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.)
Absorber treatment for an anechoic range is designed to attenuate the potential reflections from the walls, ceiling, and floor and to keep a certain level below the direct path between the range antenna (or probe) and the quiet zone (or minimum radiated sphere for spherical near-field ranges). There are, however, some antenna measurement systems where the range changes or moves as the data is acquired. In some cases, the probe moves around the antenna-under-test (AUT) along a section of circle supported by an arch or a gantry. In other ranges, the multiple probes are switched on and off; these probes are supported by an arch. Because the direction of the range moves with respect to the walls, ceiling, and floor, it is a bit more complex to arrive to an optimal absorber layout, as well as locating the preferred placements for the instrument rack, door, and vents in the range. In this paper, a higher-order-basis-function method of moments approach is used to model a gantry-supported probe as it moves around the location of the AUT. The power density at the walls as the probe moves is analyzed to arrive to an optimal absorber layout that will provide adequate levels of reflections for measuring an antenna. The paper looks at a gantry that moves from +135° to -135° with the AUT rotating 180° and for a gantry that moves from 0° to +135° with the AUT rotating 360°. The latter will require a smaller range with one of the walls closer to the location of the antenna under test. A series of recommendations based on the electrical size of the absorber at different areas of the range are provided.
The reduction of acquisition time in planar near field systems is a high interest topic when active arrays or multi beam antennas are measured. Different solutions have been provided in the last years: multi-probe measurements systems and the PlanarWide Mesh (PWM) methodology, which implements a non redundant sampling scheme that reduces the number of samples required for the far-field transformation, are two of the most well known techniques. This paper proposes the combination of both approaches to derive a multi-probe PWM grid which reduces the measurement times to the minimum. The method is based on treating the near-field to far-field transformation as an inverse source problem. The multi probe PWM is designed with a global optimization process which finds the best measurement locations of the probe array that guarantee a numerically stable inversion of the problem. A simulated measurement example with the VAST12 antenna is presented where the total number of samples is reduced by a factor of 100 using a 4×4 probe array
Mode filtering has been shown to be very effective in suppressing spurious reflections in antenna measurements. Specifically, it has been well documented that in the quasi-far-field, the two polarizations are decoupled, making it possible to apply standard cylindrical near-field theory on the amplitude and phase data acquired from a single polarization measurement on a great circle cut [1]. The method was further extended to allow data collected from an unequally spaced angular abscissa by formulating the solution as a pseudo-inversion of the Fourier matrix [2]. This formulation, however, can be prone to spectral leakage because of nonorthogonality of the Fourier basis on an irregularly sampled grid, especially when the positions deviate significantly from the regular grid [2]. In this paper, we propose to use Compressed Sensing (CS) to compute the Cylindrical Mode Coefficients (CMCs), which improves the signal to noise ratio, allowing more accurate recovery of the prominent modes. The CS recovery is tenable because with the coordinate translation of the measurement pattern to the rotation center, the Maximum Radial Extent (MRE) of the antenna under test is greatly reduced, making CMCs quite sparse in the mode domain. The novel application of CS presented in this paper further expands the generality of the mode filtering method, which is now applicable to under-sampled data (at below the Nyquist rate) acquired on positions that grossly deviate from the equally-spaced regular grid.
In this paper, the full-wave computational electromagnetic simulation of the production test, measurement, and calibration of a 5G, 24 elements, C-band, active, planar array antenna together with a representative open-ended rectangular waveguide probe with, and without, absorber collar were evaluated using a large computing cluster and a proprietary full-wave solver. In this way, various components within the measurement could be carefully and precisely examined providing a framework for further measurement optimization. Particular attention has been paid to the presence of the standing waves in the simulated near-field measurement. This is a crucial feature of most practical measurements, but is omitted from the vast majority of simulations due to the computational effort required to evaluate it, and which is absent from the standard near-field theory. Here, the presence and impact of this phenomenon has been carefully examined with a range of intensive simulations being harnessed to quantify their impact, as well as enabling various methods for their minimization to be explored in a convenient and highly controlled fashion.
This paper proposes a fast human exposure Absorbed Power Density assessment approach, based on a combination of over-the-air radiative field measurements and fullwave electromagnetic simulations. This so-called augmented OTA technique relies on the computation of an equivalent source or digital twin, which reproduces the radiation properties of the device under test. At short separation distances, the interaction between the human model and the device is however not negligible. A novel solution to model the influence of multiple reflections is introduced, where the inside of the equivalent source box is filled with a perfect electric conductor, thereby creating a reflective digital twin model. Simulation results demonstrate the relevance of this approach for enabling accurate absorbed power density evaluations.
Extrapolation method is regarded as one of the most accurate methods for obtaining the absolute far-field gain of an antenna. This paper will compare the efficacy of several data processing techniques for calibrating low frequency antennas with long ring down time. Traditionally, measurement data are preprocessed to remove ripples from multipath reflections before a curving fitting is applied. We will first investigate two traditional data processing techniques. The first technique is to apply time domain gating to the vector response vs. frequency data at each separation distance. Then the gated data as a function of distance is fitted to the polynomial equation. The second technique is spectrum domain filtering. The vector response as a function of distance is transformed to k domain at each frequency. A band pass filter is applied in k domain to keep only the direct antenna response. In this study, we propose a new approach - the magnitudes of the antenna response as a function of distance including the ripples is fitted to a more complete generalized antenna response equation with the antenna-to-antenna multiple reflection terms included. This paper will compare the three techniques using a set of measurement data on double-ridged waveguide horn antennas in a fully anechoic extrapolation range.
The success and efficiency of many classical iterative plane-to-plane based phase retrieval algorithms is to a large extent dependent upon the fidelity of the initializing, i.e. guiding, phase estimation [1], [2]. This is especially so when using these techniques to recover the phase of active electronically scanned array antennas such as those employed within beam-steering mm-wave Massive MIMO antenna systems intended for 5G New Radio applications where the performance of the algorithm, and its ability to not become trapped within one of the (possibly many) local minima, is particularly dependent upon the quality of the initializing guess where access to a phase reference is not always convenient, or even possible. Many traditional phase recovery iterative Fourier methods employ simulation or passive measurement supported phase initialization [1], however this information is not always readily available, or in the measurement may require a destructive, invasive, examination of the device under test (DUT). In this work we address this issue by presenting a proof of concept which employs a machine learning based neural network [3] to estimate the initializing phase function based on the assessment of the measured amplitude only near-field pattern. Here, we show that there is sufficient information contained within the difference between the two near-field amplitude only scans to be able to determine the antenna beam steering characteristics. A simplified beam steering case with electronic scanning in one, or more, scanning axes is demonstrated and verifies the power of the novel method, as well as illustrating its inherent resilience to noise within the amplitude only measurements, and verification of the robustness of the approach thereby extending the range of measurement applications for which this class of iterative Fourier algorithms may be successfully deployed [4].
This paper proposes a range-Doppler imaging method based on FFT-MUSIC method for FMCW radar systems. With the growing significance of vehicle and human motion recognition in automotive radar, the accuracy of conventional deep learning network-based recognition methods is reduced because it depends only on distance, speed, and angle information provided by conventional radars. Therefore, various types of imaging radar methods have recently been proposed. Among them, the range- Doppler imaging algorithm is widely used. This algorithm can simultaneously analyze both distance and velocity characteristics of a vehicle or person. However, conventional range-Doppler imaging based on the FFT algorithm has limited resolution, which cannot obtain detailed information on the target. Although the FFT algorithm is widely used in many applications, its lowresolution characteristics can limit its ability to provide detailed information. In particular, improving velocity resolution often requires the extraction of a significant amount of data. To address this issue, a range-Doppler imaging method based on FFT-MUSIC is proposed in this paper. This technique has been simulated using Remcom’s WaveFarer® software package. The proposed algorithm is effectively able to distinguish between two moving vehicles in several cases in which the ranges and velocities are too close to be resolved by conventional FFT methods. We can observe that the proposed algorithm enhances the velocity resolution by approximately twice as much as the conventional algorithm. Additionally, in indoor environments, the proposed algorithm provides a detailed representation of the indoor multipath, outperforming conventional algorithms. The high-resolution radar imaging offered by the proposed method will enable improved target recognition and thus enhance overall performance in practical applications.
This paper proposes a range-Doppler imaging method based on FFT-MUSIC method for FMCW radar systems. With the growing significance of vehicle and human motion recognition in automotive radar, the accuracy of conventional deep learning network-based recognition methods is reduced because it depends only on distance, speed, and angle information provided by conventional radars. Therefore, various types of imaging radar methods have recently been proposed. Among them, the range- Doppler imaging algorithm is widely used. This algorithm can simultaneously analyze both distance and velocity characteristics of a vehicle or person. However, conventional range-Doppler imaging based on the FFT algorithm has limited resolution, which cannot obtain detailed information on the target. Although the FFT algorithm is widely used in many applications, its lowresolution characteristics can limit its ability to provide detailed information. In particular, improving velocity resolution often requires the extraction of a significant amount of data. To address this issue, a range-Doppler imaging method based on FFT-MUSIC is proposed in this paper. This technique has been simulated using Remcom’s WaveFarer® software package. The proposed algorithm is effectively able to distinguish between two moving vehicles in several cases in which the ranges and velocities are too close to be resolved by conventional FFT methods. We can observe that the proposed algorithm enhances the velocity resolution by approximately twice as much as the conventional algorithm. Additionally, in indoor environments, the proposed algorithm provides a detailed representation of the indoor multipath, outperforming conventional algorithms. The high-resolution radar imaging offered by the proposed method will enable improved target recognition and thus enhance overall performance in practical applications.
Absorber fences have been used on compact ranges since their first implementations. The purpose of this fence is to hide the feed positioner and reduce the direct coupling between the feed and the device under test (DUT). A known problem caused by such a fence is that it diffracts the plane wave generated by the reflector, creating an interfering ripple on the illumination of the DUT in the quiet zone. Traditionally, fences have serrated edges to direct this diffracted signal away from the quiet zone. However, this redirection is not always achievable or even repeatable from one facility to the next. Often low frequency requirements drive absorber physical size, leading to very large absorbing surfaces that cannot be optimized to reduce this interfering signal. In this paper, the fence design presented in a recent publication [1] is further optimized by modifying its shape and absorbing material parameters. The performance of this new design is compared with traditional fences.
The spherical wave expansion-based transmission formula allows to accurately evaluate the coupling (or S21 parameter) between a transmitting and a receiving antenna. Its use as tool for probe corrected spherical near-field to far-field transformation is well accepted and documented. On the other hand, its direct use in the evaluation of antenna measurement performance has been exploited only in recent years. In this paper we will show how measurement performances predicted with the transmission formula compare with actual measurements. Taking as examples relatively complex antenna measurement systems like spherical near-field, plane wave generators and CATR, we will focus on the prediction of the accuracy of the measured radiation patterns, also including the approximation of reflections from the test environments, and on the evaluation of link budgets.
This communication provides an effective two-steps strategy to compensate for known 3-D probe positioning errors occurring in the non-redundant (NR) cylindrical near-to-far-field (NTFF) transformations. As first step, a phase correction, here denoted as cylindrical wave correction, is employed to perform the correction of the positioning errors relevant to the deviations of the measured NF samples from the nominal scanning cylinder. Then, an iterative procedure will be applied to retrieve the NF samples at the points specified by the adopted sampling representation from those obtained at the previous step and affected by 2-D positioning errors. Finally, after properly reconstructing the correctly distributed cylindrical samples, the data necessary to apply the classical cylindrical NTFF transformation can be restored in accurate way by employing a 2-D optimal sampling interpolation (OSI) formula. It should be noticed as, to derive the NR sampling representation, as well as the OSI scheme, it is necessary to provide a proper modeling of the antenna under test. This modeling has been got by shaping the source with a prolate spheroid. Numerical tests will show the capability of the procedure to compensate these 3-D positioning errors.
This work aims to propose and optimise a non-redundant spherical spiral near-to-far field (NTFF) transformation for elongated AUTs from spiral near-field (NF) data acquired over the upper hemisphere due to the presence of an infinite perfectly electric conducting (PEC) ground plane. Such a technique properly exploits the principle of image and the theoretical foundations of spiral scan for non-volumetric AUTs to develop the non-redundant representation along the sampling spiral in presence of PEC ground plane and to synthesise the voltage NF data which would be acquired over the spiral wrapping the lower hemisphere. Once these voltage NF data have been synthesised, then an efficient 2-D optimal sampling interpolation scheme allows the recovering of the NF data required by the classical NTFF transformation. In the hypothesis that the AUT and its image exhibit a predominant dimension as compared to the other two ones, a prolate spheroidal source modeling is here adopted. Numerical tests show the accuracy of the developed non-redundant spherical spiral NTFF transformation.
—Cylindrical hubs with fan blades are inserted into a pipe inside a modified camera box—a recently introduced structure intended to host differently-shaped ducts behind an aperture. The resulting structures increase the reproducibility of commonly used simplified jet-engine inlet models and are designed to serve as precisely-defined radar cross section (RCS) benchmarks with reliable reference results. The design, manufacturing, and assembly of the measured structures are detailed; the RCS measurement setup, data collection, and post processing are documented; and the uncertainty in measured RCS data is quantified with the help of simulations. Results show that the fields scattered by the structures, while highly sensitive to geometric and material perturbations, can be both measured and simulated accurately even at frequencies with many propagating modes inside the pipe.
The application of multi-probe (MP) technology in near-field (NF) measurement scenarios is well-known for its ability to significantly reduce test time. This is achieved by electronically sampling the radiated field using different probes in the array, eliminating the need for mechanical probe movement. However, in planar near-field (PNF) measurements, the accuracy is contingent on probe correction (PC) during post-processing. Characterizing the pattern of each individual sensor in a PNF MP system presents an additional challenge, often being impractical or impossible. Previous publications have explored various approaches to address this challenge and achieve an accurate characterization of the MP equivalent pattern. In this paper, we focus on the average probe pattern (APP) technique, which involves the experimental determination of the MP pattern. To validate the effectiveness of the APP technique, we conducted experiments on a large PNF MP system equipped with a 4.65m probe array. Our measurements focused on an electrically large 1.5m diameter reflector antenna (MVG SR150 reflector, fed by a quad-ridge horn) operating in the 1.8–6.0 GHz frequency range. The validation process involved the comparison of MP measurements processed with the APP technique and conventional open-ended waveguide (OEW) PNF measurements. To ensure the reliability of the validation, we conducted the comparative tests within the same frequency range and test setup. This minimized the impact of measurement errors, enabling a robust and accurate comparison between the techniques. By validating the APP technique's effectiveness, we aim to establish its suitability for improving accuracy in PNF MP system measurements.
There are several applications that require the Fourier transformation from data obtained on a regular polar grid to a regular sine-space grid, and one of these is the processing of plane-polar near-field data. The most common approach to this task is to interpolate from the polar grid to an X-Y grid and then use the conventional 2D Fourier transform. This paper revisits an alternative algorithm, referred to herein as the polar-coordinate Fourier transform (PCFT), for doing the same transformation from polar input to a regular sine-space output grid. This PCFT has some advantages when processing data undersampled in their angular phi spacing, and appears to offer the possibility of probe correction without having to counter-rotate the probe. The process of the PCFT is similar to that of the conventional 2D Fourier transform. These two processes are compared. Rather than interpolating the polar data to the X-Y grid, the PCFT starts with a 1D transform along each diametric spoke of the polar wheel. Each spectral output is then rotated by the corresponding phi angle and interpolated to a common sine-space output grid. If probe-pattern correction is needed for a probe with no extra roll axis, then a notional approach is also described.
In this contribution a simple algebraic approach is discussed on the minimum of required sample points for either a nearfield or a farfield configuration to calculate the antenna’s current distributions to accurately reconstruct the antenna’s radiation pattern anywhere in space. The proposed algebraic approach comprises a Gaussian quadrature sampling scheme for a set of Hertzian dipoles with unkown amplitudes representing the antenna current distribution. The algebraic equation system with the number of unknown amplitudes then suggests the minimum of required sample points in the radiated field. In this initial study simulation examples of a dipole antenna and a horn antenna are presented validating the proposed algorithm.
In this work, a tunable frequency-selective surface is designed with a center frequency that can be varied with an applied DC voltage. An equivalent circuit representation of the FSS is derived from a finite-difference time-domain (FDTD) simulation of the passive FSS, allowing tuning circuits for the FSS to be designed in common circuit simulation tools such as SPICE. A comparison between the spectral response obtained from FDTD and equivalent circuit modeling (ECM) in SPICE shows that under certain conditions, ECM provides good agreement with full-wave analysis, is less computationally intensive, and provides physical insight. The ECM technique enables rapid design and analysis of various trade-offs, such as those between resonant frequency tunability and bandwidth. The ECM-designed circuit is then validated with full-wave analysis of the designed structure with active components using an FDTDSPICE hybrid co-simulator. Finally, applicability of the chosen active FSS topology as a metasurface for free-space dielectric material characterization is discussed.
The IEEE Std P2816 recommended practice for computational electromagnetics applied for the modeling and simulation of antennas is currently being developed by the IEEE Antennas and Propagation Standards Committee (APS/SC), sponsored by the IEEE Antennas and Propagation Society (APS). The document provides guidance on the numerical modeling of antennas deployed in free space using commonly adopted computational electromagnetics (CEM) techniques such as the finite element method (FEM), the finite difference time domain (FDTD) method, the Method of Moments (MoM), the finite integral technique (FIT) and the transmission line matrix (TLM) method. Benchmark models and comparisons of numerical simulation results are included for potential users of the standard to better understand the uncertainties and limitations of these techniques. A biconical antenna was previously proposed as a benchmark model. The numerical simulation results showed a good overall agreement among the participating laboratories and against the analytical solution. Herein, a 5G New Radio (NR) FR1 ultrawide band (UWB) antenna is proposed as another benchmark model for the development of IEEE Std P2816. In addition to the comparison of the numerical simulation results obtained from the participating laboratories, the simulation results are confronted with preliminary measurement results.
Vehicular wireless power transfer (WPT) systems conforming to the SAE J2954 standard are thought to operate as inductive WPT systems. As such, they should be able to be accurately represented by a magnetic multipole source. For example, the magnetic field of the “circular” coupler, which is described in the SAE standard, can be represented by a combination of a vertical, linear magnetic quadrupole and a horizontal magnetic dipole. However, it has been recently shown that a significant conservative electric field exists in such a WPT system due to the multi-turn windings. This can lead to a significant electric multipole contribution, predominantly a vertical linear electric quadrupole for the circular coupler. In fact, the circular coupler electric field (not the magnetic field) is somewhat similar to that of a coaxial aperture. Here, we carry out a detailed analysis of the electric multipole representation of the “DD” coupler which is also described in the SAE standard. The analysis of the electric multipole of the DD coupler is more complex than that of the circular coupler. Because the DD coupler is composed of two side-by-side spiral windings, it is possible to obtain two different electric multipoles from configurations that produce nominally the same magnetic multipole and the same magnetic performance. Fortuitously, the configuration used in the DD coupler very nearly cancels the conservative electric field, the associated electric multipole, and the attendant emissions.