In this article, machine learning (ML) metamodels have been developed in order to predict the per-unit-length parameters of hybrid copper–graphene on-chip interconnects based on their structural geometry and layout. ML within context article include artificial neural networks, support vector machines (SVMs), least-square SVMs. The salient feature all these is that they exploit prior knowledge p.u.l. obtained from cheap empirical models reduce number expensive full-wave electromagnetic (EM)...

This work presents a novel approach for the accurate estimation of multiple time-delays from frequency response distributed system. The proposed is based on powerful and flexible machine learning technique, namely, least-square support vector (LS-SVM). LS-SVM regression used to construct metamodel transfer function describing generic linear time-invariant system in delayed-rational form. Specifically, after some manipulation model precisely identifies dominant propagation delays original...

Nonlinear phenomena in electrical contacts deteriorate the quality of communication systems with production passive intermodulation (PIM). The theoretical evaluation PIM as a function physical parameters contact is rather complicated. Standard linear and macroscopic models do not take into account all microscopic aspects responsible for its nonlinear behavior. For above reason, an accurate analysis should be carried out by using models, defined dozens parameters, some which cannot precisely...

This paper presents a study on an induction planar actuator concept. The device uses the same principles as linear motor in which interaction between travelling magnetic field and conducting surface produces eddy currents that leads to generation of thrust force can result movement over metallic surface. benefit inspection surfaces based driving platform provided by actuator. Equations electric fields are presented and, means these equations, forces involved were calculated. behaviour normal...

This paper focuses on the development and experimental validation of a physical-based model for SpaceWire (SpW) cable (Variant 01) terminated by two micro D-type connectors. The SpW relies frequency-dependent per-unit-length parameters computed from available information its cross-section materials via 2D simulation. On other hand, connector is obtained simulation results provided 3D representation in CST MICROWAVE STUDIO. models are then combined together implemented HSPICE means their...

This paper focuses on the application of Least-Square Support Vector Machine (LS-SVM) regression for modeling frequency responses complex interconnect structures. The goal is to obtain a delayed-rational model (DRM) structure accounting multiple time-delays generated by wave propagation and reflections along channel. A novel approach estimation based LS-SVM introduced. delays are estimated using dual space formulation with an ad-hoc kernel that considers possible delay interval. results...

This paper presents the studies on an induction planar actuator. The device uses same principles as a linear motor, where interaction between traveling magnetic field and conducting surface produces eddy current, which leads to generation of force. However, actuator has fields in two perpendicular directions, combination these can produce movement any direction surface. movement, combined with inherent normal force over it moves, provides adhesive allow utilization this system for inspecting...

This paper describes the application of method separation variables and use Fourier series for solving Laplace's Poisson's equations on study analysis magnetic field produced in a linear motor with Halbach array. Equations predicting 2D flux density distribution air gap were developed. The model was validated by means finite element measurements carried prototype motor. Theoretical results helped understand behaviour to obtain values static propulsion force normal such machine.

Planar motion over a surface is commonly acquired using combination of rotatory motors and appropriate mechanical transmissions to convert the torque produced by motor into linear motion. Typically, assembly employs two motors: one responsible for in x-axis other, y-axis [1]. A planar actuator achieves this kind from only device. Several different types have been studied. These utilized actuator, such as switched reluctance actuators [2], permanent magnet DC devices [3, 4, 5], AC [1, 6]....

This paper deals with the development of a surrogate model for uncertainty quantification and stochastic analysis passive intermodulation (PIM) in an Aluminum-Aluminum contact based on least-squares support vector machine (LS-SVM) regression. Starting from small set training pairs collecting configuration un-certain parameters corresponding PIM level, LS-SVM allows to build closed-form approximation such non-linear relationship. Such model, can be suitably used within Monte Carlo (MC)...

This paper presents an automatic procedure for the optimization of hyperparameters a delay rational model approximating frequency-domain behavior highspeed interconnects. The proposed is built via kernel-based regression, such as Least-Square Support Vector Machine (LS-SVM), by considering ad-hoc kernel with two related to propagation delays introduced system. Such hyperparameters, along Tikhonov regularizer used LS-SVM are carefully tuned approach based on k-fold cross-validation and...

This paper presents an automatic procedure for the optimization of hyperparameters a delay rational model approximating frequency-domain behavior high-speed interconnects. The proposed is built via kernel-based regression, such as Least-Square Support Vector Machine (LS-SVM), by considering ad-hoc kernel with two related to propagation delays introduced system. Such hyperparameters, along Tikhonov regularizer used LS-SVM are carefully tuned approach based on k-fold cross-validation and...