Support vector design of the microstrip lines

Abstract

In this article, the support vector regression is adapted to the analysis and synthesis of microstrip lines on all isotropic/anisotropic dielectric materials, which is a novel technique based on the rigorous mathematical fundamentals and the most competitive technique to the popular artificial neural networks (ANN). In this design process, accuracy, computational efficiency and number of support vectors are investigated in detail and the support vector regression performance is compared with an ANN performance. It can be concluded that the ANN may be replaced by the support vector machines in the regression applications because of its higher approximation capability and much faster convergence rate with the sparse solution technique. Synthesis is achieved by utilizing the analysis blackbox bidirectionally by reverse training. Furthermore, by using the adaptive step size, a much faster convergence rate is obtained in the reverse training. Besides, design of microstrip lines on the most commonly used isotropic/anisotropic dielectric materials are given as the worked examples.

Prediction of slot‐position and slot‐size of a microstrip antenna using support vector regression

“…In most recent years, the support vector regression (SVR) modeling has been admired in optimizing the engineering problems because of its several attractive features . ANNs utilize empirical risk minimization method whereas SVRs use structural risk minimization technique . ANNs may have multiple local minima problem whereas SVRs have a distinctive solution and SVR find a solution of constrained quadratic optimization problem which is on the basis of small‐sample statistical learning theory .…”

“…A rectangular microstrip antenna has been designed for the computation of the resonant frequency, bandwidth as well as the input impedance with the help of SVM formulation . The SVR analysis as well as synthesis have been done for microstrip lines on various dielectric materials . A knowledge based SVR method has been applied in order to synthesize the transmission lines .…”

“…15 ANNs utilize empirical risk minimization method whereas SVRs use structural risk minimization technique. 16 ANNs may have multiple local minima problem whereas SVRs have a distinctive solution and SVR find a solution of constrained quadratic optimization problem which is on the basis of small-sample statistical learning theory. 17 ANNs are more complex than SVRs which result over fitting problem in statistical models.…”

“…Thus, SVR is capable of finding the solution of various types of problems successfully [15][16][17][18][19][20][21][22][23][24][25] but unlikely, the available literature of SVR formulation in electromagnetic and antennas problems is really restricted. [15][16][17][18] A rectangular microstrip antenna has been designed for the computation of the resonant frequency, bandwidth as well as the input impedance with the help of SVM formulation. 15 The SVR analysis as well as synthesis have been done for microstrip lines on various dielectric materials.…”

Knowledge-Based Support Vector Synthesis of the Microstrip Lines

“…Furthermore, SVM is based on small sample statistical learning theory, whose optimum solution is based on limited samples instead of infinite sample that ensures enormous computational advantages. Typical applications of the SVRM to the microwave modeling can be found in [4][5][6].…”