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.)
In this paper, a method is presented that allows for the experimental verification of far-field conditions in a direct far-field measurement setup. The method is based on a relativedistance sweep (i.e., increasing the distance by linearly translating one antenna) and on the Friis equation. The presented method is only valid for one specified direction and is therefore well suited to assess whether or not far-field conditions are achieved when performing an absolute measurement, such as a maximum gain or effective-isotropic-radiated-power (EIRP) measurement. It is shown that antenna measurement uncertainties due to the finite antenna separation, scattering, positional inaccuracies, drift and noise on the order of hundredths of dBs around 30 GHz for a separation on the order of 1 m can be obtained. Using this method, it is also experimentally shown that whether or not farfield conditions are met depends not on one but on both antennas in a two-antenna measurement setup. This implies that, strictly speaking, the far-field distance cannot be determined by solely considering the largest antenna in a two-antenna measurement setup.
An active S-band dual-polarized multifunction phased array radar (MPAR), the Advanced Technology Demonstrator (ATD), has recently been developed for weather sensing and aircraft surveillance. The ATD is an active electronically scanned array (AESA) with 4864 transmit/receive (T/R) modules and was installed in a spherical radome. Simulations and a novel phased array measurement technique have been explored to assess the impact of high reflectivity from a wet radome during rain that can potentially induce voltages exceeding the transmit amplifier breakdown voltage. The measurement technique uses array elements radiating one at a time to illuminate the radome, and uses superposition to quantify the received signal power in a reference antenna on the face of the array. It is shown that when the radome surface is wet and highly reflective, certain electronic steering angles sum to a large reflected signal focused on the array face. This measurement technique can be used prior to high-power phased array radar operation to monitor the magnitude of reflections and help avoid element transmit amplifier failures.
The equivalence principle allows for the replacement of a radiating structure with a Huygens’s box resulting from a near-field scan obtained from either simulations or measurements. The near-field on the surface enclosing the structure is replaced by a set of equivalent electric and magnetic dipoles that can be used as sources for numerical simulations. In this manner, problems of high complexity can be handled, using less unknowns, by separately addressing the radiating structure and the surrounding diffracting objects. Furthermore, sometimes the antenna model is not available, and it must be considered as a black box. However, important challenges arise when the Huygens’s box approaches a surrounding structure. Indeed, the near-field resulting from discretized equivalent dipoles presents a singularity at each dipole’s position. Therefore, the obtained fields with the equivalent dipoles are not valid in the close vicinity of the Huygens’s box. A novel method is proposed to handle the aforementioned singularity. It is based on the interpolation of the electromagnetic field between the non-singular region and the surface of the box. The field on the non-singular region is obtained using equivalence principle whereas the near field scan data is used to retrieve the field on the box surface.
With the event of integrated and multi-standard wireless links, phaseless antenna measurements are attracting more and more interest in research. Especially in the context of connected and automated driving, antennas, frontends, and digital signal processing units merge into telematic units and require new methods for performance evaluation in the installed state. The measurement of the phase diagram and the exact absolute positioning of electrically large antennas, i.e., antennas interacting with the car body, present challenges for safety-relevant applications and reliable test methods. This paper describes a way to determine the position of automotive antennas in the installed state with sub-wavelength precision from phaseless measurements. Realistic LTE uplink signals were used as test signals as they would be transmitted by an active device in a real-world scenario. The localization algorithm is based on orthogonal power measurements of the transmitted signal on a cylinder surface and a non-linear optimization. By comparison with a conventional localization based on spherical far-field data, an accuracy of the approach of less than 1 cm was achieved, which is less than λ/16 at the considered frequency of 1870 MHz.
Diagnostic techniques are crucial in antenna development and testing to enhance the Device Under Test (DUT) performances and identify the cause of possible failures in the qualification process. Among different approaches [1]-[8], it has been demonstrated that the equivalent currents method (EQC) [8]-[9], implemented in [10], is one of the most efficient for investigations in various application areas [11]-[13]. Indeed, the generality of the 3D reconstruction surface enclosing the DUT is a key feature, it ensures that this technique is unique and highly suitable for diagnostics, respect to traditional methods based on plane wave expansion. To handle electrically large problems, the EQC method has been initially based on a Fast Multipole Method (FMM) [14]. The recent advent of 5G technologies has led to an increasing need in terms of antenna electrical dimensions. Therefore, a novel technique based on a Nested Skeletonization Scheme (NSS) has been implemented to guarantee a further reduction of memory requirements and computational time. The new capability has been demonstrated in the past for a patch array antenna [15]. In this paper, the diagnostic capabilities of the EQC approach are applied to an early prototype of an electrically large array antenna for 5G antenna measurements applications [16].
An antenna measurement system at the Institute of High Frequency Technology at RWTH Aachen University is being established containing a six-axis robot arm allowing the realization of numerous measurement geometries. Room scattering is one of the most crucial uncertainty terms in every antenna measurement which becomes even more interesting in a dynamic scattering situation. To determine the scatteringinduced interference caused by the robot, a motion-capable model is developed and firstly simulated using the multi-level fast multipole method between 8GHz and 12GHz to qualitatively assess the surface currents. Secondly, asymptotic simulations are carried out using physical optics for the most important robot positions at 60GHz which is in the frequency range where the system is operated. For example, differences in the same simulation points of up to 20dB are shown for different robot positions. Based on the simulation results, the measurement sequences can be optimized by selecting a trajectory which reduces the scattering effects. In addition, the strongest scattering sources of the robot are identified in order to cover these parts by absorbers. Therefore, the knowledge gained from the simulations can be applied to the measurement system to improve the performance of antenna measurements.
The validity and viability of using frequency domain mode filtering to qualify an EMC chamber above 1 GHz has been demonstrated in a previous study [1]. The novel approach overcomes the difficulties with under sampling encountered in the traditional spatial sampling method adopted by CISPR 16-1-4, and it also has the distinct advantage over the time domain method adopted by ANSI C63.25.1 in that broadband and low ring-down antennas are not required. In this study, we further examine one of the assumptions made in the previous study to translate the quasi-far-field pattern to the rotation center. The approximate method is compared to a more rigorous method by using a quasi-far-field to far-field transformation first before applying the phase translation and subsequent mode filtering. In this paper, we further validate the method by conducting an intercomparison study based on measurements conducted in a 3 m anechoic chamber to show the correlations of the mode filtering method to the CISPR and the time domain (TD) site VSWR methods. We demonstrate how the proposed method improves the test repeatability, lowers measurement uncertainties, and increases measurement efficiencies.
In Multi-Probe (MP) based measurement systems, the standard procedure is to calibrate the probe array with a well-known reference antenna [1]. This procedure equalizes amplitude, phase, and polarization characteristics of each probe array element. In Planar Near Field (PNF) systems, the probe pattern impact is usually more pronounced than in other near field scan geometries, such as spherical. Thus, the probe pattern must be compensated during post-processing for more accurate measurements at wider angles. While the probe array calibration ensures the on-axis equalization, the probe array elements still have individual pattern difference due to finite manufacturing accuracy and absorber interaction. Probe compensation using an equivalent probe pattern of the array has been shown to be very effective and accurate for MP PNF systems [2]. In this paper we compare two methods to determine the equivalent probe pattern for a given MP PNF system. We also discuss the acceptable limits of pattern variation within the array versus measurement accuracy as a design parameter for MP PNF systems.
Accurate characterization of locomotive antennas is key to safe and robust railway signaling and control communication. With the introduction of new technologies and the foreseeable migration from the GSM-R standard towards FRMCS, new wireless applications and specifications arise, and suitable antenna solutions need to be developed and tested. Moreover, the rooftops of modern locomotives present a dense and harsh environment; therefore, potential antenna mounting spaces should be carefully evaluated to avoid undesirable degradations of the antenna radiation patterns. Due to the electrically large and complex structure of locomotives, full-scale testing is challenging to perform, especially under laboratory conditions. Antenna measurements with geometrically scaled models present a powerful alternative to address this issue. In this paper, we present and discuss antenna measurement results of a scaled locomotive mockup. The mockup incorporates two different cabin geometries, one with a step-like rooftop contour, and one with a smooth slightly tilted geometry. First, the optimum scaling factor was identified and validated through numerical simulations. Afterwards, antenna measurements with a scaled locomotive mockup were carried out in our automotive antenna measurement facility VISTA. The measured results were compared with the numerical simulations, where a good correlation above 80% was found. Secondly, the impact of the rooftop geometries, and superstructures on the roof has been investigated for a range of operational frequencies between 700 and 2600 MHz. The results reveal that the parasitic impact of the antenna environment becomes more pronounced at higher frequencies.
Over-the-Air (OTA) measurements of modern mmWave User Equipment (UE) should be performed under plane-wave conditions which require sufficiently large measurement distances. Alternatively, shorter distances can be considered but special plane-wave generator devices should be used instead of conventional probes/range antennas. The use of probes/range antennas at reduced distances would offer advantages in terms of cost effectiveness and improved dynamic range but in general, they would not provide the proper plane-wave condition. To set the proper measurement distance the electrical size of the whole UE is usually considered. However, in most cases only a smaller portion of the UE actively contributes to the radiation. Reduced distances can thus be considered without significant loss of accuracy, unless the source of radiation is offset from the center of the measurement system. This latter scenario is called parallax and often causes distortions of the pattern if the distance is not sufficiently large. In this paper parallax compensations techniques applicable to amplitude-only measurements will be investigated considering realistic OTA measurement emulations of modern devices equipped with mmWave phased arrays placed in different positions. The investigation is performed focusing both on the measurement of the single beams and on the overall spherical coverage provided by the antennas.
This paper presents a wide incident angle metasurface unit cell element applied to a reconfigurable intelligent surface (RIS) for beamforming and beam-steering applications in the 26 GHz frequency band from the fifth generation of mobile communications (5G) frequency range 2 (FR2). Each metasurface unit cell is based on a printed frequency selective surface (FSS) loaded with a varactor diode. The FSS-based structure is based on a circular loop at the top and a slot-based ground plane at the bottom resulting in a 0.25x0.25λ0 total area. The complete unit cell element encompasses four conducting layers, in which the first two ones form the FSS. RF chokes are printed at the middle layer to isolate the DC circuit, and the bias lines are routed at the fourth layer. The unit cell has been conceived using the full-wave electromagnetic solver ANSYS HFSS. Its numerical results demonstrate a reflection phase shift up to 180º and reflection magnitude higher than 0.4 at the 26 GHz frequency band for incident wave angle from 0 to 50º. The proposed reconfigurable intelligent surface might be applied to future wireless communication systems, planar antenna reflectors, and vortex beam generation.
This paper experimentally investigates the ping pong channels resulting from a narrow but divergent beam of the horn antenna as witnessed in non-line-of-sight (NLoS) scenario for the 240−300 GHz frequency range. A THz vector network analyzer (THz−VNA) extender measurement setup equipped with a ∼ 25 dBi horn antenna as transceiver (TRX) is employed to retrieve and evaluate the reflections prompting a ping pong influence on THz wireless channels. This ping pong effect being uncommon at lower frequencies is studied from extensive measurement campaigns for monostatic measurement setups in the frequency range of interest. Corner reflector (CR) as well as metal plate reflectors (MRs) are employed to analyze the resultant ping pong channels in the manifold scenario setups. This ping pong effect may lead to an irreversible ambiguity in the channel transfer functions (CTFs) and certainly demands understanding of such sub-harmonics.
Producing uncertainty estimates is an integral part of every measurement procedure. This is a time consuming process in spherical near-field antenna measurements, because for a few factors in the uncertainty list it is necessary to perform additional full-sphere measurements. In this paper we propose an original method to simulate in a computer the effect of two important items in the uncertainty list, namely θ-zero and axes intersection errors, by taking advantage of the fact that after a measurement the antenna under test is completely characterized. These sources of errors are associated with the rotating positioner of the anechoic chamber and, therefore, are more prone to change between campaigns. Consequently, they need to be checked and assessed for every antenna under test, which can be inefficient if the measurement time is excessive due to the test antenna’s size. With the techniques presented in this work, the time spent estimating the impact of these errors in the measurands is greatly reduced, since additional full-sphere measurements are not needed. Furthermore, the errors can be isolated from each other and the degree of linearity between measurand and error source can be assessed. Finally, it is no longer necessary to occupy the antenna under test in order to perform the uncertainty estimation.
This paper presents antenna design and packaging consideration for near field communication (NFC) system that is being used in automotive security systems, and, more specifically, to an NFC reader for obtaining access to, and controlling activation of, a transportation vehicle such as a motor vehicle. Various important studies for automotive applications were performed in this work such as a magnetic wall method. This magnetic wall method can prevent the reduction in NFC reading range caused by proximity to body sheet metal. It provides a unique and superior magnetic shielding effect as compared to ferrite sheets because it is not temperature dependent and can be implemented with minimal cost and complexity. The proposed design can be easily fabricated on the back face of the NFC reader antenna PCB using conventional PCB techniques.
In this work, a simulation-based study is presented exploring the effect of surface roughness of 3D printed plastics on the accuracy of material parameter extraction. A homogenous sample material is placed inside the W-band waveguide and the S-Parameters are simulated. Two different methods for estimating the dielectric properties of the sample using the simulated S-Parameters are presented (i) NRW (Nicolson-Ross-Weir) technique-based estimation method, and (ii) Feko optimization-based estimation method. An error analysis study is presented to understand the percentage of error due to the surface roughness of the 3D printed plastics. For N7 grade surface roughness, NRW predicts 14% error in material parameters due to surface roughness, whereas Feko optimization method predicts 10% error compared to estimation without any surface roughness. Process outlined in the paper can be used to estimate effect of surface roughness of waveguides on material property measurements at mm wavebands such as W-band.
We tackle the problem of recovering 2-D complex fields starting from the spectral amplitude data, the support of the source, and a few additional information. In particular, we further elaborate on our ‘crosswords-solution like’ approach where the solution is found by solving 1D problems and congruence arguments. Its argued that intersecating lines and circles (rather than just lines) is more effective, and we show how the resulting approach, initially developed for the case of continuos (aperture) sources, is also effective in determining the excitations of discrete (array) sources.
Spherical wave function expansions involving a TM- R and TE-R field decomposition are widely used. Here we examine a spherical wavefunction expansion based on a TM- z and TE-z field decomposition. The impetus is the straightforward relationship between the TM-z and TE-z spherical wavefunctions and the cartesian multipoles which, in turn, have contemporary application in wireless power transfer. However, there are potentially other useful features of such a TM-z and TE- z spherical wavefunction expansion. For example, the TE/TM-z decomposition results in the ????-polarized far electric field being associated exclusively with the TM-z wavefunctions while the ????-polarized far electric field is associated exclusively with the TE-z spherical wavefunctions. Unfortunately, from the analysis given here, it appears that it is not possible to represent a generalized finite source in TE/TM-z spherical wavefunctions exclusively. One succinct way to show this is by attempting to expand the far field of an x or y directed electric or magnetic dipole in TE/TM- z spherical wavefunctions. The dipole can be represented by a single TE/TM-R spherical wavefunction, but the expansion in TE/TM-z wavefunctions is problematic.
Cellular LTE MIMO downlink performance, for 4x4, 4x2, and 2x2 LTE MIMO architectures, in terms of average data throughput and availability, were investigated in an urban canyon environment of Frankfurt, Germany at 2110 MHz on a Sport Utility Vehicle (SUV) with metal and glass roofs for a virtual route. This study utilized the following measured antenna radiation patterns for total polarization on the SUV at 2110 MHz for the mobile station: 1) roof-mounted antenna on metal roof; 2) roof-mounted antenna on glass roof; 3) interior-mounted planarinverted F antenna; and 4) interior-mounted planar-inverted F antenna rotated 90 degrees. This research was carried out using a three-dimensional simulation software suite that enabled users to simulate electromagnetic wave propagation and wireless network planning. The following observations were obtained from this research. First, the MIMO architectures for the SUV with metal roof exhibited approximately 5% higher average data throughput levels compared to the same MIMO architectures on the SUV with glass roof. Second, the throughput availability for the 4x4 and 4x2 MIMO systems were comparable. Lastly, the average throughput for the 4x4 MIMO system was higher than the 4x2 and 2x2 MIMO systems for the SUV regardless of roof material.
In this paper, we show how Prolate Spheroidal Wave Functions (PSWFs) can be reliably and accurately numerically computed to represent the field radiated by an oblong antenna. We show the convenience of the PSWFs representation by comparing its performance against a spherical harmonics expansion. We also point out how the PSWFs representation can be fruitfully exploited to face a Near-Field/Far-Field (NFFF) problem in a cylindrical geometry for the mentioned class of antennas.
Resistive materials are often employed in antenna or absorber design for radio frequency (RF) applications. Causal material models are needed when modeling wideband RF systems using time-domain numerical models (e.g. FDTD). To this end, the frequency-dependence, from 10’s of MHz to 10’s of GHz, of spatially patterned and un-patterned resistive-cards (R-cards) were measured using free space and specialty materials measurements fixtures. Specifically, the complex sheet-impedance of two R-card specimens were measured at VHF frequencies using either an 8.5-inch slotted rectangular-coax (R-coax) or a recently developed resistive material mapping probe (RMMP). At GHz frequencies measurements were conducted using a standard 2’ focused beam lens system. The multi-band complex-impedance data were fit using a set of causal sheet material models. Typically, the fit errors are in the 1-3% range for causal models of measured data over two-plus decades of bandwidth.