AMTA Paper Archive

The ESA Compact Payload Test Range (CPTR) has been designed to allow scanning of the range axis by means of the movement of the feed in the focal region. This capability has been applied to the verification of the performance of the Inter Orbit Link Antenna (IOLA) signal acquisition and tracking system, known as the IAPS, of the ARTEMIS communication satellite. The feed of the CPTR was used to simulate the Ka band signal transmitted from a low earth orbit satellite. The paper describes the test scenario and requirements, as well as presenting the scan performance of the ESTEC CPTR. The scan performance was verified by comparing azimuth scanned patterns of the main beam with those made translation of the feed in the focal region.

This paper presents a new technique for performing range compensation of full sphere antenna patterns measured on fixed line-of-sight antenna ranges where pattern measurements are made over a spherical surface. Such ranges include far-field, compact, and spherical near-field ranges. A plane wave model of the range field illuminating the antenna under test (AUT) is determined as described in another paper. This plane wave model consists of a small, selectable number of plane waves. Equations are given describing the transformation of range coordinates to AUT coordinates. This allows the response of an AUT to a plane wave from an arbitrary direction to be defined using only the far-field pattern of the AUT. The error pattern added to the pattern measurement by the extraneous plane waves is then estimated using the plane wave model and the measured pattern. This error pattern is subtracted from the antenna pattern measurement to obtain a compensated pattern. The compensated pattern and error pattern are improved iteratively. This paper demonstrates the technique using simulated data. The rotation of the spherical AUT grid with respect to the range grid during the measurement requires an interpolation of the measured fields to estimate the error pattern. Investigations of interpolation error are presented. The computational complexity of the compensation algorithm, excluding the plane wave model, is on the order of the number of measurement points on the spherical measurement grid. K

Cellular and PCS basestation antennas are basically arrays with highly directive elevation patterns and broad azimuth patterns. This causes measurement problems because they are large but not directive in both principal planes. As a result, the pattern measurements of these antennas that have been performed outside have been unreliable in many cases because they are very receptive to interference and range clutter. Thus, one wants to move inside but the antenna size can significantly impact the overall range cost. This paper describes a very practical solution to this problem. Since basestation antennas are long and narrow, one can use a near field scanner approach to deal with the length. In fact by using a sectorial horn probe, the narrow dimension of the antenna-under-test is illuminated by a cylindrical wave. Thus, the scanner need only probe the field along the antenna length. This linear scan data can then be transformed to generate the desired far field elevation pattern. The details of this novel design will be described as well as the results, to illustrate the system capability and accuracy.

Until now the mobile phones have been qualified by power measurement at the RF-connector of the handset without any regard to the antenna characteristic and the losses caused by the mismatch of the impedance matching network. IMST is exammmg, via measurements, the user's influence on the antenna pattern of the mobile phone. These measurements were performed in the transmit situation and in the receive situation of the mobile phone at different elevation angles and for different channels of the German E­ Plus-Network. Due to the differences between human bodies and due to the body's movement during a measurement, the emphasis of this investigation was on the development of a model with dimensions and electromagnetic characteristics similar to those of the average human body. By comparing measurement results using different test persons and the model, the validity of the model has been evaluated.

A method is presented for computing far field antenna patterns from spherical near field antenna measurement data. The new method utilizes a novel Uniform Geometrical Theory of Diffraction (UTD) based transformation of spherically scanned antenna tangential electric (or magnetic) near field measured values to more efficiently obtain the antenna far field. Examples illustrating the accuracy and speed of UTD based spherical near to far field transformations for large to moderately large antennas are presented.

The far-field parameters of an antenna are obtained from near-Field measurement with an accuracy that is limited by the sampling area and the sampling rate used to collect the measurement data. It is therefore important to know the relation between the far-field parameters and the sampling parameters. A parametric study of the far field parameters accuracy versus the sampling parameters was made. In order to determine the optimal choice of the sampling parameters to achieve the desired far-field accuracy, planar near-field measurements of a linear array were performed in an anechoid chamber at the Canadian Space Agency. A program performing Fast-Fourier Transform was used to process the data and to obtain spectral domain and reconstruct the far­ field patterns. A methodology developed in [1] was used to compare different spectral and far­ field patterns obtained from different sampling conditions. Parametric curves were developed for the far-field parameters such as gain, beam pointing, beam width, sidelobes, etc.

We review the development and recent performance results of a stand-alone fiber optic based EM field sensor system. The sensor heads are miniature (lcm), electrically passive, and are directly coupled to optical fibers at the remote sensing site. Sensor conversion of EM fields to optical intensity is carried out by mounting small antenna structures directly onto high speed lithium niobate electro-optic modulator chips. Optical power to the sensor head is derived from a stabilized laser which is located within a system chassis at a control room location. Sensor and fiber temperature drift effects are compensated by specialized remote bias control electronics. Recent broad spectrum tests have demonstrated a system bandwidth of about 20 GHz, and a minimum detectable field in the lO's of mV/m. Ultra wideband pulse measurements have demonstrated real time pulse signals of about 2 Vpp for 3 KV/m fields. The sensor system is slated for application in EMI effects such as EM compatibility, and for pin-point near-field and far-field mapping of radiation patterns. The technology is readily scaleable to frequencies exceeding 20 GHz.

Image Editing and Reconstruction (IER) is used to estimate the RCS of component parts of a complex target. We discuss the general areas of controversy that surround the technique, and present a set of practical data processing procedures for assisting in validation of the process. First, we illustrate a simple technique for validating the end-to-end signal processing chain. Second, we present a procedure that compares the original unedited, but fully calibrated, RCS data with the summation of all IER components. For example, if we segregate the image into two components - component of interest, remainder of the target mounting structure plus other clutter - we require that the two patterns coherently sum to the original. This indirectly references the results to the calibration device. In addition, it provides a quantitative means of assessing the relative contribution of the component parts to overall RCS. We demonstrate the procedures using simulated and actual data.

Gabor holography is an appropriate technique for near­ field measurements at THz frequencies when apertures of the order of thousands of wavelengths are involved. The method permits pattern prediction over a restricted angular range from intensity measurements, providing a direct method of recovering phase which overcomes cable, planarity and atmospheric effects; problematic to conventional near-field phase measurements. We demonstrate the feasibility and convenience of the method with an example planar near-field measurement at 94GHz for a 1.1m Cassegrain reflector and we determine the relationships governing dynamic range and the requirements for sampling. Finally, two-dimensional numerical simulations for a lm antenna at 0.5THz, with a 10m scan distance, will be presented to demonstrate the feasibility of the method for large terahertz antennas.

For frequencies above 30 GHz, RCS reference target method is, in general, more accurate than scanning the field by a probe. Application of mechanically calibrated targets with a surface accuracy of 0.01 mm means that the phase distribution can be reconstructed accurately within approximately 1.2 degrees across the entire test zone at 100 GHz. Furthermore, since the same result can be obtained for both azimuth and elevation patterns, all data is available for the characterization of the entire test zone. In fact, due to the fact that the reference target has a well known radar cross-section, important indication of errors in positioning can be obtained directly from angular data as well. In the first place the data can be used in order to recognize the first order effects (+/- 5 degrees in all directions). Applying this data, defocussing of the system reflector or transverse and longitudinal CATR feed alignment can be recognized directly. Furthermore, mutual coupling can be measured and all other unwanted stray radiation incident from larger angles can be recognized and localized directly (using time­domain transformation techniques). Inmost cases even a limited rotation of +/- 25 degrees in azimuth and +/- 10 degrees in elevation will provide sufficient data for analysis of the range characteristics. Finally, it will be shown that sufficient accuracy can be realized for frequencies above 100 GHz with this method.

A simulation tool used during the design of near-field ranges for phased array antenna testing is presented. This tool allows the accurate determination of scanner size for testing phased array antennas under steered beam conditions. Estimates can be formed of measured antenna pointing accuracy, side lobe levels, polarization purity, and pattern performance for a chosen rectangular phased array of specified size and aperture distribution. This tool further allows for the accurate testing of software holographic capabilities.

Antenna measurement techniques historically have been dominated by an assumption that an antenna is a discrete component of the overall electronic system into which it is built. Under this assumption, the measurement technique is to remove the antenna from its host electronic system and place it in a generic test system to measure the gain, pattern, etc. Although this technique still applies to many antenna measurements, it does not work well in cellular/PCS handset measurement applications because cellular/PCS handsets exhibit significant electromagnetic coupling to the human holding the phone. Therefore, the antenna should be measured in situ with a person holding the phone or, for practical reasons, with a mannequin arranged such that it can hold the phone. The mannequin is placed on an azimuth positioner and a near-field probe is moved on a very accurate circular arch from zenith to a significant angle below the mobile phone horizon plane. A description of the chamber and system, and measured results are provided.

In this paper, a near-field time domain scattering measurement technique is described. Near-field measurements are typically performed for radiation applications but not scattering applications. This time domain measurement approach borrows from many of the principles developed in the frequency domain and is ideally suited for broadband scattering characterization. The goal of determining the scattered far-fields of a structure is accomplished by the transformation of near-field data collected over a planar sampling surface. The scattered near-fields were generated with a probe excited by a fast rise time step. In particular, the near-fields were sampled with a second probe and digitized using a digital sampling oscilloscope. The bandwidth of the excitation pulse was approximately 15 GHz. The overall accuracy of this approach is examined through a comparison of the transformed far-field pattern to a numerical calculation.

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.

Two normalized pattern functions appropriate to Ultra-Wideband (UWB) antennas were discussed at the 1992 AMTA meeting [1]. The normalized pattern spectrum is an image showing radiated signal intensity as a function of azimuth or elevation angle and frequency. The spectrum is complex, and thus requires both an amplitude spectrum image and a phase spectrum image to be complete. It is normalized by dividing by the complex radiated signal at the defined boresite angle for the designated antenna. Therefore, on boresite, the normalized pattern spectrum is unity. The normalized impulse response pattern function is the Fourier Transform of the normalized pattern spectrum. This image plots intensity(and polarity) of the real impulse response of the antenna vs time and angle. On boresite, it is a band-limited impulse corresponding to the normalized pattern spectrum. This paper will discuss measurements of seven UWB antennas, and present normalized pattern results of these antennas. The antennas include both off-the-shelf products and experimental prototypes. Included are antennas which have been used for wide-angle UWB SAR imaging, a coherent UWB application where both signal attenuation and dispersion vs angle are important. The results show how pattern behavior can be separated from boresite transfer function information, and how antennas compare in this compact image format.

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.

Probe correction is required to accurately determine the far-field pattern of an antenna from near-field measurements. At Raytheon Primary Standards Laboratory (PSL) in El Segundo, CA, data acquisition hardware, instrument control software, and a mechanical positioning system have been developed and used with an HP Network Analyzer/Receiver system to perform these measurements. Using a three antenna technique, the on-axis and polarization parameters of a linearly (or circularly) polarized probe are calibrated. The relative far-field pattern of the probe is then measured utilizing the two nominal, orthogonal polarizations of the source antenna. All measurements are stepped in frequency and use a time domain gating technique. The probe and the source antenna are optically aligned to the interface and unique, kinematic designed interface flanges allow repeatable mounting of the antennas to the test station.

Practical antenna applications require accurate characterization of the antenna, including both the amplitude and phase performance. Recent advances in antenna measurement technologies allow the antenna to be measured in various indoor facilities with a well controlled environment. However, measurements that take a long time to complete can still suffer phase drift and variation due to the movement of RF cable as well as changes in the chamber environment. Without proper phase correction, the measured antenna pattern performance may not satisfy the desired requirement. Consequently, it is very important to have appropriate methods for phase correction in order to obtain more accurate results. In this paper, a simple procedure for phase correction of volumetric spherical near field antenna measurement is presented. In this method, only a few additional measurements are needed for correcting the phase variation observed in the original volumetric pattern. Application of the phase corrected pattern has been found to satisfy the desired antenna performance.

The Sirius 2 telcommunication satellite was build in France by Aerospatiale. As a subcontractor Saab Ericsson Space (SES) developed the telecommunication antenna for direct television broadcast. The satellite was successfully launched November 13, 1997. Three antennas were manufactured by SES: a quality model (QM), a flight model (FMl) and a flight spare (FM2). Each antennas consists of a 1.4 meter in diameter shaped main reflector fed by a shaped subreflector and a dual polarized feed horn. For the test of the antennas, spherical near-field antenna test ranges located at Ericsson Microwave System (EMW)/SES in Sweden and at the Technical University of Denmark (DTU) were used. Each of the three antennas was measured twice. Between the two measurements mechanical and thermal tests were performed. The paper presents the measurements on the satellite antennas together with a discussion of the advantages of using the spherical near-field technique for this type of measurements. Compared to a far-field range the advantages are evident: At both SES and DTU a measurement distance of ten and six meters respectively were used on the indoor ranges. On a far-field range a measurement distance in the excess of 250 meters must be applied. To decrease the measurement time the near fields were only measured in a certain region on the near field sphere. The influence of this truncation will be discussed. Coordinate systems for the antennas were defined using mirror cubes. The RF measurements as well as the optical measurements on the cubes were performed without dismounting the antenna from the antenna positioner. The radiation patterns are therefore precisely decined with respect to the coordinate systems of the cubes.