Martin Gimersky

Imaginary part of antenna's admittance from its real part using Bode's integrals

The imaginary part of an antenna input admittance is calculated from its real part using Bodes integrals. Since the real part is typically a smoother function of the frequency than the imaginary part, the procedure presented here requires computation at a smaller number of frequency points, thus saves time, and is ideal for systems whose input conductance exhibits sharp peaks. A numerical procedure to evaluate the singular Bodes integral is also presented. Numerical examples using a wire antenna are used to illustrate the advantages of this approach compared to calculations involving a densely scanned frequency range. The noise stability and robustness of the algorithm are demonstrated through the successful prediction of the susceptance and the resonant frequencies of the antenna in the presence of random noise in the conductance. >

Numerical evaluation of the two-dimensional generalized exponential integral

An exact technique for the evaluation of the two-dimensional generalized exponential integral-as required in structures involving rectangular planar radiators and transmission lines-is presented. It is shown that the singularity is integrable if integration is performed in polar coordinates. The integral can be calculated exactly and with minimal computational effort, even if integration over the origin of the coordinate system is required-regardless of whether the origin is an internal or boundary point of the integration region. Comparison with a standard technique proves the presented approach superior. Stability of the algorithm and convergence is discussed. Performance is demonstrated for the example of an asymmetrically edge-fed patch antenna.

Simple and efficient numerical evaluation of the one-dimensional generalized exponential integral for ultra-thin wire antennas

A new, simple, and efficient technique for the accurate calculation of the one-dimensional generalized exponential integral is presented. The method isery robust, suitable for een ultra-thin wire antennas, which are of great interest for communications with sub- mersibleehicles, for example. Q 1998 John Wiley & Sons, Inc. Microwave Opt Technol Lett 19: 255)257, 1998.

A cellular-space-division-based method of moments algorithm for the pattern analysis of printed-circuit radiators

A cellular-space-division-based method of moments (MoM) algorithm for the analysis of geometries involving imperfectly conducting planar radiators as well as lossy and finite-extent dielectric substrates is presented. Since the technique, via the volume equivalence theorem, replaces the structure under analysis with an equivalent structure composed of thin-wall cells, modeling of the surrounding environment is not required, hence, completely avoiding the need for absorbing boundary conditions. Real (as opposed to perfect) material parameters are incorporated via properly defined surface impedances. Several examples of radiation patterns (including radiation underneath the ground plane of a finite-extent substrate) of planar geometries are presented. The calculated patterns are compared with measured results and are found to be in good agreement.

A modified method-of-moments technique for the full-wave analysis of imperfect conductors on lossy and finite-extent substrates

A modified method-of-moments technique with general field-solver capability is presented. The structure to be analyzed is subdivided into a number of thin-wall cells. Surface impedance concepts are used to represent the material characteristics of each cell. The outstanding advantages of this method include: the absence of absorbing boundary conditions, as material parameters are defined with respect to a surrounding environment, e.g., free-space, thus minimizing the computational domain; conductor and dielectric losses are readily incorporated via the surface impedance concept; and radiation into any direction, even below the ground-plane of a finite-extent substrate, is included. Several examples involving imperfect conductors as well as lossy and finite-extent dielectric substrates are presented. The method is compared with measured results and is found to be in good agreement.

Investigation of mutual coupling effects on the radiation pattern of rectangular patch antennas

Different scenarios of including mutual coupling effects in the analysis of rectangular patch antenna arrays are investigated. A standard method of moments technique is utilized to present three different models of mutual coupling interactions. The corresponding radiation patterns of a 3*3 array are compared with those of neglected mutual coupling. The results illustrate the movement in the radiation patterns as the modeling of the contribution of coupling effects increases from one level to another. It is found that considering only the adjacent patches of the individual elements results in inaccurate radiation pattern predictions.<<ETX>>

Tuning-free analysis of planar radiators based on an exact numerical evaluation of the two-dimensional generalized exponential integral

It is demonstrated that it is possible to accurately calculate the two-dimensional generalized exponential integral in the method-of-moments formulation. By performing the integration in polar coordinates over the radial variable p, the singularity existing in Cartesian coordinates is removed from the integrand. This in effect means that the commonly applied introduction of an offset into the integrand and subsequent tuning of that offset can be eliminated entirely. As a result, the calculated current distribution is unambiguous and exact. A comparison between calculated and measured input impedance data of a patch antenna shows good agreement and, hence, verifies the integral evaluation, even if a simplified model of non-overlapping subsectional expansion functions is assumed. The proposed integral calculation requires minimal computational effort; only 12 points in the Gaussian quadrature are needed to achieve results accurate to better than 0.1 percent.

Design and measured performance of advanced Satcom dielectric lens antenna

This paper describes design techniques and presents calculated as well as experimental performance of a multiple-beam dielectric lens antenna for the Ka-band Advanced Satcom application. The lens is spherical on the outside and zoned on the inner surface. A dual-CP, single-horn-per-beam mode of operation is assumed. In order to examine mutual-coupling effects, however, a 19-element feed cluster, with only one element active, was built and tested. Primary and secondary radiation patterns are presented. Superior electrical performance of the lens and excellent agreement of experimental and predicted results are demonstrated.

Feed development for advanced Satcom RX-antenna application

The paper presents the results of the feed development for the Advanced Satcom Rx-antenna. Two design variants have been developed and described in this paper. The first feed design was used to illuminate a dielectric lens antenna, whereas the second is compatible with a multiple reflector antenna system. Both antenna concepts require the use of a feed cluster. A 19-element feed cluster was designed, built and tested for the dielectric lens. A single feed chain was developed for the reflector system. The feed design issues, as well as the measured results are presented in the paper. Excellent feed performance is demonstrated.

Design and implementation of the Koreasat-3 Ka-band FSS antenna

The Ka-band antenna subsystem discussed in this paper forms a part of an active-repeater communications spacecraft for use in geosynchronous, circular, equatorial orbit for the Korean domestic service. The antenna is a Gregorian-type configuration with a shaped main reflector fed by a circularly polarized feed and it is used for transmission and reception. The paper first presents the antenna main performance requirements and the description of the design. They are followed by a description of the key implementation issues. Finally, feed breadboard results and antenna performance predictions derived from the feed results are presented.