G. Angiulli

Microwave Devices and Antennas Modelling by Support Vector Regression Machines

Development of fast and accurate models of microwave devices and antennas is of paramount importance in computer-aided design and circuit analysis. At this purpose, artificial neural networks (ANNs) have been extensively exploited in technical literature. However, in the last years support vector machines (SVMs) developed by Vapnik are gaining popularity due to many attractive features capable to overcome the limitations connected to ANNs. In this work, support vector regression machines (SVRMs) modelling performances are investigated and compared with ANNs performances by means of several cases of study

Resonant frequency evaluation of microstrip antennas using a neural-fuzzy approach

The accurate evaluation of the resonant frequency of microstrip antennas is a key factor to guarantee their correct behavior. To this aim, the Method of Moments technique is currently employed. A fast technique to evaluate the resonant frequency of microstrip antennas using neuro-fuzzy networks, is proposed. Numerical results obtained by using this technique agree quite well with the Method of Moments results. The proposed technique can be fruitfully used in microwave CAD applications.

An improved synthesis algorithm for reflectarrays design

An improved synthesis procedure is presented in this letter for the design of microstrip reflectarray antennas. The iterative projection based algorithm is modified to include the effect of different incidence angles on the reflectarray elements in the evaluation of the design curve. Experimental results on an X-band prototype of unequal microstrip patch elements are reported.

Resonant Frequencies of Circular Substrate Integrated Resonators

In this letter, the characterization of circular resonators realized in substrate integrated waveguide technology is presented. Resonances are found by setting up the equations of the scattering for an ensemble of metallic posts in parallel plates waveguide and considering the nontrivial solution to the corresponding homogeneous problem. Resonant frequencies are then located by means of an efficient strategy, based on an estimate of the minimum singular value. In the last part of the letter a data fit formula of practical use is presented for circular resonators with a wide range of geometrical and electrical parameters resonating on TM010 and TM110 modes.

On the Computation of Nonlinear Eigenvalues in Electromagnetic Problems

The numerical evaluation of parameters having practical interest in electromagnetic applications is often related to the determination of the set of values, named nonlinear eigenvalues, for which the nonlinear eigenvalue problem L(γ)f = 0 admits a non trivial solution. These values are usually computed, after properly discretization of L(γ), by SVD. However this approach has a main drawback that is very time consuming. In this work we propose a method for quickly evaluating these parameters. An application of the developed technique to a well-known electromagnetic problem is then presented and discussed.

SAR imagery classification using multi-class support vector machines

In this paper, we present the application to SAR imagery classification of a novel pattern recognition technique named Multi-class Support Vector Machines (M-SVMs). M-SVMs are a n-ary extension of Support Vector Machines (SVM), introduced by Vapnik within the framework of the Statistical Learning Theory. In this article we use the M-SVMs in order to classify an ERS-1 SAR multi-frequency survey of Torre de Hercules coast, Spain (December 13, 1992). The main objective of this work is to evaluate the classification performances of M-SVMs in comparison with the most frequently employed Neural Networks and Fuzzy classifiers. M-SVMs provided interesting results with respect to Neural Networks and Fuzzy classifiers, having a reliability factor around to 94%.

Fast Nonlinear Eigenvalues Analysis of Arbitrarily Shaped Substrate Integrated Waveguide (SIW) Resonators

In this paper, the problem of evaluating resonant frequencies of arbitrarily shaped substrate integrated waveguide resonators is considered. Resonances are found regarding them as nonlinear eigenvalues of the operator describing the scattering by metallic posts in a parallel plate waveguide. Nonlinear eigenvalues are usually computed by singular value decomposition. In order to save computational time, we propose an alternative approach derived by some general consideration on the theory of nonlinear eigenvalue problems in the discrete case. Results on rectangular, circular, and elliptical cavities are presented and compared with High Frequency Structure Simulator software package simulations. It is shown that the proposed method is efficient and accurate.

Support Vector Regression Machines to Evaluate Resonant Frequency of Elliptic Substrate Integrate Waveguide Resonators

In this paper an efficient technique for the determination of the resonances of elliptic Substrate Integrated Waveguide (SIW) resonators is presented. The method is based on the implementation of Support Vector Regression Machines trained using a fast algorithm for the computation of the resonant frequencies of SIW structures. Results for resonators with a wide range of parameters will be presented. A comparison with results obtained with Multi Layer Perceptron Artificial Neural Network and with full wave simulations will show the effectiveness of the proposed approach.

A Neuro-Fuzzy Network for the Design of Circular and Triangular Equilateral Microstrip Antennas

A fast method for the design of circular and equilateral triangular microstrip antennas, based on a Neuro-Fuzzy Network, is presented. Numerical results obtained by using this technique agree quite well with the Moment Method results. The proposed technique can be fruitfully used in microwave and millimeter wave CAD applications.

COMPUTATION OF THE RESONANT FREQUENCY AND QUALITY FACTOR OF LOSSY SUBSTRATE INTEGRATED WAVEGUIDE RESONATORS BY METHOD OF MOMENTS

This paper presents a technique for the e-cient and accurate determination of resonant frequencies and quality factors of Substrate Integrated Waveguide (SIW) resonators. To consider resonators of a general shape the SIW structure is modelled as a parallel plate waveguide populated with metalized via holes. The fleld into the SIW cavity is found by solving the scattering problem for the set of vias into the parallel plate. Resonances are determined searching for the complex frequencies for which the determinant of the system of equations pertinent to the scattering is zero. To speed up the search, a flrst guess for the resonance frequency is found using an estimate of the minimum singular value of the system of equations. A Muller search in the complex plane is later used to accurately determine both frequencies and quality factors. Results relevant to resonators of various shapes are presented and compared with results obtained with a commercial code.