AMTA Paper Archive

The need for a well-defined accuracy estimate in antenna measurements requires identification of all possible sources of inaccuracy and determination of their influence on the measured parameters. For anechoic chambers, one important source of inaccuracy is the reflection from the absorbers on walls, ceiling, and floor, which gives rise to so-called stray signals that interfere with the desired signal. These stray signals are usually quantified in terms of the reflectivity level. For near-field measurements, the reflectivity level is not sufficient information for estimation of inaccuracy due to the stray signals since the near-to-far-field transformation of the measured near-field may essentially change their influence. Moreover, the inaccuracies are very different for antennas of different directivity and with different level of sidelobes, and for different parts of the radiation pattern. In this paper, the simulation results of a spherical near-field antenna measurement in an anechoic chamber are presented and discussed. The influence of the stray signals on the directivity at all levels of the radiation pattern is investigated for several levels of the chamber reflectivity and for different antennas. The antennas are modeled by two-dimensional arrays of Huygens' sources that allow calculation of both the exact near-field and the exact far-field. The near-field with added stray signals is then transformed to the far-field and compared to the exact far-field. The copolar and cross-polar directivity patterns are compared at different levels down from the peak directivity.

Calibrated probe complex pattern data is used in planar NFR (near field range) data processing to remove the effects of the probe on the measurement. In a prior paper [1] we proposed a procedure to estimate the measurement error (uncertainty) introduced into a near field antenna radiation pattern measurement due to test frequencies that do not coincide with available calibration frequencies of the range probe. Our prior paper resulted in a “19th term” which was added to the well known NIST NFR 18 Term Error Table used to evaluate the unavoidable uncertainty of far-field radiation patterns derived from a near field scan of a given AUT (antenna under test). A limitation of this procedure, pointed out in our prior paper, is that it was most accurate for a test frequency falling midway between two nearest neighbor probe calibration frequencies. The estimated uncertainty became overly pessimistic as the test frequency of interest moved closer to one of the neighboring calibrated frequencies. The procedure is improved in the present paper by the inclusion of a new term that is a function of the test frequency and the two nearest neighbor probe calibration frequencies. Examples are shown of the use of the new procedure to obtain an improved estimate of this measurement uncertainty and to create the 19th term for use with the standard 18 Term Error Table.

In previous AMTA presentations, we developed and evaluated an image-based near field-to-far field transformation (IB NFFFT) algorithm for monostatic RCS measurements. We showed that the algorithm’s far field RCS pattern prediction performance was quite good for a variety of frequencies, near field measurement distances, and target geometries. In this paper, we quantify the statistical RCS prediction performance of the IB NFFFT using simulated data from a generalized point scatterer model and method of moments (MoM) code, both of which allow modeling of targets with single and multiple interactions. It is shown that the predicted RCS statistics remain quite accurate under conditions where the predicted far field patterns have significantly degraded due to multiple interactions and other effects.

This paper describes a unique three axis gimbal assembly used to test the performance of small radomes. The gimbal assembly supports thin, low profile antennas within the radome, and achieves the correct orientation of the antenna relative to the radome as the entire radome/antenna assembly is rotated for measurement of parameters such as transmission loss and boresight error. The gimbal provides roll, elevation, and azimuth compensation for the antenna within the radome using a small package, with the gimbal point located very close to the rear of the antenna. The axes are equipped with high resolution encoders to provide the very accurate antenna positioning required to demonstrate compliance with tight boresight error tolerances for the radome under test. The entire assembly is removable from the master positioning system for the purpose of switching the test range configuration from radome testing to standard antenna pattern testing.

This paper refers to a special type of antenna, called frequency independent antenna, used in Stepped Frequency Continuous Wave (SFCW) radar employed for humanitarian demining. The radar transmits 128 frequencies within the frequency range from 400 MHz to 4845 GHz, in groups of 8 simultaneously transmitted frequencies. It has been built at the International Research Center for Telecommunications transmission and Radar (IRCTR), Delft University of Technology. Two Archimedean spiral antennas with opposite sense of rotation, in order to decrease coupling signal below –55dB, have been chosen. Precise antenna behavior characterization is needed because SFCW radar is phase sensitive. The paper is focused on antenna footprint measurements, translating data from frequency domain to time domain and gating in order to remove any unwanted signals. Some phase and amplitude pattern using gating measurements are presented.

A prototype design of the dielectric rod antenna is discussed. This novel design is suitable for nearfield probing application in that it provides broad bandwidth, dual-polarization and low RCS. The design details are provided in this document along with measurement data associated with important antenna characteristics such as VSWR and far-field radiation pattern

Occasionally, antenna patterns have discontinuities or “glitches” in them. While most of these glitches are obvious to a human looking at the plot, it can be difficult for a computer to automatically identify glitches while ignoring sidelobes and other real features of the antenna pattern. This paper will present a technique for accurately identifying and removing antenna pattern glitches through the use of higher order derivative information.

Probe position errors, specifically the uncertainty in the theta and phi position of the probe on the measurement sphere, are one of the sources of error in the calculated far-field and hologram patterns derived from spherical near-field measurements. Until recently, we have relied on analytical results for planar position errors to provide a guideline for specifying the required accuracy of a spherical measurement system. This guideline is that the angular error should not result in translation along the arc of the minimum sphere of more than ?/100. As a result of recent simulation and analysis, expressions have been derived that relate more specifically to spherical near-field measurements. Using the dimensions of the Antenna Under Test (AUT), its directivity, the radius of the sphere (the minimum sphere) enclosing all radiating surfaces and the frequency we can estimate the errors that will result from a given position error. These results can be used to specify and design a measurement system for a desired level of accuracy and to estimate the measurement uncertainty in a measurement system.

The direction of arrival of multiple coherent electromagnetic signals can be determined by measuring the pattern of an antenna probe when it is rotated off its phase center and then exciting a synthetic array with the same geometry as the probe measurement points using the signals received in the measurement. The offset and angle of sweep, which defines the aperture size required for separating the waves, depends upon the resolution required. The sampling resolution must also fall within the Nyquist sampling criteria.

This paper describes a family of new measurement systems, termed “test cells”, designed to satisfy the certification requirements of the Cellular Telephone & Internet Association’s (CTIA) “Method of Measurement for Radiated RF Power and Receiver Performance” test plan for wireless subscriber stations. These test cells employ simultaneous dual-axis mechanical scanning and operate in both far-field and near-field modes over the 750MHz to 6 GHz frequency range. Operation can be extended to higher frequencies through the use of suitable sampling antennas. Test cell facility configuration is detailed. Scanner layout and RF sampling antenna designs are discussed. Anechoic chamber characterization data is presented along with typical measured pattern and efficiency data for both broadbeam and directive AUT’s. Measurement test times for various test scenarios are discussed.

The Cellular Telecommunication and Internet Association has developed a ripple test measurement for qualifying the quiet zone of wireless pattern measurement systems for their Mobile Station Over the Air Test Plan. The data produced by this ripple test provides a very thorough characterization of the worst possible contributions to an antenna pattern measurement performed on the qualified system. However, the characterization represented by the maximum ripple significantly overestimates the ripple seen on typical pattern measurements produced by the qualified system, and greatly overestimates the actual uncertainty involved in the determination of integral quantities such as Total Radiated Power (TRP). In order to better account for the results of this test, a statistical analysis method referred to as the Surface Standard Deviation (SSD) has been developed to determine an expected uncertainty for surface integral quantities. This paper will present the background and formulation of the SSD method and show some typical results.

A novel structure for accurate measurements of antennas mounted on an infinite ground plane has been designed and built. The structure is eight feet in diameter and can be used to measure antennas as big as fourteen inches at the base at frequencies as low as 1 GHz. The structure is defined by blending a planar surface with an elliptical surface such that near the antenna under test the surface resembles a planar surface and then it slowly rolls back to minimize any diffractions due to discontinuities in the surface. Patterns of a few antennas mounted on the structure are presented and compared with the expected patterns of the antennas mounted on an infinite ground plane.

In this paper, the electromagnetic (EM) characteristics of various antennas implanted in both the human head and the human body are analyzed for biomedical applications such as hyperthermia and biotelemetry. The implanted antennas are studied in two ways: the near- and far-field patterns of the antenna are calculated and the potential effects on the human body are observed. To ensure the correctness of the results, we apply two simulation methodologies: dyadic Green’s function (DGF) expansions and finite difference time domain (FDTD). We characterize the performances of the low profile antennas designed for biomedical applications in terms of specific absorption rate (SAR), radiation patterns, maximum available power and safety issues. These results should also provide a good basis for validating the results of experimental data.

In a previous AMTA paper [1], we presented a firstprinciples algorithm called wavenumber migration (WM) for estimating a target’s far-field RCS and/or far-field images from extreme near-field linear (1-D) or planar (2-D) SAR measurements, such as those collected for flight-line diagnostics of aircraft signatures. However, the algorithm assumes the radar antenna has a uniform, isotropic pattern on both transmit and receive. In this paper, we describe a modification to the (1-D) linear SAR WM algorithm that compensates for nonuniform antenna pattern effects. We also introduce two variants to the algorithm that eliminate certain computational steps and lead to more efficient implementations. The effectiveness of the pattern compensation is demonstrated for all three versions of the algorithm in both the RCS and the image domains using simulated data from arrays of simple point scatterers.

A fully-functional Fresnel Zone (FZ) antenna was designed and measured using PO simulation programs and the bi-polar near-field facility. The results from these measurements and simulation are presented in this paper. First, a detailed description of an FZ antenna and its operation is given. Then, a discussion of the design and construction procedure for both the FZ antenna and supporting structure is included. The resulting far-field pattern, near-field plots, and holographic images are shown in this paper. The antenna was measured with different feed positions to observe how it affects the overall antenna performance.

A new spherical near-field probe positioning device has been designed and constructed consisting of a large 5.0 meter fixed arc. This arc has been installed in a near-field test facility located at Alenia Marconi Systems on the Isle of Wight, UK. As part of the nearfield qualification, testing was performed on a ground based radar antenna. The resultant patterns were compared against measurements collected on the same antenna on a large outdoor cylindrical near-field test facility also located on the Isle of Wight [1]. These measurements included multiple frequency measurements and multiple pattern comparisons. This paper summarizes the results obtained as part of the measurement program and includes discussions on the error budgets for the two ranges along with a discussion on the mutual error budget between the two ranges.

A Wideband Optically Multiplexed Beamformer Architecture (WOMBAt) was developed and characterized at the Crane Naval Surface Warfare Center Active Array Measurement Test Bed (AAMTB) facility. The project includes development and integration of the true-time delay (TTD) WOMBAt photonic beamformer with the Active Array Measurement Test Vehicle (AAMTV). The AAMTV is a 64-channel transmit-receive (TR) module based phased array beamformer that is integrated with the AAMTB facility 12’x9’ planar near-field scanner. The AAMTV provides phase trimming and a small amount of delay using electrical components while the WOMBAt provides longer delays using commercial-off-the-shelf (COTS) optical components typically manufactured for the telecommunication industry. By integrating the WOMBAt with the AAMTV, a highly flexible test environment was achieved that includes system calibration, multi-frequency scanning, and antenna pattern analysis. Phase I receive tests for this system were previously described and presented to AMTA[1] in 2002. This paper will describe the results of reconfiguring the AAMTV into a transmit architecture for Phase II. WOMBAt successfully demonstrated wideband TTD in both receive and transmit configurations at angles greater than the system goal of ±65º while exceeding all other system level performance goals. System level performance included a beam squint of less than 1.1º for receive and 0.5º for transmit, a worse case amplitude variation of 2.4 dB receive and 1.6 dB transmit and differential delays of less than 3.5 picoseconds.

The recent launch and successful orbiting of the EO-1 Satellite has provided an opportunity to validate the performance of a newly developed X-Band transmitonly phased array aboard the satellite. This paper will compare results of planar near-field testing before and after spacecraft installation as well as on-orbit pattern characterization. The transmit-only array is used as a high data rate antenna for relaying scientific data from the satellite to earth stations. The antenna contains distributed solid-state amplifiers behind each antenna element that cannot be monitored except for radiation pattern measurements. A unique portable planar near-field scanner allows both radiation pattern measurements and also diagnostics of array aperture distribution before and after environmental testing over the ground-integration and pre-launch testing of the satellite. The antenna beam scanning software was confirmed from actual pattern measurements of the scanned beam positions during the spacecraft assembly testing. The scanned radiation patterns on-orbit were compared to the near-field patterns made before launch to confirm the antenna performance. The near-field measurement scanner has provided a versatile testing method for satellite high gain data-link antennas.

This paper deals with different aspects of the on-ground antenna pattern characterization of the MIRAS radiometer for ESA’s SMOS mission. Various technical challenges of the project are briefly described. Special attention is given to the effect of multiple reflections between the antenna under test and the measurement probe. A series of antenna measurements of the MIRAS radiometer antennas is now on-going at the DTU-ESA Facility. The main objectives of these are to investigate the accuracy of the forthcoming antenna characterization, to find solutions to the already known problems, to identify new possible difficulties, and to establish an optimal measurement strategy, which should allow for the tight error requirements and minimize the overall time of the measurement campaign.

As known, spacecraft antenna design is an everdemanding task, due to the tight requirements on performance and the small available space. Moreover, the trend of installing array antennas onboard makes this task even more complex. For this reason, every antenna design must be verified against its installation constraints, in term of pattern distortion, due to interaction with spacecraft structure, inter-antenna coupling with nearby antennas, which can be remarkable due to the reduced available space, and generation and propagation of passive inter-modulation products (PIM), that can seriously affect the performance of transponders. For the above reasons, a Design Framework has been developed in the frame of an ESA contract. In this activity, European universities, industry and research centers cooperated in order to integrate within a single environment different prediction codes providing the required modeling capabilities. The system is able to guide the user from the antenna design phases through antenna installation simulation. The framework also allows for storage and management of experimental data in the more common formats or in user-defined ones, making able the designer to validate its numerical models with measured data obtained from intermediate breadboards. A validation activity is in progress and comparisons between simulation and measurements are reported, together with main characteristics of the design system. The system has been applied, among others, in the design and compatibility analysis, of Galileo antennas.