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.)
We examine how accurately the transmit and receive parameters of a radar cross section measurement system can be determined by use of a rotating dihedral as the polarimetric calibration device. We derive expressions for the errors due to misalignment in the angle of rotation. We obtain expressions for the angles a0,hv and a0,vh for which the measured cross-polarization ratios of a target vanish. Since the theoretical cross-polarization of a cylinder is 0, we can .nd the calibration bias-correction angles. We use simulated and real data to demonstrate the robustness of this bias-angle correction technique. We derive expressions for the uncertainty in the polarimetric system parameters.
In previous work [1], we presented an antenna pattern compensation technique for linearly-scanned near field measurements. In this paper, we present a similar technique to mitigate the errors from uncompensated azimuthal antenna pattern effects in circular near-field monostatic radar measurements. The antenna pattern co mpensation is implemented as part of an improved algorithm for transforming the near-field measurements to the far-field RCS. A description of this improved circular near field-to-far field transformation CNFFFT technique for isotropic antennas is presented in a companion paper [2]. In this paper, we formulate the near-field signal model in the presence of an azimuthal antenna pattern under the same scattering approximation used in the isotropic CNFFFT. Using this model, we derive a modified version of the CNFFFT that includes antenna pattern compensation. Numerical simulations are presented that demonstrate the ability of the technique to remove antenna pattern errors and improve the accuracy of the far field RCS patterns and sector statistics.
ABSTRACT A fast and accurate technique is proposed in this work for the far field evaluation from a nonredundant number of voltage data collected by using the planar wide-mesh scanning (PWMS). It relies on the nonredundant sampling representations of the electromagnetic field and on the optimal sampling interpolation expansions of central type. By using a very flexible source modelling, which fits very well a lot of actual antennas, a new sampling technique is developed to recover the plane-rectangular data from the knowledge of the PWMS ones. It must be stressed that the so developed near-field–far-field transformation requires a number of data remarkably lower than that needed by the standard plane-rectangular scanning. Some numerical tests, assessing the accuracy of the technique and its stability with respect to random errors affecting the data, are reported.
Antenna systems are increasing in complexity at a rapid pace as advances are made in electronics, signal processing, communication, and navigation technologies. In the past, antenna design requirements have focused on parameters such as gain, efficiency, input impedance, and radiation pattern (e.g., beamwidth and sidelobe level). For some new systems, the group delay characteristics of the antenna are important, where the group delay is proportional to the derivative of the insertion phase as a function of frequency. The group delay is required to stay within certain bounds as a function of frequency and pattern angle. Unfortunately, there are not well established methods or standards for calibrating antenna group delay like the standard methods used for gain and input impedance. This paper presents a method for calibrating the group delay of three antennas based on an extension of the widely used three-antenna gain and polarization calibration methods. No prior knowledge of the gain or group delay of the three antennas is required. The method is demonstrated by a measurement example where it is shown that multipath errors and time gating can be critical for calibrating the group delay.
This paper presents a modification to the standard three-antenna polarization measurement method. The new technique solves for the sense, axial ratio, and tilt angle utilizing a least-squared errors routine and multiple measurements of the response at different roll angles between antennas. The paper compares the results of this method to Allan Newell’s well known modified three-antenna polarization measurement technique. Four antennas were measured two at a time and in several different arrangements to get twenty-four measures of the polarization parameters for each antenna. The work shows this method had a more repeatable measure of the axial ratio than the parameters determined using Newell’s technique.
Parallax occurs in antenna measurements when the antenna under test (AUT) is located off the center of rotation (COR) of the test positioner axis. As the AUT is rotated while located off of the COR of the axis, the angle to the AUT as viewed from the source antenna is different than the angle to which the positioner is commanded. This results in a distortion of the antenna pattern, and can result in errors in beam shape and beam width. Knowledge of the test geometry allows for the determination of an appropriate mapping from the recorded test angles to the actual angles to the AUT as viewed from the source. This, in turn, allows for the possibility that the antenna pattern may either be corrected for the parallax error, or measured at the correct angles in order to avoid pattern distortion. ORBIT/FR has implemented a parallax correction in the 959Spectrum Antenna Measurement Workstation software that allows for flexibility in positioning angle correction, and in addition provides a useful tool for implementing unusual measurement test scenarios, such as measuring antenna data at a “list” of angles. This paper describes the parallax problem, the implemented solution, and provides examples of use of the implemented software feature.
Reflections in antenna test ranges can often be the largest source of measurement errors, dominating all other error sources. This paper will show the results of a new technique developed by NSI to suppress reflections from the radome and gantry of a large hemi-spherical automotive test range developed for Nippon Antenna in Itzehoe, Germany. The technique, named Mathematical Absorber Reflection Suppression (MARS), is a post-processing technique that involves analysis of the measured data and a special filtering process to suppress the undesirable scattered signals. The technique is a general technique that can be applied to any spherical near-field test range. It has also been applied to extend the useful frequency range of microwave absorber in a spherical near-field system in an anechoic chamber. The paper will show typical improvements in pattern performance and directivity measurements, and will show validation of the MARS technique using data measured on antennas in a conventional anechoic chamber.
ABSTRACT A unified theory of near-field spiral scans is proposed in this work by introducing a sampling representation of the radiated electromagnetic field on a rotational surface from the knowledge of a nonredundant number of its samples on a spiral wrapping the surface. The obtained results are general, since they are valid for spirals wrapping on quite arbitrary rotational surfaces, and can be directly applied to the pattern reconstruction via near-field–far-field transformation techniques. Some numerical tests, assessing the accuracy of the technique and its stability with respect to random errors affecting the data, are reported with reference to the case of the helicoidal scan.
Inversion of the material parameters for a sample usually requires that the sample fill the waveguide cross-section. Alternative methods require that a non-filling sample be aligned along the center-line of the waveguide. However, it is not known how errors in placement impact the accuracy of the inversion. Hence, a numerical simulation to assess these errors is beneficial to the community. The extraction of the S-parameters from a rectangulardielectric-filled waveguide is conducted numerically by means of the Finite Element Method (FEM) and the Thru-Reflect-Line (TRL) calibration technique. Three different ratios of dielectric sample width (d) to waveguide width (a) are primarily studied. The results are then validated with experimental data on the X-band. An assessment of error with respect to position will be presented at the meeting.
Standard Gain Horns (SGH) are normally used as reference antennas in antenna measurements. Gain charts for SGH are provided by the supplier. These charts give the gain of the SGH in dBi versus frequency but do not provide any information on the phase variations versus frequency. For complete antenna calibration, one needs the phase as well as gain data for SGH over the frequency band of interest. To obtain the gain and phase data, one can use the three-antenna method which requires three independent measurements and, therefore, is more susceptible to measurement errors. Note that if one has access to two identical antennas, the three-antenna method reduces to a single measurement which is more desirable. In practice, however, one does not have access to two identical antennas. In this paper, a novel method which mimics measurements with two identical antennas is described. In the method, one performs S11 type measurements on the antenna of interest by placing the antenna in front of a large conductive flat plate. The late term in the S11 measurements is then used to obtain the boresight gain and phase of the antenna under test. The measured gain and phase data of several antennas obtained using the proposed method is presented and compared with the results obtained using the three-antenna method as well as with analytical results.
Abstract Historically, most commercially available Local Oscillator (LO) distribution systems have used a closed loop control system to set LO power levels. As frequency switching times decrease, the time constant of the control loop can introduce errors to precision antenna measurements. This paper will discuss the basics of the remote mixing test system, and then discuss the limitations of various current approaches. Finally, it will introduce an open loop LO distribution system, and discuss the design considerations for the various components of the system.
ABSTRACT This paper describes a broadband radome measurement method that provides insertion loss performance referenced to circularly polarized radiation. The measurements are performed using linearly polarized sources and post processing is employed to convert to circular polarization. The method reduces measurement errors encountered using circularly polarized sources that traditionally have poor cross polarization isolation.
We estimate uncertainties in the test antenna transmitting function due to uncertainties in the near- field measurements and in the probe receiving function.
In this paper, we present an analysis of the impact of positioning errors on the performance of the GDAIS circular image-based near field-to-far field RCS transformation (CNFFFT). The analysis is part of our continuing investigation into the application of near fieldto-far field transformations to ground-based signature diagnostics. In particular, the analysis focuses on the errors associated with ground-to-ground, near-field, whole-body measurements where the radar moves on a nominally circular path around the target. Two types of positioning errors are considered: slowly-varying, long term drift and rapidly-varying, random perturbations about the nominal circular path. The analyses are conducted using simulated data from a target comprised of an array of generalized point scatterers which model both single and multiple interactions on the target. The performance of the CNFFFT was evaluated in terms of the angle sector cumulative RCS statistics. The analyses were performed as a function of frequency for varying amounts of position error, both with and without (approximate) motion compensation. As expected, the results show that the CNFFFT is significantly more sensitive to rapidly-varying position errors, but that acceptable performance can be achieved with motion compensation provided an accurate estimate of the errors is available.
There is a trend within the RCS community to use squatty cylinders in place of spheres for calibration. A higher degree of accuracy can be achieved; however, cylinder calibrations require much more precision in the alignment procedures. This effort is doubled when the dual calibration target is also a cylinder. The dual calibration test article could be a sphere thus reducing calibration efforts as long as good correlation exists between theory and measurement sphere data. A series of measurements were collected at the NASA Langley Research Center Compact Range Pilot Facility to study measurement errors of spheres atop foam columns to determine their feasibility for dual calibration use.
An internal research and development project at the Georgia Tech Research Institute (GTRI) focused on cohering multiple apertures into a single distributed aperture. Cohered distributed aperture antenna patterns were collected on the GTRI far-field range for a 1.5 GHz bandwidth at X-band frequencies. Both 1-way and 2-way antenna patterns were measured, with the 1-way antenna pattern measurement requiring coherence on receive only and the 2-way antenna pattern measurement requiring coherence on transmit and receive. The resulting data were compared with the ideal angular resolution and power-aperture gain product improvements from a perfectly cohered distributed aperture, and the results are presented. As measurement techniques were developed for collecting 1-way and 2-way antenna pattern data, sources of potential errors in measurement collection and aperture coherence were identified, with potential methods of error mitigation outlined.
Measurements of antenna array’s patterns are usually taken with feed networks, but for some digital beam forming arrays, feed networks are not included. To measure such arrays in traditional way, we must design a feed network first, which is too complex and inefficient when steering is required. A VNA can act as a multichannel receiver to form a digital beam forming array for testing. Today’s VNA may have many test ports (for example, 9-ports in Agilent E5091A multiport test set and 11-ports in Advantest R3986 multiport test set.) designed for multiport devices, which is very suitable for measuring small arrays without using feed networks. Another advantage of this method is that it can eliminate errors from imperfect feed networks. Automatic measurement program is required to calculate array patterns from S-parameters, which is easily developed using Labview or Matlab. Test results of a 3-elements adaptive anti-jam array with different jam DOA are demonstrated.
For enhancing the performance of existing near field antenna test facilities it is quite reasonable to use both conventional (the amplitude and phase measurements) and the phaseless measurements techniques during electrically scanning phased array antennas (PAA) testing. This simple yet critical approach helps to improve the quality of PAA alignment and testing reducing measurement errors and saving costs. In this way many difficulties related to precise phase measurements are overcome. Both simulation and measurement results will be presented to demonstrate the utility of such approach to PAA alignment and determination of its parameters. Comparison will be made between the PAA patterns for electrically scanned beams calculated using traditional near field - far field (NF/FF) transformations, the phaseless methods and the results obtained applying both measurement techniques.
Geometries for measuring radome characteristics can usually be split into two categories. The first category always has the antenna inside the radome pointing along the range axis. The second category has the antenna maintaining a fixed relationship with respect to the radome during each scan of data. A facility can generally be designed to minimize measurement errors in one of the two geometries, but not both. Many facilities that permit collection of data in both geometries would benefit from the ability to dynamically capture data that lead to measurement errors, then compute and remove the associated errors. This paper discusses some of the primary error contributors in a dual-geometry radome measurement system, and suggests some mechanisms for capturing and potentially removing those errors.
We present a first-order analysis of the RCS errors resulting from non-uniform ground-bounce illumination in mobile, ground-to-ground, diagnostic RCS measurements of aircraft. For the case of a non-planar ground surface, these errors are a function of both aspect angle and position on the target. We quantify the errors in terms of their impact on the sector mean RCS as a function of position on the target. For typical targets, we show that the mean RCS error increases significantly for points displaced (either horizontally or vertically) from the calibration point. Conversely, the sector mean RCS is relatively insensitive to small-scale variations in the height of the ground, even though the errors at a single frequency and aspect angle can be quite large.