Andrea Randazzo

Optimization of the difference patterns for monopulse antennas by a hybrid real/integer-coded differential evolution method

The optimization of difference patterns of monopulse antennas is considered. The synthesis problem is recast as an optimization problem by defining a suitable cost function. In particular, in this paper, the cost function is based on constraints on the side-lobe levels. A subarray configuration is adopted and the excitations of the difference pattern are approximately determined. The optimization problem is efficiently solved by a differential evolution algorithm, which is able to contemporarily handle real and integer unknowns. Numerical results are reported concerning classical array configurations previously considered in the literature.

An inexact-Newton method for short-range microwave imaging within the second-order Born approximation

A new approach to noninvasive inspection of dielectric targets at microwave frequencies is proposed. Cylindrical dielectric objects are reconstructed under the second-order Born approximation. A multi-illumination configuration is considered. The continuous model is discretized by the moment method and an efficient inexact-Newton method is applied. The dielectric profile is iteratively reconstructed starting from the measured scattered data, which are related to the unknown target through the inverse scattering equations written in a variational setting. Several numerical results are reported, which are aimed at assessing the capabilities of the approach in dealing with the nonlinear ill-posed inverse problem associated to the short-range microwave imaging. Single, multilayer, and separate cylinders are reconstructed in noiseless and noisy environments.

An Inexact Newton Method for Microwave Reconstruction of Strong Scatterers

An approach to solve the inverse scattering problem for imaging applications is presented. The proposed algorithm is based on an inexact Newton method applied to the Lippmann¿Schwinger integral equation. Inhomogeneous dielectric scatterers with high contrast in a tomographic arrangement are considered. The capabilities of the approach are assessed with reference to several different unknown objects. In particular, both noisy synthetic data and experimental data have been used. Furthermore, a comparison with a model based on the second order Born approximation is also reported.

Detection of buried inhomogeneous elliptic cylinders by a memetic algorithm

The paper studies the application of a global optimization procedure to the detection of buried inhomogeneities. The object inhomogeneities are schematized as multilayer infinite dielectric cylinders with elliptic cross sections. An efficient recursive analytical procedure is used for the forward scattering computation. A functional is constructed in which the field is expressed in a series solution of Mathieu functions. Starting with the input scattered data, the iterative minimization of the functional is performed by a new optimization method called memetic algorithm.

A Novel Microwave Imaging Approach Based on Regularization in

Inverse problems arising in microwave imaging suffer from high ill-posedness. As it is well known, it is necessary to employ regularized inversion methods, in order to mitigate such behavior. Usually, such approaches are formulated in standard Hilbert spaces. Recently, a more generic regularization theory, working in Banach spaces, has been investigated, in order to overcome some limitations of the Hilbert-space regularization. In this paper, a novel imaging algorithm, performing a Lp Banach-space regularization, is proposed for 2-D electromagnetic inverse scattering problems. The reconstruction capabilities of the methods are evaluated by using numerical and experimental data.

L^{p}

A new inverse scattering method is assessed against some of the real input data measured by the Institut Fresnel, Marseille, France. The method is based on the application of an Inexact-Newton method to the Lippmann-Schwinger integral equation of the inverse scattering problem within the second-order Born approximation. The regularization properties of the approach are evaluated by considering the reconstruction of multiple dielectric cylinders.

Banach Spaces

In this paper, the use of a smart antenna system for the estimation of the directions of arrival (DOAs) of multiple waves is considered. An efficient method based on the support vector regression is proposed, in which the mapping among the outputs of the array and the DOAs of unknown plane waves is approximated by means of a family of support vector machines. Several numerical results are provided for the validation of the proposed approach, considering multiple impinging waves both in noiseless and noisy environments.

A Two-Step Iterative Inexact-Newton Method for Electromagnetic Imaging of Dielectric Structures from Real Data

The optimization of the directivity of the difference patterns in monopulse array antennas is considered. A subarray configuration is used in order to avoid the need for the implementation of two separate feed networks. An hybrid differential evolution method is applied to contemporarily determinates the weights of the subarrays and the group membership of the elements. Some numerical results are provided together with a validation comparison with data previously published in the literature.

A smart antenna system for direction of arrival estimation based on a support vector regression

Phaseless data are used to evaluate the application of an electromagnetic inverse-scattering-based procedure for the detection of cylindrical inhomogeneities, which are schematized as multilayer infinite dielectric cylinders with elliptic cross sections. The electromagnetic inverse problem is recast as a global optimization problem and iteratively solved by an efficient memetic algorithm, which combines deterministic and stochastic concepts. Moreover, a recursive analytical procedure is used for the forward-scattering computation. The possibility of localizing and reconstructing the scatterers by using phaseless input data, which would greatly simplify the design of the imaging apparatus, is evaluated both with reference to synthetically produced data and by means of experimental data obtained by a microwave tomograph.

Optimization of the Directivity of a Monopulse Antenna With a Subarray Weighting by a Hybrid Differential Evolution Method

Indoor localization of targets by using electromagnetic waves has attracted a lot of attention in the last few years. Thanks to the wide availability of electromagnetic sources deployed for various applications (e.g., WiFi), nowadays it is possible to perform this task by using low-cost mobile devices, such as smartphones. To this end, in order to achieve high positioning accuracy and reduce the computational resources used in the position estimation, fingerprinting approaches are usually employed. However, in this case, a time-consuming training phase, where a great number of measurements must be performed, is needed. In this letter, a novel approach, where the training data are obtained by means of finite-difference time-domain (FDTD) simulations of the electromagnetic propagation in the considered scenario, is presented. The performances of the method are assessed by means of experimental results in a real scenario.