Efficient Implementation of the Vector-Valued Kernel Ridge Regression for the Parametric Modeling of the Frequency-Response of a High-Speed Link

This paper presents an efficient implementation of the vector-output kernel Ridge regression (KRR). The proposed approach is applied to model the frequency-domain behavior of the magnitude of the transfer function of a parametric high-speed link as function of 11 parameters. The accuracy and the computational cost of the proposed technique are assessed on noisy samples and compared with the ones of a state-of-the-art modeling technique based on the combination of the principal components analysis (PCA) and the least-squares support vector machine (LS-SVM) regression.