AMTA Paper Archive

Mars rover Direct-to-Earth (DTE) communication is an exciting new development that can maintain transfer of high volumes of scientific data from Mars to Earth. Currently, large orbiting assets are used as a relay to return scientific data, often containing higher data rates than current DTE systems. Therefore, the goal of this paper is to investigate antenna array topologies to augment DTE systems to support high data rates. The antenna design is complex, having to simultaneously support dual-band, high gain, high power handling, and circular polarization capabilities. An exhaustive study of patch elements in literature shows that current geometries are infeasible for a Mars rover DTE system. A CP Half E-shaped patch element is developed, containing important dual band S11/AR performance in the required RX and TX bands while featuring a single-feed single-layer design. Moreover, various subarray architectures are evaluated to determine if the gain requirements can be achieved. To meet this gain requirement, a 4x4 subarray topology is designed which allows a modular, scalable, and high gain design. To feed each of the 4x4 element subarrays, a stripline feed network is developed, consisting of a binomial impedance transformer and a four stage 1:2 power divider. This feed network supported a broadside radiation pattern for the subarray topology. These components are then integrated, first through a full wave simulation in HFSS. This rigorous study showed support for Mars rover DTE communications systems. The integrated subarray design is then fabricated and measured using a spherical near-field chamber in the UCLA Center for High Frequency Electronics (CHFE) facilities where measurements showed a very good comparison to the simulation results. Overall this integrated subarray design was successful, showing dual-band, high gain, high power handling, and CP performance.

Wireless connectivity is rapidly expanding in both popularity and potential. Incorporating antenna arrays on both ends of the wireless channel realizes this potential by facilitating beam steering and increased directivity. From an operations vantage point, these capabilities reduce transmit power, increase data rates, and extend communication range. Antenna arrays also facilitate forming nulls toward antagonistic regions to hide information and thwart easily accessible jamming devices. These performance characteristics of antenna arrays address several critically important challenges for Unmanned Aerial Vehicle (UAV) operation, which is becoming attractive in both military and commercial sectors. Maintaining wireless communication channels over extended ranges that can potentially cross into antagonistic regions helps accomplish precise, adaptable mission objectives. In addition, efficiently utilizing power, a scarce commodity typically drawn from solar panels, facilitates extended flight durations. Finally, the reduced transmit power also reduces the aircraft weight, which can further extend flight duration. Although antenna arrays offer extensive advantages, the final design must account for the presence of the aerial platform including other electronic systems. Strong mutual coupling can then result from operating multiple wireless systems within a physically confined space. In addition, the surrounding environment can change the electrical characteristics of the antenna (e.g. input impedance and radiation pattern). Analyzing these electrical characteristics on a physically-confined platform becomes an electrically-large problem when operating a communication channel over the 2.4 GHz ISM band. Simulating the installed performance potentially requires significant computational resources unless research is conducted to understand the trade-offs between numerical methods in existing commercial software. The installed performance of an antenna array on an electrically-large platform can then be optimized in the most efficient manner. In this paper, we demonstrate a process that efficiently navigates the typical trade-offs engineers encounter when conducting an antenna placement study. This process involves designing a conformal antenna array in an isolated environment, analyzing potential installation locations and further optimizing the antenna array for the chosen location.

Antennas utilized as probes, sources, and for gain comparison are typically specified to have excellent cross polarization levels, often on the order of 50 dB below the primary polarization component. In many cases, these antennas are fed with a waveguide-to-coaxial adapter, which can be sourced from a multitude of vendors. Depending on the design and construction of the adapter, and the distance from the excitation probe to antenna aperture, the adapter itself can contribute significantly to the degradation of the polarization purity of the antenna. These adapters typically use one of several methods to achieve a good impedance match across their bandwidths, including tuning screws, posts and stubs. These tuning elements may be arranged asymmetrically and can cause the waveguide to be overmoded locally. Additionally, there is wide variance in the separation of the adapter excitation probe and waveguide electrical flanges, which may not be long enough to suppress the higher order modal content. In this paper, we study the effects of adapter to antenna aperture coupling, including the coupling of fields local to the current probe as well as those that are induced by design asymmetries. The results of the analysis lead to a number of rules of thumb which can be used to ensure that the antenna polarization purity is optimized.

Flared notch (“Vivaldi”) arrays have been a subject of great interest since the mid 1990s for use in broadband phased-array systems. These arrays are popular in large part due to their ultra-wide bandwidths, which can span multiple octaves, exceeding the bandwidths of the individual flared notch elements themselves. This effect is achieved via strong inter-element coupling, a departure from the conventional wisdom of minimizing mutual coupling between elements in a phased array. The benefits of this design choice have been widely reported on in the literature - however, this dependence on element coupling also places serious constraints on array performance, especially with regards to scan angle, active impedance, and array efficiency, which often go unreported. In addition, reliance on inter-element coupling necessitates an array that can be safely approximated as “infinitely” planar. If an array does not strictly meet this condition, significant VSWR issues can result, especially for elements near the edges of the array. This paper discusses the common pitfalls inherent in practical flared-notch array design that are often overlooked in the literature. To aid in this analysis, a network-centric approach to array modeling is demonstrated that allows for an examination of both element- and array-level performance metrics in a way that minimizes computation time and resources. Special attention is paid to parameters such as active impedance as a function of scan angle, which, though vital to array performance, are often mischaracterized by “infinite array” approximations commonly used by engineers in the design phase. The effects of mutual coupling on different array performance metrics, both beneficial and detrimental, are examined in detail so that an informed decision can be made on the suitability of the flared-notch topology for a given application.

Microstrip patch antennas are well known in the field of communications and other areas where antennas are used. They consist of a metallic conducting surface deposited onto a grounded dielectric substrate and are widely used in situations where a conformal antenna is desired. They are also popular antennas for array applications. But most patch antennas are typically resonant structures owing to the standing wave of current that forms on them. This resonant behavior limits the impedance bandwidth of the antenna to a few percent. In this paper we shall present an approach for improving the bandwidth of a resonant patch antenna which employs an engineered anisotropic superstrate. By proper design of this superstrate and its tensor, and proper alignment of it with the axis of the patch, an antenna with improved impedance bandwidth results. Some of the challenges associated with the measurement of the anisotropic superstrate will be discussed, ranging from 3D simulations to physical models tested in the laboratory. A final working model of the antenna will be discussed; this model consists of a stacked patch arrangement and was designed to operate at the GPS L1 and L2 frequencies. Data collected from 3D simulations using CST Microwave Studio along with laboratory and anechoic chamber measurements will be presented, showing how the bandwidth at both of these frequencies can be increased while maintaining circular polarization in both passbands. Tolerance to errors in alignment and fabrication will also be presented. Additionally, some lessons learned on anechoic chamber measurements of the antenna’s gain and axial ratio will be discussed.

Gain is a principal property of antennas; it is essential in establishing the link budget for communication and sensing systems through its presence in Friis’ transmission formula and the radar range equation. The experimental determination of antenna gain is most often based on a gain-transfer technique involving a reference antenna for which the gain has been calibrated to high accuracy; this is typically a pyramidal horn antenna [1]. The required accuracy of antenna gain obviously depend on the application; in some cases it can very high, ±0.1 dB or less, and this implies an even higher accuracy, of the order of ±0.01dB, for the gain reference antenna. This work investigates the accuracy to which a gain reference antenna can be calibrated; the investigation is based on experimental spherical near-field antenna measurements [2] and computational integral equation / method of moments simulations [3]. While calibration of gain reference antennas has been studied in many previous works, even works from early 1950s [4]-[6], this work is novel in systematically supporting measurements with full-wave simulations. Such simulations facilitate the study of e.g. the effect of multiple reflections between antennas at short distances. We study two absolute calibration techniques for the gain of pyramidal horn antennas. The first technique determines gain as the product of directivity and radiation efficiency; this technique has been referred to as the pattern integration technique [7] (which is not an entirely adequate designation since gain cannot be determined from the radiation pattern). The second technique determines the gain from Friis’ transmission formula [8] for two identical antennas; this technique is generally referred to as the two-antenna technique [1]. These two calibration techniques involve very different steps and contain very different sources of error; for both techniques our investigation involves measurements as well as simulations. For the pattern integration technique we compare experimental and computational results for the directivity and demonstrate agreement within one-hundredth of a dB. The radiation efficiency is calculated by different techniques based on the surface impedance boundary condition for the metallic walls of the pyramidal horn. This technique is not influenced by proximity effects or by impedance mismatch between the measurement system and the gain reference antenna. For the two-antenna techniques we compare experimental and computational results for the gain and we compare the calculated distance-dependence with that of the extrapolation technique [9]. It is demonstrated how the use of the phase center distance in Friis’ transmission formula notably decreases the necessary separation between the antennas for a required accuracy, but that multiple reflections may then become a limiting factor. This technique is highly influenced by the impedance mismatch that must be accurately accounted for. We compare the gain values resulting from the pattern integration technique and the two-antenna technique, including their very different uncertainty estimates, for a C-band standard gain horn. The work is related to an on-going ESA project at the DTU-ESA Spherical Near-Field Antenna Test Facility for the on-ground calibration of the scatterometer antennas of the EUMETSAT MetOp Second Generation B-series satellites. IEEE Standard – Test Procedures for Antennas, Std 149-1979, IEEE & John Wiley & Sons, 1979. J.E. Hansen, “Spherical Near-Field Antenna Measurements”, Peter Perigrinus Ltd., London 1987. www.wipl-d.com W.C. Jakes, “Gain of Electromagnetic Horns”, Proceedings of the IRE, pp. 160-162, February 1951. E.H. Braun, “Gain of Electromagnetic Horns”, Proceedings of the IRE, pp. 109-115, January 1953. W.T. Slayton, “Design and Calibration of Microwave Antenna Gain Standards”, Naval Research Laboratory, Washington D.C., November 1954. A. Ludwig, J. Hardy, and R. Norman, “Gain Calibration of a Horn Antenna Using Pattern Integration”, Technical Report 32-1572, Jet Propulsion Laboratory, California Institute of Technology, October 1972. H.T. Friis, “A Note on a Simple Transmission Formula”, Proceedings of the I.R.E. and Waves and Electrons, pp. 254-256, May 1946. A.C. Newell, R.C. Baird, P.F. Wacker, “Accurate Measurement of Antenna Gain and Polarization at Reduced Distances by an Extrapolation Technique”, IEEE Transactions on Antenna and Propagation, vol. 21, no. 4, pp. 418-431, July 1973.

This paper discusses design and fabrication of a low cost, combined pressure / thermal test-bench engineered for environmental tests of UAV mounted antennas. Both test-beds are mainly made of commercial of-the-shelf (COTS) parts and in-house made frames. They occupy small space and do not require specific professional skills for operation or high maintenance cost. Measurement setup is designed to reliably reproduce temperature and pressure corresponding to altitudes from sea level to 6000 m (20000 ft) with dynamic load equivalent for 200 m/s (400 knots) of air speed. Experimental results of radome enclosed wideband antenna are presented.

Measurement scenarios for 5G mobile communications are nowadays challenging the industry to define suitable turn-key solutions that allow Over the Air (OTA) testing of non-connectorized devices. In order to respond to the needs of an effective measurement solution, that allow measuring all the required OTA parameters at both sub6GHz and mm-Wave frequencies and that could be deployed in a very short time, the Compact Antenna Test Range (CATR) was chosen. In this paper, we will summarize the performance and the testing capabilities of a short focal-length, corner-fed CATR design, providing a 1.5 m x 1.5 m cylindrical Quiet Zone, operating from 1.7 GHz to 40 GHz and upgradeable to 110 GHz, allowing OTA measurements of Active Antenna System (AAS) Base Stations (BS), installed at Ericsson premises in Gothenburg, Sweden in 2017.

Non-destructive testing (NDT) is a fundamental step in the production chain of aircraft structural components since it can save both money and time in product evaluation and troubleshooting. This paper presents a reflection-based imaging technique for electromagnetic (EM) testing of composite panels, with the device under test (DUT) being metal backed and both the transmitting and receiving components of the NDT system situated on the same side of the DUT. One of the key properties of the presented technique is the complete redundancy of a reference measurement, thereby making it feasible to retrieve a high quality image of the DUT with only a single measurement. Data for both a proof-of-concept DUT and an industrially manufactured composite panel is provided, and the retrieved images show the applicability of both the measurement technique and the imaging algorithms.

The unknown-thru calibration technique is being used to achieve a system level calibration at millimeter wave frequencies (>50 GHz) on the robotic ranges at NIST. This two-port calibration requires the use of a full bi-directional measurement, instead of a traditional single-direction antenna measurement. We explored the value of the additional data acquired. We find that we can use this information to verify antenna/scan alignment, image the scattering from the positioner/facility, and perform a first order correction to the transmission data for uncertainties due to LO cable flexure.

Finding an off the shelf antenna tuned for the operating frequency of a small satellite mission can be difficult, especially when the mission uses an experimental license in a frequency band that is not used for commercial or amateur radio systems. This paper discusses how electromagnetic modeling software can be used to assist adapting commercial-off-the-shelf (COTS) antennas to other operating frequencies than the ones for which they have been originally designed. The discussion is illustrated with a case study outlining how a COTS cross-polarized UHF Yagi amateur radio antenna is adapted for operation in the 400 MHz experimental bands.

The Conex Antenna, Radar, and Measurement Equipment Lab (CARAMEL) is a ten-element VHF antenna array that operates from 30 MHz-120 MHz with an attached lab space. This array was developed for use in low frequency Radar Cross Section (RCS) measurements. The antenna elements support both vertical and horizontal polarizations. The antenna was designed using a genetic algorithm, employing the fragmented aperture technique; measured and modeled data will be presented. The attached lab space is air conditioned and provisioned for rack mounted equipment. The structure uses a modified 20' Conex shipping container where an entire sidewall has been replaced with a reinforced composite radome for the antennas. The overall mechanical frame design included a Finite Element Analysis to ensure structural integrity. The system is intended for long-term standalone use as an outdoor measurement radar system but can be moved using standard shipping container methods. The structure was shipped using a standard cargo carrier from Atlanta, Georgia to White Sands, New Mexico.

Antenna gain is the product of directivity and antenna loss. Antenna gain is typically measured by comparing the antenna under test (AUT) to a standard gain horn (SGH) or direct gain measurement using a calibrated probe. This requires an accurate account of power into the AUT and SGH, the loss of all test cables and switches must be measured to obtain an accurate AUT gain. Additionally, SGH calibration uncertainty reduces the quality of the measurement. The gain measurement technique describe here exploits the near-field range capability of accurately producing the pattern of high gain antennas. The near-field range allows the full wave capture of antenna aperture fields and transformation to the far-field with high resolution. The new technique uses the directivity obtained by integrating the far-field pattern, accounts for the spill-over energy not measured by the near-field range, and uses measured network losses of the AUT. It does not require measured losses of test cables and switches. Since AUT losses are typically measured as part of antenna integration the technique reduces overall measurement burden. Accurate calculation of spill-over energy is the key to success. The technique has been shown to yield better accuracy than the typical gain calibration method for multi-beam high gain antennas.

Testing of automotive antennas are commonly performed in large Spherical Near Field (SNF) ranges [1-3] able to host the entire vehicle to test the effect of the antenna coupling with the structure [3]. The impact of a realistic ground, such as asphalts or soil, on the radiation performance of the vehicle mounted antennas is often a desired information. As long as the free-space response of the vehicle is available, such information can be obtained with fairly good accuracy considering post-processing techniques based on the Image Theory (IT). Automotive systems with absorber material on the floor [3] are thus ideal for estimating such effects because the free-space signature of the vehicle is directly measured and because the radiation pattern is usually available on more than just a hemisphere. In this paper an IT-based technique which allows for the estimation of a realistic ground is proposed and validated with simulations where the measurement setup of a typical multi-probe free-space automotive system is emulated. The impact of the truncation of the scanning area is analyzed in detail showing how advanced post-processing techniques [4-6] can be involved to mitigate the truncation errors and thus obtain a better estimation of the realistic ground effect.

This paper discusses some aspects of isolation improvement and associated measurements on a cylindrical millimeter-wave repeater operating over K, Ka and V bands. The isolation between the transmitting and receiving antennas is improved by means of reactive impedance surface implemented as tapered depth corrugations. The designed tapered depth profile broadens bandwidth of the surface compared to the traditional quarter wavelength corrugations. Required isolation of 80 dB and large electrical size of the platform make numerical analysis and actual measurements challenging. Details of the analysis and measurements are summarized. Along with external coupling, the coupling due to leakages from waveguide components and antennas is also discussed. Measurements confirm that the design goal isolation is accomplished.

A new Compact Antenna Test Range (CATR) has been built, as a turnkey facility, with a cubic quiet zone (QZ) of 4.8m x 4.8m x 4.8m in the frequency range 0.9-18 GHz. The CATR has been installed in a new building with an isolated and stable foundation. The dimensions of a traditional CATR for such QZ size becomes impractical and requires a very large chamber. A new, diagonally fed, short focal length reflector has been developed to minimize the chamber size to fit the dimensions of 22 m x 14.5 m x 14.5 m.

In this paper, we present a chip antenna in the 27GHz band, targeting 5G measurements. This antenna can be used as reference in mm-wave measurement systems, such as the MVG µ-Lab, feeding the antenna under test through a micro-probe station. The reference antenna is employed to calibrate in gain through the substitution method. The antenna shown in this paper is an array of four patches, fed through a strip-line beam forming network. A transition strip-line to coplanar waveguide allows the antenna be fed by the micro-probe.

Compliance with regulatory exposure requirements of power density for 5G systems will need accurate measurements. In this work a near field measurement technique for electromagnetic exposure of 5G communication devices is presented. The technique requires two measurements, one of a device under test and one of a small aperture as a calibration measurement. The method uses method of moments and involves reconstructing equivalent currents on a predefined surface. These currents are then used to generate and propagate the electromagnetic fields to an arbitrary plane and further compute the power density. The measurement data are obtained through a planar scan of a device under test using a probe and probe calibration using a small aperture to obtain an accurate field with absolute positioning. Measurement data is presented and compared with simulations for several distances and two antennas, operating at 28 GHz and 60 GHz. The computed power density agrees well with simulations.

The Low frequency Coaxial Reflectometer is the recommended procedure to measure the absorbers' reflectivity as per the IEEE 1128-1998 standard. The standard recommends the operable frequency range up to 500 MHz with a permissible error of 2 dB and higher error beyond 600 MHz. This paper studies and discusses the error on different types of absorber. Each of the absorber type is simulated in the square section of the reflectometer setup to compute the absorber's reflectivity using Ansys HFSS. An effective time gating technique is applied to reduce the effect of edge effects. These results are compared to the unit cell simulation results with a plane wave excitation and periodic boundary conditions. The absorbers are then simulated in the complete reflectometer setup to include the mismatch associated with the transition and compared to the unit cell model results. The errors associated with the comparison of the absorbers' simulation results for these different models are analyzed. The combination of these different absorbers is simulated in unit cell model. The absorbers are placed in different regions and orientations inside the reflectometer. The comparison between the unit cell results of the combination of the absorbers and the results of the absorbers inside the reflectometer in different orientations give the effect of the non-uniform field distribution inside the reflectometer.

The principles of near-field antenna measurements in Cartesian, cylindrical, and spherical coordinates are well established and documented in the literature and in standards used on antenna ranges throughout government, industry, and academia. However the measurement methods used and the mathematics that are applied to compute the gain and radiation of the pattern of the test antenna from the near-field data assume that the antenna is operating in free space. This leaves several questions open when dealing with antennas operating over a lossy ground plane, such as the ocean. In this paper, we shall discuss a possible avenue for addressing this problem : the use of Complex Image Theory (CIT). The CIT approach allows the lossy earth to be removed and an image of each equivalent source point in the space above it to be constructed in the now empty space below it, but where the depth of that image is in general a complex number. While it might appear confusing to define a complex depth, such a depth is merely a mathematical construct that accounts for a magnitude and phase shift that occurs due to the presence of the lossy ground. The depth is computed so that the boundary condition at the surface of the original lossy ground is maintained; in this way, an equivalent problem is formulated. We propose an approach based on CIT that can be applied to the problem of a spherical nearfield antenna measurement taken over seawater. A limiting case of measurements taken over a metal ground plane shall be presented, along with thoughts about some practical concerns involved in the performance of such measurements.