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.)
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.
Errors arising in the measurement of reflection coefficient are identified and analyzed. The presence of multiple reflections due to poor connectors, transmission line discontinuities, and terminal loads is described, modeled and applied. Various measurement scenarios are analyzed, and measured results are presented as a guide for laboratory troubleshooting and as a validation of the measurement models. Improvements to Vector Network Analyzer calibration methods are proposed, including computer corrected calibration for one-port radiating elements and elementary improvements to two-port TRL calibration. An extensive error evaluation of the somewhat forgotten slotted line measurement is finally presented as a robust alternative, and computer automation, acquisition, and calibration of this measurement is outlined.
Measurements of the ERP radiated by an antenna and the ERP received from a distant antenna are addressed. Alternative measurement techniques are described and correction for polarization mismatch loss, pointing error and propagation loss is discussed. The statistics of the measurement errors are presented for error budget projections of measurement accuracy.
The close proximity of the ground to the radar antenna and the target under test is often hard to avoid at an outdoor RCS measurement range. Ground reflection of energy from the antenna leads to target illumination errors, and target-ground interactions lead to multipath errors. By proper positioning of the antenna and target, ground reflections of the antenna illumination can be exploited to increase overall system sensitivity by concentrating more energy on the target; however, this is only effectivefor narrowband measurements over a limited target region [1]. Reducing target-ground interactions by increasing the target height above the ground generally has limits due to mechanical restrictions on both the radar antennas and the target. This paper will present a model-based data post-processing technique to mitigate illumination errors and target-ground interactions in ground plane range RCS measurements. The algorithm is an extension of the network model multipath mitigation technique previously developed for indoor RCS measurement ranges [2,3,4]. The technique will be described and demonstrated using a numerical simulation of the RCS measurement of a canonical target over a ground plane.
In-situ pattern measurement of JHU/APL's 60-foot parabolic reflector antenna (S-band), using a low-earth orbit satellite as the source is described. The signal strength and X and Y tracking error voltages are measured as the antenna dish sweeps a matrix of points around the position of the moving satellite. The swept region is approximately ±0.30° from the antenna's boresight. This technique was evaluated during April 1998. This measurement was used to baseline the current performance of the ground station before the feed underwent significant modifications. Before the new feed assembly was installed, the position of the current feed was translated to the new feed assembly. Once installed the performance of the reflector was verified. Misalignment of the feed broadens the main beam and increases the sidelobes. More importantly, the inclusion of new components inside the feed also has the potential to introduce phase errors onto the tracking signals. These phase errors will be translated by the auto-track electronics into pointing errors causing the antenna system to inaccurately follow a target. This paper describes the measurement of the reflector antenna pattern and tracking pattern before the new assembly was installed. Results of pattern measurements with the new assembly will be presented at the conference
The use of dual polarization in meteorological radars offers significant advantages over single polarization. Recently a standard single-polarization Cuband radar was upgraded to operate in dual-polarization mode. The antenna has a 4.2m diameter parabolic reflector with a prime-focus feed. A spherical Fresnel-zone holographic technique was used to obtain the radiation pattern for the upgraded antenna. The sidelobes were higher than predicted and so the data was analyzed to identify the relative contributions of shadowing from the feed crook and surface errors in the dish. This paper describes practical considerations in the measurement of this antenna and the analysis of the results.
Three common methods of measuring circularly antennas on a far-zone range are: using a spinning linear source antenna (SPIN-LIN), measuring the magnitude and with a linearly polarized source antenna in two orthogonal positions (MAG-PHS), and using a circularly polarized source antenna (CIRC-SRC). The MAG-PHS and CIRC-SRC methods are also used in a near-field or com pact range. The SPIN-LIN method is useful because an accur te measurement of the axial ratio and gain can be made without the need to measure phase. The MAG-PHS method is the most general method and can also completely characterize the polarization of the test antenna. The CIRC-SRC method is the simplest and least time-consuming measurement if the antenna response to only one polarization is needed. The choice of measurement method is dictated by schedule, accuracy requirements, and budget. An analysis is presented that provides errors in the measured gain, relative gain pattern, and phase of the test antenna depending on the polarization characteristics of the source and test antennas. These results are useful for deciding which measurement method is the most appropriate to use for a particular job. These results are also useful when constructing more complete error budgets.
Concise mathematical relations have been derived for Planar Near-Field measurements that quantify the effects of x, y and z-position errors on antenna parameters such as gain, sidelobe level, pointing, and cross polarization. Because of the complexity of the theory, similar relations for spherical near-field measurements have not been developed. The requirements for the spherical coordinate system are generally defined in terms of the alignment parameters such as orthogonality and intersection of axes, q-zero, x zero and y-zero rather than individual errors in q , f and r. Mechanical, optical and electrical techniques have been developed to achieve these alignments. This paper will report on the development of methods to estimate the antenna parameter errors that will result from spherical alignment errors for typical antennas.
Recently, RCS measurements were made of several common calibration objects of various sizes in the Boeing 9-77 Range. A study was conducted to examine the accuracy and errors induced by using each as a calibration target with a string support system. This paper presents the results of the study. Two of the objects, i.e., the 14"-ultrasphere and the 4.5"-dia. cylinder, are found to perform the best in that they exhibit the least departures (error) from theory. The measured departures of 0.2 to 0.3 dB are consistent with the temporal drift of the radar in several hours.
Calibration of monostatic radar cross section (RCS) has been studied extensively over many years, leading to many approaches, with varying degrees of success. To this day, there is still significant debate over how it should be done. It is almost a certainty, that if someone proposes a way to calibrate RCS data, someone else will come up with reasons as to why the "new" approach will not yield results that are "good enough." In the case of full scattering matrix RCS measurements, the lack of information concerning calibration techniques is even greater. The Air Force's Radar Target Scattering Facility (RATSCAT) at Holloman AFB, NM,has begun an effort to refine monostatic and bistatic cross polarization measurements at various radar bands. For the purposes of this paper, we have concentrated on our monostatic cross polarization developments. Such issues as calibration targets and techniques, system stability requirements, etc. will be discussed. During several programs we have attempted to collect sufficient data to do full scattering matrix corrections. In a previous paper, "Bistatic Cross-Polarization Calibration," our collected data had a high background which obscured much of the cross polarized return. The data presented here is from a program conducted at RATSCAT recently which utilized the Ka band. Because of the sensitivity of measurements at Ka to many effects, an error estimate was required. This paper presents this error estimation and some results of full scattering matrix correction of RCS data. This analysis is based upon "The Proposed Uncertainty Analysis for RCS Measurements", NISTIR 5019, by R. C. Wittmann, M. H. Francis, L. A. Muth and R. L. Lewis. This paper was aimed at principle pole measurements, e.g. HH and VV. The tabular data presented in the paper are from this paper with additions for errors associated with cross polarization and cross polarization correction.
In the last few years, a change has occurred in the RCS metrologist concerns for error analysis and the quantification of measurement uncertainty. The specific methods for range characterization and uncertainty estimation are the topics of many passionate technical discussions. While no single treatment can please everyone, most agree a measurement uncertainty program is critical to the understanding of measurement quality, the development of error reduction strategies, and to the planning of range improvement paths. We present the statistical case for the natural grouping of errors into multiplicative and additive classes. We will derive the two cases where one class dominates as presented by LaHaie [1], and then expand the analysis to include the general case of competing classes. We summarize the role and applicability of this method in estimating measurement quality and discuss how this procedure offers a logical and comprehensive error propagation solution to both top-down and bottom-up range characterization approaches.
Nearfield Systems, Inc. (NSI) has delivered the world's largest vertical near-field measurement system. With a 30m by 16m scan area and a frequency range of 1GHz to 50GHz, the system consists of a robotic scanner, laser optical position correction, computer and microwave subsystems. The scanner and microwave equipment are installed in an anechoic chamber 40m in length by 24m in width by 25m in height. The robotic scanner controls the probe positioning for the 33m by 16m vertical scanner using X, Y, Z and polarization axes. The optical measurement package precisely determines the X and Y axes position, alignment errors along the X and Y axes, and Z-planarity over the XY scan plane.
Offset parabolic reflector Compact Ranges are limited for cross polarization measurements in comparison to compensated dual reflector systems. This means that, in some cases, the crosspolar measurements at low levels show a significant content of the compact range reflector cross polar. An investigation has been carried out at INTA to reduce the crosspolarization measurement errors levels to those of a compensated dual reflector system by the application of vector deconvolution techniques. Results are shown of the validation of the algorithm in a far-field range where a crosspolar field is introduced by depointing the transmitter antenna.