Identfication of Main Factors of Uncertainty in a Microstrip Line Network

Abstract

This paper deals with uncertainty propagation applied to the analysis of crosstalk in printed circuit board microstrip traces. Complex interconnection networks generally are affected by many uncertain parameters and their point-to-point transfer functions are computationally expensive, thus making Monte-Carlo analyses rather inefficient. To overcome this situation, a metamodel is highly desirable. This paper presents a sparse and accelerated polynomial chaos approach, which proves to be well adapted for high-dimensional uncertainty quantification and well suited for the sensitivity analysis of crosstalk effects. We highlight the significant advantage of the advocated approach for the design of microstrip line networks of complex topology. In fact, we demonstrate how a small number of system simulations can help to quantify the statistics of the output variability and identify a reduced set of high-impact parameters.

Pce-Based Approach to Worst-Case Scenario Analysis in Wireless Telecommunication Systems

“…This method provides high accuracy in determining probability distributions of EM fields; however, it requires a lot of iterations to obtain satisfyingly accurate results. An alternative method to the Monte Carlo algorithm is the application of polynomial chaos expansion (PCE) which originates from the work of Norbert Wiener in 1938 [11] and since then was discussed in numerous articles and books, and widely used in scientific engineering, including electromagnetics, e.g., [12][13][14][15][16][17]. Our work, according to the best knowledge of the authors, is the first in which the total impact of uncertainties associated with antennas and the propagation channel on the distribution of a random electromagnetic field in a telecommunication system is analyzed.…”

Fast Uncertainty Quantification in Low Frequency Electromagnetic Problems by an Integral Equation Method Based on Hierarchical Matrix Compression

“…In many electromagnetic (EM) applications the values of the material parameters are affected by unavoidable uncertainties [1], [2]. Moreover, conductivity, permittivity, and permeability of media are often strongly influenced by technological uncertainties and external uncontrolled phenomena, such as temperature, pressure, humidity, and other environmental quantities [3].…”