Search

The design level of channel physical characteristics has a crucial influence on the transmission quality of high-speed serial links. However, channel design requires a complex simulation and verification process. In this paper, a cascade neural network model constructed of a Deep Neural Network (DNN) and a Transformer is proposed. This model takes physical features as inputs and imports a Single-Bit Response (SBR) as a connection, which is enhanced through predicting frequency characteristics and equalizer parameters. At the same time, signal integrity (SI) analysis and link optimization are achieved by predicting eye diagrams and channel operating margins (COMs). Additionally, Bayesian optimization based on the Gaussian process (GP) is employed for hyperparameter optimization (HPO). The results show that the DNN–Transformer cascaded model achieves high-precision predictions of multiple metrics in performance prediction and optimization, and the maximum relative error of the test-set results is less than 2% under the equalizer architecture of a 3-taps TX FFE, an RX CTLE with dual DC gain, and a 12-taps RX DFE, which is more powerful than other deep learning models in terms of prediction ability. [ABSTRACT FROM AUTHOR]

High-speed flows over a 34° compression corner subjected to several types of applied electromagnetic fields are numerically investigated. A coupled model for the turbulent flow field and the externally applied electromagnetic field is established based on the low magnetic Reynolds number assumption. The advection upstream splitting method with pressure-based weight function (AUSMPW) scheme and the lower upper symmetric Gauss Seidel (LUSGS) method are implemented to solve turbulent magnetohydrodynamic (MHD) flow equations. Numerical results demonstrate that the performance of MHD separation flow control is determined mainly by the Lorentz force in the turbulent boundary layer. For the counter-flow Lorentz force, the velocity fluid in the boundary layer can be retarded, and a significant temperature increase was observed. The decelerating Lorentz force significantly brings negative effects on the turbulent skin friction coefficients and increases the static pressure locally. Furthermore, the position of the MHD zone has a significant impact on the control efficiency of the ramp-induced separation. With an external electric and magnetic field applied, the low-velocity fluid in the boundary layer can be accelerated, and the Lorentz force directed along the stream can reduce the size of the separation bubble. [ABSTRACT FROM AUTHOR]