct planar square spiral sensors in PD detection. The primary objective of the study is to compare inductive loops with loop antennas by varying the geometric parameters to achieve each approach. Two single-turn square loop antennas were designed, with a circumference of 30 cm and 75 cm, and compared with four multi-turn square inductive loops with an exterior side of 31.2 mm and varying geometric parameters. The figures of merit of all sensors, such as inductance, quality factor, and resonant frequency, are extracted through measurement and simulation to identify the most influential parameter that maximizes the sensing capability toward PD.
Convolutional Neural Network (CNN), a feedforward neural network, is widely used in large-scale machine learning tasks such as image recognition, image classification, and object detection. Although CNN has strong per...
Convolutional Neural Network (CNN), a feedforward neural network, is widely used in large-scale machine learning tasks such as image recognition, image classification, and object detection. Although CNN has strong performance, its application in daily embedded systems is limited due to the complex network hierarchy and a large amount of data. In this paper, the LeNet-5 network structure is optimized. The recognition accuracy of the optimized network on the MNIST dataset is 98.32%, but the number of weights is reduced by more than ten times compared with the LeNet-5 network, and data quantization is used to convert the weights into 8-bit fixed-point numbers, which is more suitable for deployment in the hardware with limited resources and power consumption. At the same time, different hardware structures and optimization strategies are designed for each layer of the network, and the convolution, pooling and fully connected layers are computed in parallel and optimized using methods such as loop unrolling and pipelining. Finally, the feasibility of the hardware design is verified by simulating each layer using the MNIST dataset and quantized weights.
This paper describes the design and simulation of a small volume, low insertion loss, and high isolation Band1 SAW duplexer. Compared to mainstream Band1 SAW Duplexer, for example Murata SAYEY1G95HA0F0A, our simulatio...
详细信息
ISBN:
(纸本)9798350320817
This paper describes the design and simulation of a small volume, low insertion loss, and high isolation Band1 SAW duplexer. Compared to mainstream Band1 SAW Duplexer, for example Murata SAYEY1G95HA0F0A, our simulation results show that the design enables a package size reduction of ~25%, with the length and width shortened by 200 microns respectively. The design has a similar 1.8dB insertion loss and a superior 60dB isolation. Exacerbation of internal electromagnetic coupling on the performance caused by the reduction of package size is successfully resolved with iterative optimization of the PCB layer and main circuit structure. This compact Band1 SAW duplexer design has the potential to meet the more and more stringent size requirement of radio frequency communication in future mobile devices.
暂无评论