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.)
This paper describes the use of a two-dimensional chirp z-transform (2D-CZT) to efficiently concentrate a large number of sample points in a single portion of the far zone without interpolation. This work presents the equivalence of transforms calculated from measured near-field data using both the 2D-CZT and 2D-fast Fourier transform (FFT). The paper also shows that the 2D-CZT is computationally more efficient than a zero-padded FFT when one requires a high resolution over a small area of the pattern.
ABSTRACT An iterative probe correction technique for spherical near-field antenna measurements is examined. This technique has previously been shown to be ideally-suited for non-ideal first-order probes. In this paper its performance for other probes is examined.
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.
ABSTRACT Electronic devices designed for purposes other than transmitting and receiving electromagnetic fields nonetheless act as unintentional antennas. Measurements methods are needed to characterize these antennas for electromagnetic compatibility tests; however, the rigor of precision antenna measurements is typically too costly and time consuming for electromagnetic compatibility applications. Alternate approaches are needed. This paper presents analytical estimates for the directivity of unintentional antennas based on the assumption that unintentional antennas will only randomly excite the available propagating spherical modes at a given frequency. This directivity estimate is then compared to simulated and measured data. Good agreement is shown. Directivity estimates combined with simple total radiated power measurements represent a useful alternative to direct antenna measurements for electromagnetic compatibility tests.
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.
Multiple-Input Multiple-Output (MIMO) systems have been introduced as a solution to improve channel capacity and link reliability. Several MIMO real implementations have been developed and reported in the literature. Some of the implemented MIMO systems are channel sounders, used to perform channel measurements, and do not include any MIMO algorithm. As an improvement of reported testbeds, this paper summarizes a MIMO demonstrator that offers a number of features that make it especially suitable for research and education. Its main advantage is the possibility of testing MIMO algorithms under real channel conditions with different antenna array schemes and antenna configurations in a straightforward way. Other possible applications are channel measurements on different MIMO schemes. The basic configuration is a M x N MIMO (max(M,N)=4) scheme where the antenna module have been designed to allow the user easily change several physical features, such as the spacing and the type of antenna elements, offering the possibility to test the effect of such parameters in the algorithm under study.
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.
An antenna measurement system was developed to complement a new rectangular anechoic chamber (20’L x 10’W x 9’7”H) that has been established at California Polytechnic State University (Cal Poly) through donations and financial support from industry and Cal Poly departments and programs. Software algorithms were written to provide four data acquisition methods: continual sweep and step mode for both single and multiple frequencies. Log magnitude and phase information for an antenna under test is captured over a user-specified angular position range and the antenna's radiation pattern is obtained after post processing. Pattern comparisons against theoretical predictions are performed. Finally an RF link budget is calculated to evaluate the performance of the antenna measurement system.
This paper will discuss the development of a VHF/UHF near field test range for the case where there are reflections from a realistic ground surface. We will show the results of a direct computation algorithm where a far field pattern is computed using plane wave synthesis. The performance of a C++ program that implements this algorithm will be discussed.
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.
ABSTRACT The article discusses the performance and design of a SCARA type robot with counter balanced arms for portable near-field antenna testing. An X-band 43” by 93” antenna on its’ system trailer was tested. A SCARA robot uses rotating joints with parallel axis on linked arms to achieve straight line (or arbitrary) probe movement in a plane. For a horizontal movement plane counterbalanced arms allow movement without change in stress in the scanner structure or foundation, therefore probe movement stays in a plane and structure and foundations can be lightweight and more portable. Probe movement stayed within .004” of a flat surface. Graphite-epoxy tubular arms were used for lightweight, stiffness, and vibration damping. A clockspring like cable carrier was used for each rotary axis. This design kept the center axis free for a directly connected rotary encoder (providing greater accuracy). The diameter of the cable carrier housing at the rotary joint, between arms, enhanced safety by reducing the hazard of a scissoring effect. A dimension touch probe mounted in place of the RF probe was used to align the scanner to the antenna while on its’ system trailer.
Direct far-field transformation is developed from bi-polar near-field samples. As compared to conventional interpolation and expansion methods, a significant reduction in the computation time is obtained by the efficient use of the Fast Fourier Transform and the related shift theorem. Numerical simulations on array of Huyghens sources are considered as validations.
At the Institut Fresnel in Marseille (France), we created an original experimental setup in order to test antennas and carry out scattering measurements in both monostatic and bistatic configurations. The main advantage of this setup is, of course, the multipurpose feature. Two main mechanical systems are installed in a large anechoic chamber. The first system is a spherical positioning setup which allows measurements of antennas and scattered fields for both bi-dimensional (2D) and three-dimensional (3D) targets. This setup consists of two carriages moving on a circular vertical arch and a third carriage which follows a circular path on a horizontal plane. A transmitter and a receiver can be fixed on any of these three carriages. A fourth rotating stage in the center of the spherical setup fixes the angular position of the antenna under test or of the scattering target. The second system is a far-field positioner which allows the measurement antenna patterns and RCS. To illustrate our activities with this original setup, we first show measurements of a metamaterial antenna prototype and then some results of scattered fields obtained on 2D and 3D targets used in studies of electromagnetic direct and inverse problems.
The gain and efficiency of electrically small antennas at long wavelengths can be difficult to measure. On the other hand, it can be shown that both the gain and efficiency of such antennas behave in a predictable asymptotic manner as the antenna becomes electrically smaller. Specifically, in free-space, electric-type antennas have an asymptotic frequency behavior given by f 3/2 and magnetic-type antennas behave as f 7/2. On a Bode plot, the break-point for the gain and efficiency asymptotes occurs when the gain or efficiency is 3 dB down from their electrically small maximum values. This occurs at the frequency where the efficiency is 50% which is where the radiation resistance equals the ohmic loss resistance. It is proposed in this paper that a knowledge of the asymptotic behavior could be of use in extending
Abstract. Scientists at the National Institute of Standards and Technology (NIST) have measured the gain of several antennas using two different methods. The first method is the three-antenna extrapolation method developed at NIST in the early 1970s. The second method is the far-field pattern integration method. We compare gain results and gain uncertainties for several antennas using these two methods.
The measurement uncertainties for base station antenna gain measurement are in general very high, ± 1dB could normally be expected and there are examples of much higher uncertainties. Applying the uncertainties above to the cell planning tools gives at the end a very large uncertainty on the number of cells needed to cover an area. The extra cost for this uncertainty could be an extra 15-20% of the site costs or 10-20% less coverage than expected. This paper identifies the different uncertainty sources and suggests how to optimize the measurement set-up to reduce uncertainties as much as possible during the measurement and compensate for the remaining uncertainties after the measurement.
This paper describes the design and testing of a feed network for a transparent flat plate array antenna. This antenna is the top of a stack of three antennas that must occupy the same volume while pointing in different directions. At many pointing angles, the antenna will create blockage for the antennas underneath. In order to minimize the blockage, the array and its transmission lines must be as transparent as possible to the antennas underneath. The flat plate array consists of active elements over a frequency selective surface (FSS) ground plane that is transparent at the frequencies of the antennas below. The feed lines must also be transparent to the antennas below. This is achieved by minimizing the total area occupied by the feed lines. Rather than the traditional corporate feed network, a series feed network was designed. Such a network requires that each individual feed point must be fed with a coupler where the coupling coefficient is adjusted to distribute the same power to each array element. We will show the details of the design of the network as well as a set of measurements that show the performance.
A low-profile, high efficiency sixteen-element stacked patch microstrip array operating in the L-band frequencies of 1.26GHz and 1.413GHz was designed, fabricated and tested for use in applications to airborne sensors operating on small aircrafts. The array was optimized for element spacing, excitation amplitude taper, low cross-polarization and high beam-efficiency using Particle-Swarm Optimization (PSO) and Finite-Difference Time Domain (FDTD) methods. The design and measurement of sixteen-element array topology, stacked patch elements, and power-divider beam forming network are presented in detail. The study highlights the repeatability measurements and characterization of array with the effect of dielectric radomes in a spherical near-field test facility at UCLA. The results met the requirements of center-frequencies and frequencybands(1.26GHz ± 10MHz, 1.413GHz ± 15MHz), side-lobes, very good beam-efficiency (>90%) and low-cross polarization (<-40dB) in main-beam region of array. The measured results compared well with simulations for the two frequencies. Based on measurement results, the microstrip array design has a potential to be used as a feed for deployable mesh antennas for future spaceborne L-band passive and active sensing systems that can operate at integrated active radar (1.26GHz) and passive radiometer (1.413GHz) frequencies with dual polarization capabilities to study soil-moisture and sea-surface salinity.