Welcome to the AMTA paper archive. Select a category, publication date or search by author.
(Note: Papers will always be listed by categories. To see ALL of the papers meeting your search criteria select the "AMTA Paper Archive" category after performing your search.)
High Power Superposition (HPS) is a method for measuring high power active array antennas in the transmit mode using the near-field technique. Due to the substantial field emitted by array antennas it is not practical, due to safety reasons, to power all of the elements at once. Therefore, the nearfield of smaller element groups are measured individually and the results are added together using complex mathematics. This forms the mathematical equivalent of the full power nearfiield, which is then processed using conventional near-field techniques. The results from the HPS testing method will be discussed with consideration of all the errors introduced. In addition requirements, issues, and solutions for accurate HPS testing will be discussed.
One problem in a RCS ground bounce range is that the direct signal can be interfered with by the ground reflected signal. The undesired ground bounce signal will cause errors in the RCS measurement. This paper presents a study of ground bounce reduction using a tapered and stepped resistive sheet fence. In order to show that the proposed R-card fence technique can reduce the ground reflected signal significantly, both experimental and theoretical studies are performed. The resistance of the R-card varies based on a Kaiser-Bessel taper function. The experimentall results with and without the R-card fence show that the ground reflected signal can be attenuated by about 20dB. Both vertical and horizontal polarization cases are considered. This paper also the results of a simulation using NEC-BSC (Numerical Electromagnetic Code - Basic Scattering Code, developed at The Ohio State University ElectroScience Laboratory). Comparison of the results between measurements and simulations will be shown in this paper.
For greatest efficiency and accuracy in measuring patterns of a circularly polarized antenna on a planar near field range (NFR), a recommended procedure is to use a fast switched, dual circularly polarized probe. With such equipment one obtains complete pattern and polarization data from a single scan of the antenna aperture. For our task of measuring high gain shaped beam apertures, measurement efficiency is further improved by using a moderately high gain (about 12 dBi) probe that has been accurately calibrated for patterns, polarization, and gain over the test frequency band. Such a probe allows scan data point spacing to be typically at least one wavelength, thus keeping scan time minimized with acceptably small aliasing (data spacing) error. The measured near field amplitude and phase data is transformed via computer to produce the angular spectrum that is further processed to remove the effect of the probe patterns, i.e. probe correction. The final output is a set of (principal and cross) circular polarized far field patterns. However on one occasion, due to fast breaking changes in requirements, we were unable to obtain a calibrated circular polarized probe in the available time. For this test we used an available calibrated 12 dBi fast-switched dual linear-polarized probe with software capable of processing principal and cross circular-polarized far field patterns. As anticipated, we found from preliminary tests that the predicted low cross-polarized shaped beam pattern was not achieved when using the calibrated fast Ku band probe switch. Further tests showed the problem to be due to small errors in calibration of the probe switch. This paper will discuss test and analysis details of this problem and methods of solution.
We evaluate radiation characteristics of a millimeter wave reflector antenna by using two sets of phaseless data measured in its near field. This is called a near field phase retrieval method (NFPRM). To apply this method to millimeter or submillimeter antennas, we have to pay more attention to the relations how the estimation errors and convergence are affected by the interval of two planes and SIN ratios of the measurements. This is because the difference between the two measured amplitude distributions is usually very small. In this paper, this method is applied to a case for 90-GHz reflector antenna with an extremely small interval between the two planes. The results show some clear correlation between the estimation errors and measurement conditions.
Considerable progress has recently been made in the application of phase retrieval methods for phaseless near-field antenna measu rements. These techniques have sufficiently matured so that accurate antenna measurements can be performed when the phase information is either unavailable or inaccurate. A comparison of conventional (amplitude and phase) and phaseless (amplitude only) planar near-field measurements for non-ideal measuring probe locations is examined via simulated array antenna case studies involving both random and systematic errors. It will be demonstrated that the presented phase retrieval algorithm can more accurately reproduce the true pattern of the antenna under test because of the diminished sensitivity of the amplitude of the near field, as compared to the phase, with respect to the measuring probe locations. This phase retrieval approach requires no knowledge of the actual measurement locations, other than the nominal location of the two required measurement planes, and is suitable for relatively large probe position errors.
The majority of precision spherical positioner alignment techniques used today are based on procedures that were developed in the 1970's around the use of precision levels and auto-collimation transits. Electrical alignment techniques based on the phase and amplitude of the antenna under test are also used, but place unwanted limitations on accurately characterizing an antenna's electrical/mechanical boresight relationship. Both of these techniques can be very time consuming. The electrical technique requires operator interpretations of data obtained from amplitude and phase measurements. The autocollimation technique requires operator interpretations of optically viewed measurement data. These results are therefore typically operator dependent and the resulting error quantification can be inaccurate. MI Technologies has recently developed a mechanical alignment technique for Spherical Near-Field antenna measurements using a tracking laser interferometer system. Once the laser system has been set-up and stabilized in the operational environment; the entire spherical near-field alignment may be completed in a few hours, as compared to the much more lengthy techniques used with level/transit or electrical techniques. This technique also simplifies the quantification of the errors due to the inaccuracy of the alignment. This paper will discuss the effect of the alignment error on results obtained from spherical near-field measurements, and the procedures MI Technologies developed using a tracking laser interferometer system to obtain the precision alignment needed for a spherical near-field measurement.
A software's user-interface design determines how productive someone will be in a accomplishing a given task. This is particularly true in the area of antenna measurement and analysis. The MiDAS software package is used as an example of how software can be specifically designed to focus on enhancing efficiency by implementing an advanced human-machine interlace. Simple yet critical aspects such as minimized menu access, integrated, user friendly error checking and help, and clear, consistent, and integrated features help to improve productivity, reduce errors and save time. In addition, design principles such as having only one interface for all antenna measurement disciplines (e.g., near-field and far-field), reduces the time needed for training which, in turn, lowers costs. This paper explores how the implementation of such a user interface can be used as a paradigm for increasing efficiency in the field of antenna measurement and analysis.
As a result of Government and Industry RCS Teaming, initial RCS range certification exercises are underway. One critical element of certification exercises is the modeling and characterization of error terms according to the unique properties and requirements of individual RCS ranges, and the development of a method for propagating these errors into overall RCS measurement uncertainty. Previously, we presented the statistical model for the case where errors are grouped into multiplicative and additive classes, as well as a robust methodology for the propagation of errors in both the signal space and RCS (dBsm) domains [1-3]. Initial data at the National RCS Test Facility (NRTF) RAMS site located in the White Sands Missile Range near Holloman AFB, NM, have been collected for range certification exercises. Preliminary analysis has been accomplished on certain dominant error terms for calibration uncertainty characterization only. A general approach [7] has been followed here, with the exception that multiplicative and additive error terms are treated separately. In addition, only variance effects are treated (not bias). This paper is a status of work in progress. The ultimate goal of this work is the full implementation of previously described concepts [1-3]. We plan to demonstrate an improved ability to capture the effects of both error bias and variance (as has been demonstrated mathematically to date) using a more complete set of data collections.
This is a report on work in progress to understand the wide variations in sphere calibration data observed on dynamic radar cross section measurement ranges. The magnitude of these fluctuations indicate an uncertainty of greater than 2 dB in some cases. The range of fluctuations in the received power (which is well beyond fluctuation due to received noise) underlines the need for a thorough understanding of sources of uncertainties in dynamic radar cross section measurements. In addition to the fluctuations, we observe a systematic error with respect to the mean of the data segments, possibly due to drift, pointing errors and I or target-background interactions. Understanding the error mechanisms in these measurements allows us to reduce the overall uncertainty and to improve data quality.
Boeing is currently pursuing certification of their 9-77 indoor compact range facility as a voluntary industrial participant in the ongoing DoD/NIST RCS certification demonstration program. In support of that process, V EI has applied a novel statistical method for analysis to VHF measurements of a canonical target from the Boeing 9-77 range. The dominant error sources in the range were identified and categorized according to their dependence on frequency, aspect angle, and the target under test. Range characterization data were collected on canonical targets and then used to estimate the statistical parameters of each of the errors. Finally, these were incorporated into expressions for the combined RCS measurement uncertainty for a test body whose RCS exhibits many of the characteristics of modern, high-value targets. The results clearly demonstrate the importance of accounting for the target-dependence of the errors and the bias they introduce into the overall uncertainty.
An important source of error in a compact range antenna pattern measurement is the deviation of the quiet-zone field from the perfectly fiat amplitude and phase of a plane wave field. Although some guidelines and rules of thumb exist that relate the quiet-zone field to the error in the measured antenna patterns, the error or perturbation is dependent on the particular type of antenna that is being measured. For example, the non-ideal quiet zone field will produce very different errors for a small horn than for a large phased array. A realistic error budget or uncertainty analysis of the compact-range measurement requires knowledge of the antenna pattern uncertainty as a function of the quiet-zone field and the particular antenna of interest. A simulation method is derived using reciprocity that allows one to quantify the perturbations induced in a given antenna pattern when the quite-zone field distribution is known. This is particularly useful, since one typically has a fair estimate of the antenna pattern and has measured data of the quiet-zone field. The simulation is tested by modelling the antenna as a collection of elemental current sources and simulating the quiet-zone field as generated by elemental current sources. Using this simple simulation model, a closed-form near-field antenna pattern may be calculated for comparison with the more general computer simulation derived from reciprocity.
Error budget projections of measurement accuracy require statistical descriptions of the individual error sources. Thermal noise errors are well known and commonly used. Such statistics, however, have a zero mean Gaussian distribution and sadly are misapplied to the distributions of other error sources. The class of coherent RF errors, for example, has non-zero mean values and variances that differ from Gaussian values. Such statistics are described.
RATSCAT has been heavily involved, as part of the DoD Range Certification Demonstration Program, in examining and documenting the underlying principles of all aspects of the outdoor measurement process. Our goal is to replace historical or "anecdotal" measurement approaches with processes founded on validated and documented procedures. This paper reports on the results of two areas of study. These are the effects on measurements caused by wind and calculation of target rotation rates. When RCS targets are measured outdoors on pylons or columns, some uncertainty will be introduced due to the effect of wind on the target and target support structure. This paper will present the results of an investigation into the errors introduced by wind motion on targets mounted on pylons or columns. When rotation rates are determined for target collection, the usual procedure is to employ a rule of thumb like "collecting three points per lobe" or "meeting the Nyquist criterion." This paper examines these common methods to determining rotation rates, and their impact on the measurement of the peak values of RCS magnitude and phase. Finally, the significance of these two measurement errors will be examined in light of their impact on outdoor range operations as well as on decisions based upon the collected RCS data.
An interlaboratory comparison is made between radar cross section (RCS) measurements at the test ranges at FOA Defence Research Establishment and SAAB Dynamics, Sweden. The comparison is made in order to increase the measurement and calibration quality at the ranges. An analysis of the deviations in the measured RCS data from the ranges provides a better understanding of the sources of errors. The RCS of two generic targets are measured at the X-band. The targets are simple airplane models, length and width are approximately 1.0 m, with no cavities. A brief comparison between some theoretical results and experimental RCS data are also presented.
To develop standards for radar cross section measurements a complete uncertainty analysis is needed. We derive the radar cross section error equation and examine sources of measurement errors that contribute to the overall uncertainty in calibrations and measurements. We obtain expressions for upper- and lower-bound errors and uncertainties that are generally valid for monostatic measurements on any unknown target using any standard calibration artifact. The general procedure can be extended to bistatic measurements. Some experimental procedures to determine the uncertainty due to background subtraction are presented and discussed.
The worldwide system now used in the aviation field as a landing aid is simply called the ILS, or instrument landing system. This paper is about the "null reference" type of vertical guidance component of the system. It operates in a frequency band near 333 MHz by transmitting signals from two antennas on a tower near the aircraft runway. The lower antenna, (and its image) produces a broad beam (the reference) along the approach to the runway. The upper antenna (also with its ground image) produces a vertical guidance signal with a null along the desired approach angle (or glide slope, typically 3 degrees). The reflection zone for these antennas is a critical component of the system. A problem has been discovered for the case of a layer of wet snow on the reflection zone. As the layer of snow warms up and changes from the frozen state to a water-snow mixture, the dielectric constant of the layer of snow changes over a very wide range. At some point in this process, the reflection coefficient of the layer of snow over the wet ground passes through zero at the design approach angle (3 degrees). At this time, the vertical width of the guidance null becomes much larger than normal. An aircraft will lose its normal tight control over the vertical approach angle, and may experience significant errors in the approach angle without any indication of the problem. The time for the phenomena to occur is so short that as of this date, no experimental proof of the phenomena has been obtained. The theory for these phenomena will be shown, and examples where aircraft crashes may have occurred in such conditions win be discussed. Some experimental evidence will be presented.
The objective of this study is to develop a new techniq ue to compensate the instrumentation errors of an antenna near-field test range. The methodology presented demonstrates that it is feasible to calculate the far-field radiation from near-field measurement with one deconvolution that will include all the errors introduced by the instrmentation. Measrements were performed on a standard gain horn as a reference and the analysis includes a theoretical comparison with a computer model of the standard gain horn, simulated using WIPL. Furthermore, four scenarios of error in the system flatness were analyzed, to verify that the technique is capable of correcting planarity errors.
A bias error in planar near-field measurement data comes from receiver crosstalk or leakage effects [1, 2, 3]. The bias error is a complex constant added to every near-field data sample. After transformation from the near-field to the far-field, the bias error becomes an easily identifiable spike located at the center of k-space. If one is measuring a horn, then the bias error produces a small bump or spike at the center of the far-zone pattern (i.e. at the center of k-space). If one is measuring a highgain antenna with the antenna beam pointed away from the center of k-space, then the bias error causes an erroneous sidelobe spike at the center of k-space. The bias constant is difficult to estimate be cause it may be more than 60 dB below the peak near-field level. Nevertheless, if the effect of the bias error can be seen in the far zone pattern of the test antenna, then it can be estimated and removed from the measured data. An algorithm is presented that is used to estimate the bias constant directly from the near-field data, then the bias constant is simply subtracted from the data. Examples using measured data are used to illustrate how the algorithm works and to show its effectiveness.
Papers were presented at the last two AMTA meetings reporting on the effect of rotator system alignment on the results of spherical near-field measurements. When quantifying the effect of non-intersection errors on the AUT directivity, these two papers presented very different results. One AMTA paper 1 and an earlier study at The Technical University of Denmark 2 found that the directivity error was extremely sensitive to non-intersection errors while the other AMTA paper3 found a very small sensitivity. During the past year, scientists at the Technical University of Denmark, The National Institute of Standards and Technology, and Nearfield Systems Inc. have been working together to determine the reasons for these differences. It now appears that the major reason for the difference is due to the method used to acquire data on the sphere. Theta scans that pass through the pole, or equivalently, phi spans of 180 degrees, produce both plus and minus phase errors that tend to cancel in the on-axis direction. Theta scans that do not pass through the pole, or equivalently phi spans of 360 degrees, produce phase errors of the same sign over the sphere which are concentrated in the on-axis direction. Examples of measurements and recommendations for using this information in spherical measurements will be presented.
The NIST 18 term error budget is used to estimate the magnitude of each individual source of error and then combine them to the total uncertainty for the planar nearlield range designed for antenna characterization of the ERIEYE Airborne Early Warning System. The ERIEYE AEW System consists of two large phased array antennas, one at each side of the Dorsal Unit which is located on the top of the airplane fuselage. T/R-modules are connected to the antenna waveguides to control the beamsteering and the very low sidelobe level. The sidelobe level is supervised by a calibration during operation, using a table of calibration data. The table of calibration data is produced by iterative computer runs of programs performing the two transformations Near-field-to-Far-field and Far-field-to-Waveguide Excitation - the characterization. Characterization to very low sidelobe level in the calculated farfield is possible when using for instance planar nearfield technique to measure an active antenna. The errors at the planar nearfield range are misleadingly compensated for by the characterization. Therefore a minimization together with a continuous control of the noise level is necessary.