G. D'Elia

Antenna pattern synthesis: a new general approach

The antenna pattern synthesis problem is of the utmost importance in almost every kind of antenna applications. Therefore, a very large number of contributions have appeared on this subject. But virtually all of them deal with simplified versions of the complete synthesis problem, wherein the degrees of freedom available in principle are strongly reduced, and/or idealized design criteria or requirements are considered. In this paper we present a formulation which allows us to overcome this fragmentation of the synthesis problem. A clear and direct description of the performance actually required by the antenna and a representation of the radiating properties of the antenna as a system allows us to formulate the synthesis problem as an intersections finding problem, i.e., to find a common element between a number of sets, each one containing elements fulfilling part of the requirements. This allows a completely general and flexible formulation of the problem, independent of the actual structure of the antenna. Then the practical implementation of this formulation is widely discussed, showing how an efficient solution procedure can be devised. The implications of the well-known ill-conditioning of the synthesis problem are also discussed. In order to show how the approach works and to assess its flexibility and power, a couple of significant examples are included, namely, a phase-only reconfigurable array and a shaped reflector synthesis. These examples are unconventional since no a priori choice of the intensity distribution (for the array case) or of the feed cluster (for the reflector case) is required. The method presented is able to exploit all the available degrees of freedom in order to fulfill the design requirements. >

Far-field pattern determination from the near-field amplitude on two surfaces

The possibility of determining the far field of radiating systems by measuring only the near-field amplitude is investigated. The main difficulties of the problem are examined in some detail and a new near-field/far-field transformation technique is developed, based on the measurement of the near-field amplitude over two surfaces surrounding the antenna under test. The accuracy of the far-field reconstruction results are related both to the distance between such surfaces and to some a priori information concerning the near-field phase and/or the radiating system. The information on the radiating system allows relaxation of the need for any information on the near-field phase provided that the distance between the measurement surfaces is high enough. Conversely, the knowledge of a more or less corrupted near-field phase allows reduction of such distances without affecting the accuracy of the far-field reconstruction. Numerical examples validating the effectiveness of the developed algorithm are provided for the planar scanning case. >

Phaseless Antenna Characterization by Effective Aperture Field and Data Representations

The problem of antenna characterization from phaseless near-field data is addressed by appropriate use of the available information on the Antenna Under Test (AUT) and on the scanning geometry to provide efficient representations for both the unknowns and the data. Such a strategy allows improving the reliability and the accuracy of the proposed characterization algorithm and, at the same time, shortens the overall measurement process. An extensive numerical and experimental analysis, together with a comparison with existing approaches, endorses the algorithm reliability and accuracy and confirms its usefulness for antennas having a general radiating (vector) behavior, i.e., either focusing or non-focusing.

Fast analysis of large antennas--A new computational philosophy

A new computational approach is presented which allows a fast analysis of radiation properties of large antennas. The radiated field is first computed using conventional techniques, e.g., physical optics and geometrical theory of diffraction, in prescribed sampled space directions, roughly one direction per lobe. Then sampling theory is used to reconstruct the complete radiation diagram. Numerical experiments are presented in the last part of the paper, showing the excellent performance of the method.

Global Optimization and Antennas Synthesis and Diagnosis, Part One: Concepts, Tools, Strategies and Performances

This is the first of two companion papers on global optimization and antenna analysis and synthesis. In Part I, an analysis of the problems involved in Global Optimization is presented by critically discussing the basic concepts and tools, the performances to be expected, the required computational complexity and the guidelines to select algorithms solving efficiently the problem at hand. The relevance of stochastic techniques is enhanced and the role of double phase algorithms is stressed. The proof of the convergence property of an idealized version of a simplified evolutionary algorithm is provided. In Part II, the selected algorithm, a hybrid evolutionary algorithm, is tested against two real world problems relevant in electromagnetics, the power synthesis of contoured beam hybrid reflector antennas and the reflector antenna diagnosis from only amplitude data. The results of an extensive numerical analysis are presented.

Power pattern synthesis of reconfigurable conformal arrays with near-field constraints

A conformal array power pattern synthesis technique is presented which makes it possible to take into account near-field constraints and to reconfigure the radiated pattern by controlling only the phases of the excitation coefficients. Far-field pattern specifications are given by masks while near-field constraints are given by prescribing the maximum allowable field intensity at points arbitrarily located in a given near-field region. The amplitude of the excitations, common to all radiated beams, and the phases corresponding to each one are obtained as a result of the synthesis algorithm.

Photonic Probes and Advanced (Also Phaseless) Near-Field Far-Field Techniques

We present innovative near-field test ranges, named compact-near-field (CNF) and very-near-field (VNF). These use photonic probes, and advanced near-field far-field (NFFF) transformations from amplitude and phase (complex) or phaseless measurements. The photonic probe allows AUT-probe distances of less than one wavelength. This drastically reduces test-range and scanner dimensions, improves the signal-to-clutter ratio and the signal-to-noise ratio, and reduces the scanning area and time. In both the cases of complex and phaseless measurements, the neat-field-to-far-field transformation problem is properly formulated to further improve the rejection of clutter, noise, and truncation error. The advantages of the compact-near-field and very-near-field test ranges are discussed and numerically analyzed. Experimental results are presented for both planar and cylindrical scanning geometries.

Dielectric Field Probes for Very-Near-Field and Compact-Near-Field Antenna Characterization [Measurements Corner]

A novel setup, based on the use of non-invasive dielectric field probes, and employing accurate and reliable algorithms for antenna characterization, is described. Experimental results show how the proposed system can provide even more accurate characterizations than standard near-field systems, enabling considerable reduction of indoor test-range dimensions (compact near-field). Also, the potential for very-near-field acquisition, as well as the sampling strategy in the reactive zone of the radiator, are pointed out.

A novel approach to scatterers localization problem

A new approach to find the number and locations of electromagnetic scatterers (sources) from one view, one frequency field data is presented. The case of scatterers (sources) with dimensions comparable to the wavelength is considered. First, in order to devise an effective technique, a new relevant property of the electromagnetic field, the point source spectral content, is introduced and its relationship with both the sliding windowed Fourier transform of the field and the local bandwidth function is discussed. To enlighten its usefulness it is also shown that a footprint of the scattering system encoding information on its geometry can be easily extracted from the scattered field by exploiting this new property. On this basis, the new localization technique is introduced. In order to restrict the search region, the minimum circular envelope enclosing the scatterers is found by purposely introducing a new technique exploiting the effective bandwidth of the radiated field. Then, the number and locations of scatterers is retrieved by using the field local quantitative feature previously introduced, without the complexities of the full inverse problem as it is usually done by the traditional approaches. In this way, not only a simplified technique is obtained but also the ill-posedness of the problem and the noise effects are significantly mitigated. The effectiveness of the proposed technique and its overall performance with respect to a singular value decomposition based approach are proved by means of a numerical analysis.

Global Optimization and Antenna Synthesis and Diagnosis, Part Two: Applications to Advanced Reflector Antennas Synthesis and Diagnosis Techniques

The paper presents the application of the hybrid global optimization algorithm, introduced in the companion paper Part I, to reflector antenna power pattern synthesis and reflector antenna surface diagnosis from only amplitude data. The synthesis algorithm determines both the reflector surface and the excitation coefficients of the array of primary feeds to meet the designing specification on the far-field pattern expressed by means of two couple of masks bounding the squared amplitude of both the copolar and crosspolar components. The diagnosis technique allows to find the reflector surface profile from the measurement of the far field power pattern by a proper formulation of the corresponding inverse problem. In both cases we take advantage of the exploring capability of an evolutionary algorithm and of the solution refinement capability of an efficient, quasi-Newton based, local search procedure. The numerical analysis shows that Global Optimization can outperform the standard local approach, by significantly improving the performance of the synthesized antenna in the first case and by enhancing the reliability of the diagnosis procedure in the second one.