AMTA Paper Archive

Two of the three algorithms require an estimate of the element pattern, which they assume to be common to all the elements. We describe our measurement of our array’s element pattern, as well as the use of the IsoFilter™ to center the element pattern and limit the edge effects.

TESTING LARGE WIRELESS DEVICES IN SMALL ANECHOIC CHAMBERS 100_0967100_0973 James D. Huff -20.00-18.00-16.00-14.00-12.00-10.00-8.00-6.00-4.00-2.000.00050100150200Relative Power (dB) Theta Angle (deg) Uncorrected Dipole Patterns 0,0,00,0,120,0,18-20.00-18.00-16.00-14.00-12.00

Azimuthal mode (µ mode) truncation of a high-order probe pattern in probe-corrected spherical near-field antenna measurements is studied in this paper. The results of this paper provide rules for appropriate and sufficient µ-mode truncation for non-ideal first-order probes and odd-order probes with approximately 10dBi directivity. The presented azimuthal mode truncation rules allow minimizing the measurement burden of the probe pattern calibration and reducing the computational burden of the probe pattern correction.

A spherical array designed for hemispherical coverage of satellite communications at S-band that is approximated by hexagonal and pentagonal planar panels. Ball Aerospace built a segment of a 10m diameter spherical array that has one center pentagonal panel and five surrounding hexagonal panels. This paper de­scribes our efforts at testing one large hex­agonal panel in a compact range.

This paper describes the behavior of the antenna radiation pattern for different planar near-field measurement errors superimposed on the near-field data. The disturbed radiating near field is processed using multilevel plane wave based near-field far-field transformation to determine the far-field. Errors like scan area truncation, transverse and longitudinal position inaccuracy of measurement points, and irregular sample spacing are analyzed for an electrically large parabolic reflector at 40 GHz. The error behavior is then compared with the standard transformation technique employing 2D Fast Fourier Transform (FFT) using the same near-field data. In order to exclude the effect of any other measurement or environmental error, electric dipoles with appropriate magnitude profile and geometrical arrangement are used to model the test antenna.

The radiation pattern of an antenna can be deter­mined accurately by near-field measurement and transformation techniques. Low numerical complexi­ty, full probe correction capabilities, and high accura­cy of the transformed far-field pattern are important features of near-field transformation algorithms. The multilevel plane wave based near-field transformation algorithm achieves an efficient full probe correction by plane wave representations of antenna and field probe and realizes the low numerical complexity by hierarchical grouping of measurement points. Field translations are carried out to the boxes on the coars­est level and are further processed to the measure­ment points by disaggregation and anterpolation. Dis­aggregation is a simple phase shift of the plane waves and anterpolation reduces the sampling rate corre­sponding to the spectral content of the plane wave spectra on the various levels. The accuracy of the transformation is influenced by several variables where the number of buffer boxes between antenna and measurement point groups is crucial. A higher accuracy due to more buffer boxes can be achieved at the cost of increased computation time. Adaptive field translations structure the measurement setup such that individual groups are transformed with the re­quired accuracy at lowest costs. A detailed investiga­tion for a planar near-field measurement will be shown.

This paper compares 3 sets of far-field patterns of an S-Band Open Ended Waveguide (OEWG). The sources for the data are measurements from NPL, an EM-Model and the commonly used NIST analytical model. Both co-polarized (co-pol) and cross-polarized (x-pol) patterns are compared. Results indicate that accuracy improvements are possible by utilizing an EM-Model in certain applications. These applications as well as the pros and cons of doing this are discussed. Understanding the differences between these 3 independent sets of data enables near-field range engineers to better understand the directional dependence of probe correction accuracies over the majority of the forward hemisphere. Information and insight gained from this comparison, along with specific AUT requirements, better equips the near-field range user to address probe correction concerns and ultimately to determine if a calibrated probe solution is required for their unique testing scenario.

The use of computational electromagnetics (CEM) prediction codes in the analysis and interpretation of RCS measurements has become increasingly prevalent. This is in large part due to rapid advances in computing capability over the last several years, particularly for rigorous techniques such as the method of moments (MoM). In many instances, however, these codes are still limited to plane waves and/or elementary dipoles as the sources of target illumination. Modeling of the illumination from an arbitrary antenna therefore requires meshing and solution of the combined antenna-target geometry for each frequency and aspect angle, with an associated increase in the computational complexity of the problem, even if the interactions between the antenna and the target are negligible. In this paper, we describe a method by which measurements or predictions of the antenna pattern are used to develop an equivalent representation of the antenna in terms of an array of non-interacting elementary dipole current sources in a MoM code that uses RWG basis functions. The representation can then be used to efficiently derive the antenna’s illumination on the target as a function of frequency and aspect angle with only a minor increase in the computational burden relative to plane wave illumination. Results are presented using antenna pattern predictions for an ETS-Lindgren 3164-01quad-ridged VHF antenna which illustrate the accuracy and efficiency of the technique.

Two novel antenna designs using inexpensive planar printing technology are presented. The goal is to obtain antennas that maintain their pattern and other antenna parameters over a large range of frequencies so they can be used efficiently in applications with either multiple frequencies or broadband frequency range. The advantage of these structures is the design simplicity, cheap cost, and capability of operating at high frequencies. The design technique, the measurements of pattern variation of electric fields E and H, bandwidth, and the reflection coefficient parameters between 2 and 6.5 GHz are presented. Although the field strength of the pattern is reduced as the frequency increases, the pattern is within a 10 dB difference between 2 and 4 GHz and is mostly omni directional in this range.

Kriging fitting, originally developed for geological exploitation, is here applied for fitting an expected pattern to noisy, irregular in-flight measurements of a satellite antenna. The noise level in in-flight measurements is often so high that only the central part of the main beam ap­pears. By the Kriging method, first a characteristic function, the regression model, is fitted to the meas­urements. For the main beam this is chosen to be de­scribed by a general second order polynomial. To this is added a more detailed correlation model which represents realistic deviations from the regression model but filters out the fast variations of the noise. The method is applied on simulated measurements on the Planck RF telescope and the presented results show a considerable reduction of the noise floor of the Figure 1 – The Planck double reflector antenna system pattern; even beam details invisible in the original with two ellipsoidal mirrors (aplanatic configuration). measurements (a shoulder) are revealed by the pattern From the antenna pattern obtained by the in-flight testing fitting1 .

In the paper [1] the principles of operation of high performance absorbing materials were described and the criterion for absorber performance optimization at UHF/VHF frequency bands was proposed and confirmed experimentally on a number of absorber components optimized for operation at low frequencies such as the VHF/UHF bands. C:\Publishing 2011\AMTA 2011\Papers\Absorbing Material Performance\freq dom 18 24 36 60.jpg The experimentally verified optimization criterion is intended to determine the optimum carbon loading of the absorber components, thus delivering optimal reflectivity of the full absorbing assembly (absorber components on a metallic backing plate) at the lowest possible operating frequency. The optimization is based on equalization of reflections in the time-domain from the front face surface of the absorbing component and from the backing metallic plate. Validity of the criteria was confirmed by measurements of the reflectivity of pyramidal absorbing components of various heights, (3’, 5’, 6’ and 8’ [3]) in a 40’ long coaxial line terminated in a metallic back wall [2,4]. In this paper, more details are highlighted explaining how the criterion is delivering the best absorber reflectivity at low frequencies. This is accomplished by implementing time gating post-processing to isolate two primary concurrent peaks corresponding to the reflections from the front surface and metallic backing substrate. It is shown that the improved reflectivity is achieved by a self-cancellation of the two signals delivering the “null” in the frequency domain, which, in turn determines the lowest operating frequency attributed to an absorber of a given height. It is shown that the “null” property of the reflectivity pattern, as well as the properties of the peaks in between “nulls”, can be scaled and, therefore, predicted based on the height of the absorber almost everywhere in the UHF band. Thus, it is possible to optimally choose the grade of the absorber necessary to meet or exceed given reflectivity specifications, or to manufacture the appropriate absorber grade which can deliver the optimum reflectivity at the specified frequency.

Since the early days of antenna pattern recorders, advances in instrumentation and computers have enabled measurement systems to become highly automated and much more capable. Automated systems have provided higher productivity, more efficient use of test facilities, and the ability to acquire more data in less time. In recent years, measurement speeds of microwave receivers and vector network analyzers have advanced considerably. However, to take full advantage of these speed improvements, the measurement system architecture must be carefully considered. Small differences in instrument timing that are repeated many times can make large differences in system measurement time. This paper describes a general method of calculating system measurement time based on the primary factors that affect system timing, including position trigger detection, frequency switching time, multiplexer switching time, receiver measurement time, and timing overhead associated with triggers, sweeps, and measurements. It also shows how key features of instruments available today can be used along with improved antenna measurement system architectures to optimize system throughput.

A comparison of antenna gain and radiation pattern for R-band (1.7 – 2.6 GHz) antennas has been performed between Korea Research Institute of Standards and Science (KRISS) and eight domestic participants including private companies and public institutes. Its purpose was to check equivalences between KRISS and participants in gain and radiation pattern measurements for antennas, particularly at R-band, to support antenna manufacturers and end users in Korea as a proficiency test program of the ‘Antenna Measurement Club’ of KRISS. This comparison uses three traveling standards (a general purpose antenna (pyramidal standard gain horn antenna), a fan-beam antenna (a sector antenna for mobile base stations), and a small antenna (a sleeve dipole antenna) and measurement parameters are the power gain and radiation pattern of the traveling standards. Gain comparison method, extrapolation method, and cylindrical near-field measurement method are used in this comparison.

ABSTRACT This work deals with the experimental validation of a direct near-field – far-field transformation with cylin­drical scanning for electrically long antennas, which requires a minimum number of measurements. Such a transformation is based on a nonredundant sampling representation making use of a flexible source model-ling suitable to deal with electrically long antennas and allows the evaluation of the antenna far field di­rectly from the acquired near-field data without inter­polating them. The good agreement between the so recovered far-field patterns and those obtained via the classical cylindrical near-field – far-field transforma­tion assesses the effectiveness of the approach.

Mathematical Absorber Reflection Suppression (MARS) has been used successfully to identify and extract range multi-path effects in a great many spherical [1, 2], cylindrical [3, 4], and planar [5, 6] near-field antenna measurement systems. This paper details a recent advance that enables the MARS measurement and post-processing technique to be used to correct antenna pattern data from far-field or compact antenna test ranges (CATRs) where only a single great circle pattern cut is taken. This paper provides an overview of the measurement and novel data transformation and post-processing chain that is utilised to efficiently correct far-field, frequency domain antenna pattern data. Preliminary results of range measurements that illustrate the success of the technique are presented and discussed.

In this work we study the near-field antenna measurement using cubical surface scanning and related near-field to far-field (NF-FF) transformations. The cubical surface scanning is a fascinating idea because it can be realized using widely used planar scanning on six surfaces of a cube, and it provides the possibility to determine the complete 3-D pattern instead of the pattern in a limited angular region as in traditional planar scanning. The NF-FF transformation presented in this paper is based on spherical vector wave expansion (SWE). The most important issue of this paper is to introduce the azimuthal mode decomposition technique to be applied as a part of the NF-FF transformation allowing a reduction in the computational burden of the transformation.

Multi-layer dielectric rod (MLROD) antenna has been shown to provide wideband, dual-polarization, symmetric patterns, and stationary phase center. The key challenge in designing a MLROD antenna is to choose proper thickness and dielectric constant of each layer, and shape of radiation tip to meet desired VSWR, pattern, and phase center requirements. This becomes even more difficult as the number of layers increases for achieving greater bandwidth. This paper discusses a design optimization procedure of a UWB 3-layer MLROD using Genetic algorithm with novel fitness functions for simultaneously controlling reflection coefficient, phase center, and pattern three key characteristics. The final design exhibited excellent desired performance throughout the desired frequency range.

Antenna pattern measurement in the presence of obstacles requires an accurate characterization of the antenna-scatterer interaction in order to retrieve the multipath effects that distort the antenna pattern. In this contribution, a new approach based on the Sources Reconstruction Method is proposed. The idea is to characterize the Antenna-Under-Test (AUT) and the scatterers through a set of equivalent currents that radiate the same fields as the original problem. Thus, it is expected that the equivalent currents retrieved on the surface enclosing just the AUT will provide the AUT radiation pattern. A comparison with modal expansion of the fields on the reconstruction domains is also performed.

The proposed method is based on fiber-optic link connected to the antenna under test in order to measure the radiation pattern of electrically small antennas without any metallic cable effects. The experimental radiation characterization of electrically small antennas reveals significant effects of the RF cable strongly disturbing the antenna gain and radiation patterns for both polarization components. With a very simple use and a low cost, the presented systems can characterize radiation of both narrow and UWB compact antennas in the 0.1-4 GHz frequency band. The proposed test-bench is firstly validated then its performances are compared to classical cable measurements and antenna simulation results. Both measured and simulated results are compared and the agreement is excellent (±1 dB) for co polarisation in the different cut planes. Limitations of the proposed method are also addressed. For very low gain levels (for the presented antennas at 950 MHz and 473 MHz, a cross polarization level below -30 dBi) the 42 x 20 x 23 mm3 autonomous optical receiver is too large and alters the antenna under test radiation. In this last case, and for a using at higher frequency, it is recommended to miniaturize the optical receiver whose size is mainly governed by battery performances (6 hours).

Fig. 1: Cylindrically conformal antenna array system This paper discusses an example of a unique cylindrical conformal array design approach. This exemplary design is composed of six subarrays of series-fed microstrip patch antenna operating in X-band. The diameter of the conducting cylinder is 7.5 inches in diameter. Each subarray contains seven patch antennas connected in series configuration and is connected to a single coaxial probe located at the center element. Such feed arrangement greatly reduces the number of feeding cables, increases the feed-line isolations, and minimizes the feed-line radiation. The elevation pattern of each subarray is controlled by the number of patches and impedance matching tapering via varying the width of connecting microstrip lines. The azimuth pattern is controlled by the subarray height above cylinder surface, substrate width, cylinder radius, and surface treatments between subarrays. Measurement results exhibited good impedance matching and broad antenna coverage in X-band.