EM-based microwave circuit design using fuzzy logic techniques

IEEE Trans. Microwave Theory Techn.

Bandler

2007

Abstract-We present a novel surrogate modeling methodology based on a combination of space mapping and fuzzy systems. Fine model data, the so-called base set, is assumed available in the region of interest. Although we do not assume any particular location of the base points, it is preferable that they form a uniform mesh. The standard space-mapping surrogate is established using available fine model data. The fuzzy system is then set up to interpolate the differences between the space-mapping surrogate and the fine model at all base points. Our new methodology offers significant advantages with respect to some of the previous space-mapping approaches to modeling, which are: 1) it handles any base set and 2) the number of space-mapping parameters does not limit the accuracy of the surrogate. Moreover, it exhibits comparable or better accuracy than the recently published modeling technique utilizing space mapping and radial basis functions. We also consider a hierarchical fuzzy space-mapping modeling, which relies on a fuzzy interpolation of space-mapping parameters and subsequent fuzzy interpolation of the residuals between the fine and surrogate model. Examples demonstrate the robustness of our approach and give a comparison with other space-mapping-based modeling techniques.Index Terms-Computer-aided design (CAD), electromagnetic (EM) modeling, fuzzy systems, microwave circuits, space mapping, surrogate modeling.

Section: Fuzzy Space-mapping Surrogate Modelingmentioning

confidence: 99%

A Space-Mapping Approach to Microwave Device Modeling Exploiting Fuzzy Systems

IEEE Trans. Microwave Theory Techn.

Bandler

2007

“…Implementation of low-cost antenna models is possible using various approximation techniques such as polynomial regression [1], radial basis function interpolation [2], Kriging [3], [4] support vector regression [5]- [8], fuzzy systems [9], [10] multidimensional Cauchy approximation [11], or artificial neural networks [12]- [15]. A common problem related to all of these methods is high model setup cost: in order to ensure usable accuracy a large number of training points is necessary, which quickly grows with the dimensionality of the design space (a problem referred to as the curse of dimensionality) [1], [16].…”

Section: Introductionmentioning

confidence: 99%

Cost-efficient modeling of antenna structures using Gradient-Enhanced Kriging

Ulaganathan

2015 Loughborough Antennas &Amp; Propagation Conference (LAPC)

Bekasiewicz

et al. 2015

Abstract-Reliable yet fast surrogate models are indispensable in the design of contemporary antenna structures. Data-driven models, e.g., based on Gaussian Processes or support-vector regression, offer sufficient flexibility and speed, however, their setup cost is large and grows very quickly with the dimensionality of the design space. In this paper, we propose cost-efficient modeling of antenna structures using GradientEnhanced Kriging. In our approach, the training data set contains, apart from the EM-simulation responses of the structure at hand, also derivative data at the respective training locations obtained at little extra cost using adjoint sensitivity techniques. We demonstrate that introduction of the derivative information into the model allows for considerable reduction of the model setup cost (in terms of the number of training points required) without compromising its predictive power. The Gradient-Enhanced Kriging technique is illustrated using a dielectric resonator antenna structure. Comparison with conventional Kriging interpolation is also provided.

“…Therefore, accurate and computationally cheap antenna models are indispensable. Low-cost antenna models can be implemented using various approximation techniques, such as polynomial regression [5], radial basis functions [6], Kriging [7], [8], support vector regression [9]- [12], fuzzy systems [13], [14], artificial neural networks [15]- [18], or multidimensional Cauchy approximation [19]. Unfortunately, in order to ensure good accuracy over the entire design space, all of these techniques require a large number of training points.…”

Section: Introductionmentioning

confidence: 99%

Variable-Fidelity Electromagnetic Simulations and Co-Kriging for Accurate Modeling of Antennas

IEEE Trans. Antennas Propagat.

Ogurtsov

Couckuyt

et al. 2013

Abstract-Accurate and fast models are indispensable in contemporary antenna design. In this paper, we describe the low-cost antenna modeling methodology involving variable-fidelity electromagnetic (EM) simulations and co-Kriging. Our approach exploits sparsely sampled accurate (high-fidelity) EM data as well as densely sampled coarse-discretization (low-fidelity) EM simulations that are accommodated into one model using the co-Kriging technique. By using coarse-discretization simulations, the computational cost of creating the antenna model is greatly reduced compared to conventional approaches, where high-fidelity simulations are directly used to set up the model. At the same time, the modeling accuracy is not compromised. The proposed technique is demonstrated using three examples of antenna structures. Comparisons with conventional modeling based on high-fidelity data approximation, as well as applications for antenna design, are also discussed.