This paper presents a prior knowledge based support vector regression modeling method to characterize the RF performance of the low temperature co-fired ceramic (LTCC) structure. A coarse surrogate is formed by multidimensional Cauchy approximation as the prior knowledge to improve the accuracy of modeling. 3D LTCC based vertical interconnection model is developed as an example using the proposed method. Experimental results show that the developed SVR model perform a better predictive ability.
Computationally Efficient Multi-Fidelity Bayesian Support Vector Regression Modeling of Planar Antenna Input Characteristics