AMTA Paper Archive

Several instances in antenna design are known where an anisotropic material is useful ; however, finding a naturally occurring anisotropic material with the required dielectric tensor is often an impossibility. Therefore, artificially engineered anisotropic dielectric materials must be designed, tested, and implemented. In a previous paper by the authors [1], the design and initial measurement of an anisotropic material in Cartesian coordinates was presented along with predictions of how the material could be used to extend the bandwidth of a simple antenna structure.             In this paper we shall present the final implementation of the anisotropic material (with a tensor implemented in cylindrical coordinates) along with data on the material properties, the resulting antenna bandwidth, and radiation pattern. Design considerations for implementation of this approach shall be discussed along with practical limitations. Data shall also be presented on an unexpected result showing that that a reduced volume of anisotropic material produces favorable results. Measured data shall be compared with values predicted using finite difference time domain (FDTD) software and applications of this new broadband antenna for range operations will be discussed. [1]. D. Tonn, S. Safford, M. Lanagan, E. Furman, S. Perini, “DESIGN AND TESTING OF LAYERED ANISOTROPIC DIELECTRIC MATERIALS”, AMTA 2015 Proceedings, Long Beach CA, October 2015.

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.

The Meteosat Third Generation series will comprise four imaging and two sounding satellites.  The MTG-I imaging satellites will carry the Flexible Combined Imager (FCI) and the Lightning Imager.  The MTG-S sounding satellites – a first for Meteosat – will carry an Infrared Sounder (IRS) and an Ultraviolet Visible Near-Infrared spectrometer, which will be provided by ESA as the GMES Sentinel-4 mission. On the MTG-I satellites, FCI will scan the full Earth disc every 10 minutes using 16 spectral channels at very high spatial resolutions, from 2 km to 0.5 km.  In fast imagery mode it will be capable of a repeat cycle of 2.5 minutes over a quarter of the disc.  The MTG-I satellites include a Data Collection System (DCS) & Geostationary Search and Rescue (GEOSAR) payload.  The DCS supports meteorology and weather prediction.  The GEOSAR transponder will be operated within the COSPAS-SARSAT system.  Distress alert signals are received by MTG-I in UHF band and transmitted to ground in L-band for distribution to rescue mission control centers. Developed by Thales Alenia Space Italy, the DCS and GEOSAR UHF and L-band patch array antennas have been designed to operate aboard MTG-I satellites. The Engineering Model of the MTG antenna assembly with mockup has been tested inside ESA’s Hybrid European RF and Antenna Test Zone (HERTZ) chamber. The spherical near field tests performed on the antenna stand-alone and on the antenna mounted on the mockup were aimed at identifying impact of the large satellite structure on radiation pattern of the two medium gain antennas at UHF- and L-band.  Taking into account the frequency of operation and the type of antenna under test, the major contributors to the measurement error are the room scattering and the probe-AUT mutual coupling.  For this reason, dedicated measurements and analysis have been performed, in order to estimate the uncertainty in the most realistic way.  The other parameters have been estimated based on past experience and knowledge on the measurement system.  Several additional measurements were performed in order to produce dedicated uncertainty budgets for the stand-alone and with mockup tests and for the two frequency bands UHF and L-Band.

Among the near-field–far-field (NF–FF) transformations, that adopting the spherical scanning is particularly interesting, since it allows the complete antenna pattern reconstruction and avoids the error due to the scanning zone truncation. The classical spherical NF–FF transformation [1] has been modified in [2] by exploiting the spatial quasi-bandlimitation properties of the electromagnetic (EM) fields [3]. In particular, the choice of the highest spherical wave has been rigorously determined by these properties instead to be fixed by a rule-of-thumb related to the minimum sphere enclosing the antenna under test (AUT). The nonredundant sampling representations of the EM fields [4] have been properly applied to develop effective NF–FF transformations, requiring a number of NF data remarkably lower than that needed by the classical transformation [1] when considering nonvolumetric antennas. In particular, a quasi-planar AUT has been modelled by an oblate ellipsoid [2] or by a double bowl [5], whereas a long AUT has been shaped by a prolate ellipsoid [2] or by a cylinder with two hemispherical caps (rounded cylinder) [5]. Unfortunately, for practical constraints, it is not always possible to mount the AUT in such a way that it is centred on the scanning sphere centre. In such a case, the number of NF data needed by the classical NF–FF transformation [1] and the related measurement time can considerably grow, due to the corresponding increase of the minimum sphere radius. To overcome this drawback, a new spherical NF–FF transformation has been recently proposed in [6], by developing a properly modified version of the spherical wave expansion, wherein the spherical wave functions are defined with respect to the AUT centre instead of the scanning sphere one. Although the number of needed NF data is drastically reduced with respect to that fixed by the rule of the minimum sphere radius, it results to be slightly greater than the one corresponding to a centred mounting. Aim of this work is to properly exploit the nonredundant representations of EM fields to develop a nonredundant spherical NF–FF transformation for long antennas, based on rounded cylinder modelling, which requires the same number of NF data in both cases of centred and offset mounting of the AUT. It will be so possible to remarkably reduce the number of NF data and the related measurement time with respect to that required by the approach [6]. [1] J. Hald, J.E. Hansen, F. Jensen, and F.H. Larsen, Spherical near-field antenna measurements, J.E. Hansen, (ed.), London, Peter Peregrinus, 1998. [2] O.M. Bucci, C. Gennarelli, G. Riccio, and C. Savarese, “Data reduction in the NF–FF transformation technique with spherical scanning,” Jour. Electromagn. Waves Appl., vol. 15, pp. 755-775, June 2001. [3] O.M. Bucci and G. Franceschetti, “On the spatial bandwidth of scattered fields,” IEEE Trans. Antennas Prop., vol. AP-35, pp. 1445-1455, Dec. 1987. [4] O.M. Bucci, C. Gennarelli, and C. Savarese, “Representation of electromagnetic fields over arbitrary surfaces by a finite and nonredundant number of samples,” IEEE Trans. Antennas Prop., vol. 46, pp. 351-359, 1998. [5] F. D’Agostino, F. Ferrara, C. Gennarelli, R. Guerriero, and M. Migliozzi, “Effective antenna modellings for NF–FF transformations with spherical scanning using the minimum number of data,” Int. Jour. Antennas Prop., vol. 2011, ID 936781, 11 pages. [6] L.J. Foged, P.O. Iversen, F. Mioc, and F. Saccardi, “Spherical near field offset measurements using downsampled acquisition and advanced NF/FF transformation algorithm,” Proc. of EUCAP 2016, paper 1570229473, Davos, Apr. 2016.

There are some antennas where rapid validation is required, maintaining a reduced measurement space and sufficient accuracy in the calculation of some antenna parameters as gain. In particular, for cellular base station antennas in production phase the measurement time is a limitation, and a rapid check of the radiation performance becomes very useful. Also, active phased arrays require a high measurement time for characterizing all the possible measurement conditions, and special antenna measurement systems are required for their characterization. This paper presents a single or dual cut near field antenna test procedure for the measurement of the gain of antennas, especially for separable array antennas. The test set-up is based on an azimuth positioner and a near to far field transformation software based on the expansion of the measurements in cylindrical modes. The paper shows results for gain measurements: first near to far field transformation is performed using the cylindrical modes expansion assuming a zero-height cylinder. This allows the use of a FFT in the calculation of the far field pattern including probe correction. In the case of gain, a near to far field transformation factor is calculated for theta = 0 degrees, using the properties of separable arrays. This factor is used in the gain calculation by comparison technique. Depending on the antenna shape one or two main cuts are required for the calculation of the antenna gain: for linear arrays it is enough to use the vertical cut (larger dimension of the antenna), for planar array antenna 2 cuts are necessary, unless the array was squared assuming equal performance in both planes. Also, this method can be extrapolated to other kind of antennas: the paper will check the capabilities and limitations of the proposed method. The paper is structured in this way: section 1 presents the measurement system. Section 2 presents the algorithms for near to far field transformation and gain calculation. Section 3 presents the validation of the algorithm. Section 4 presents the results of the measurement of different antennas (horns, base station arrays, reflectors) to analyze the limitations of the algorithm.  Section 5 includes the conclusions.

Several recent articles [1-9] have focused on assessing spherical near field (SNF) errors induced by using a non-ideal probe, i.e. a probe that has  modal content. This paper explores this issue from the perspective of the probe response constants, defined by [10], which are the mathematical representation of the effect of the antenna under test (AUT) subtending a finite angular portion of the probe pattern at measurement distance . The probe response constants are a function of the probe modal coefficients, the size of the AUT (i.e. the AUT minimum sphere radius ), and the measurement distance , and thus can be used to evaluate the relative contribution of probe  content as both measurement distance and AUT size varies. After a brief introduction, the first section of this paper reviews the theory describing the probe response constants; the second section provides some examples of the probe response constants for a simulated broadband quad-ridge horn, and the final section examines measured AUT pattern errors induced by using the corresponding probe response constants in a conventional SNF-to-FF transform.   References:   [1] A. C. Newell and S. F. Gregson, “Effect of Higher Order Modes in Standard Spherical Near-Field Probe Correction,” in AMTA 2015 Proceedings, Long Beach, CA, 2015.   [2] Y. Weitsch, T. F. Eibert, and L. G. T. van de Coevering, “Investigation of Higher Order Probe Corrected Near-Field Far-Field Transformation Algorithms for Preceise Measurement Results in Small Anechoic Chambers, in AMTA 2015 Proceedings, Long Beach, CA, 2015.   [3] A. C. Newell and S. F. Gregson, “Estimating the Effect of Higher Order Azimuthal Modes in Spherical Near-Field Probe Correction,” in EuCAP 2014 Proceedings, The Hague, 2014.   [4] A. C Newell and S. F. Gregson, “Higher Order Mode Probes in Spherical Near-Field Measurements, in EuCAP 2013 Proceedings, Gothenburg, 2013.   [5] A. C. Newell and S. F. Gregson, “Estimating the Effect of Higher-Order Modes in Spherical Near-Field Probe Correction,” in AMTA 2012 Proceedings, Seattle, WA, 2012.   [6] T. A. Laitinen and S. Pivnenko, “On the Truncation of the Azimuthal Mode Spectrum of High-Order Probes in Probe-Corrected Spherical Near-Field Antenna Measurements,” in AMTA 2011 Proceedings, Denver, CO, 2011.   [7] T. A. Laitinen, S. Pivnenko, and O. Breinbjerg, “Theory and Practice of the FFT/Matrix Inversion Technique for Probe-Corrected Spherical Near-field Antenna Measurements with High-Order Probes,” IEEE Trans. Antennas and Prop., Vol. 58, No. 8, August 2010.   [8] T. A. Laitinen, J. M. Nielsen, S. Pivnenko, and O. Breinbjerg, On the Application Range of General High-Order Probe Correction Technique in Spherical Near-Field Antenna Measurements,” in EuCAP 2007 Proceedings, Edinburgh, 2007.   [9] T. A Laitinen, S. Pivnenko, and O. Breinbjerg, “Odd-Order Probe Correction Technique for Spherical Near-Field Antenna Measurements,” Radio Sci., Vol. 40, No. 5, 2005.   [10] J. E. Hansen ed., Spherical Near-Field Antenna Measurements, London: Peregrinus, 1988.

An innovative method for the diagnosis of large reflector antennas from far field data in radio astronomical application is presented, which is based on the optimization of the number and the location of the far field sampling points required to retrieve the antenna status in terms of feed misalignments. In these applications a continuous monitoring of the Antenna Under Test (AUT), and its subsequent reassessment, is necessary to guarantee the optimal performances of the radiotelescope. The goal of the method is to reduce the measurement time length to minimize the effects of the time variations of both the measurement setup and of the environmental conditions, as well as the issues raised by the complex tracking of the source determined by a prolonged acquisition process. Furthermore, a short measurement process helps to shorten the idle time forced by the maintenance activity. The field radiated by the AUT is described by the aperture field method. The effects of the feed misalignments are modeled in terms of an aberration function, and the relationship between this function and the Far Field Pattern is recast in the linear map by expanding on a proper set of basis functions the perturbation function of the Aperture Field. These basis functions are determined using the Principal Component Analysis. Accordingly, from the Far Field Pattern, assumed measured in amplitude and phase, the unknown parameters defining the antenna status can be retrieved. The number and the position of the samples is then found by a Singular Values Optimization (SVO).

The possibility to use Near Field (NF) representation of antenna measurements in terms of equivalent currents, implemented in the commercial tool INSIGHT, is recently available in most CEM solvers. This method allows to use measured data to enhance numerical simulations in complex and/or large scenarios where antennas are installed. In the past this approach has been investigated and validated by determining the antenna radiation pattern in different antenna placement conditions. The aim of this paper is to present how this method can be extended for simulation of antenna coupling. Indeed using this innovative approach, after antennas are measured, their measured models can be imported in CEM tools and coupling with other radiators in arbitrary configurations can be simulated. No information about mechanical and/or electrical design of the measured antenna model are needed by the CEM tool, since the measured NF model in terms of equivalent currents already fully represents the antenna. Investigations have been performed on a H/V polarized array of three identical elements. Only the radiation pattern of the central element of the array has been measured, then starting from the measured data, the coupling between the other elements has been simulated by numerical tools. Accuracy of the procedure has been checked comparing the simulated results with the measured data of the entire array antenna. The testing procedure combining measurements and simulations consists of the following stages: ·      Measurement of the single element of the array and creation of the measured NF source representation. ·      Importing NF source in the CEM tool and placement in the array configuration. ·      Numerical simulation of the antenna coupling between the measured model and the other two elements of the array. Each element has two feeding ports implementing the dual H/V polarization. Preliminary analysis of the coupling is simulated and comparison with the measured data of the entire array agreement is acceptable. This study is currently under development for improving the accuracy of the results and including new test cases of different complexity.

Free Space VSWR measurement has been the de facto standard method for anechoic chamber performance evaluation for more than 50 years.  In this method, a probe antenna is kept at a fixed angle while traveling along a linear path to record the standing wave pattern.  The probe antenna is then rotated to a different angle to repeat the measurement.  Reflectivity, which is used as the chamber performance metric, is calculated for each probe rotation angle.   In this paper, we show that the reflectivity is affected by the antenna patterns of the probe antenna.  When the probe antenna is aimed at the specular reflection point of a chamber surface, measurement dynamic range is improved, and the method provides a measure of the reflectivity primarily from that surface.  When the probe is not directed at a specular point, other reflections in the chamber can contribute to the VSWR, and the chamber reflectivity becomes more dependent on the probe antenna pattern.

This contribution presents a universal antenna construction toolkit with slotted waveguide elements that can flexibly combined to form a reconfigurable antenna array capable of providing arbitrary symmetric radiation patterns. The design and the arrangement of radiating elements allow adjusting arbitrary real amplitudes of single radiating elements in a solely mechanical way without any electrical feeding network. Additional modular connecting elements even allow two dimensional and conformal antenna designs with circular and multiple polarizations. With a single toolkit in the Ku-band several design and measurement examples are presented, such as a linear array forming a desired main lobe down to -20dB, and a universal two dimensional antenna array that can switch between vertical, horizontal, LHC and RHC polarization. Given a desired antenna pattern the design procedure allows an automated generation of the physical antenna layout that can mechanically be combined without the need of additional full wave simulations. The waveguide toolkit is easily scalable to any other frequency band just being limited in the upper frequency by manufacturing issues. Another major benefit is that the modular concept of connecting and radiating elements eases the manufacturing where otherwise integral waveguide antennas require much more demanding processes. Different physical realizations of the modular waveguide concept are presented and discussed in the paper and related to the antenna performance. Beside several applications for the universal antenna toolkit, such as investigating illumination issues in scattering theory, educational aspects of teaching group antenna theory are also discussed in this contribution.

A high quality antenna feed is an essential element of a compact antenna test range (CATR) in order to ensure the range can achieve the necessary stability in beam width, phase center and the necessary purity of polarization throughout the range’s quiet zone. In order to maintain the requisite quality, such feeds are typically 1) single-port and 2) cover a relatively limited band of frequencies. It is desirable to have a single dual ported, broadband feed that covers multiple waveguide bands to eliminate the need for a polarization positioner and avoid the difficulty associated with changing feeds for a single antenna measurement. Though some such feeds exist in the market, with such feeds, we often see a reduction in polarization purity across the band of interest relative to the more band limited feeds. Previous attempts to utilize dual-port probes and/or extend the bandwidth of the feed have resulted in degraded performance in terms of beam pattern and polarization purity. In an attempt to overcome some of the deficiencies above, the authors have applied polarization processing to dual-pol antennas to correct for the impurity in polarization of the antenna as a function of frequency. We present here a broadband CATR feed solution using a low-cost, dual-port sinuous feed structure combined with polarization processing to achieve low cross-pol coupling throughout the quiet zone. In the following paper, the feed structure, polarization theory, and processing algorithm are described. We also present co- and cross-pol coupling results before and after correcting for the polarization distortion using data collected in two CATRs in Atlanta, GA and Asia.

In order to achieve high accuracy in measuring sidelobes and/or nulls in antenna patterns, it is necessary to use a test system with very high dynamic range. This is particularly important when the antenna has extremely high gain such as those used for certain satellite communications or radio astronomy applications or when transmit power is limited relative to range loss as is often the case in millimeter wave applications. For several years, commercially available antenna measurement receivers have offered a dynamic range as high as 135dB for such applications. This dynamic range has been made possible, in part, by a simple correlator in the receiver’s DSP chain. In this paper, we model the various sources of error in a test signal due to imperfections and uncertainties of the test equipment and the physical environment and analyze these models as they propagate through the receive chain. The results of that analysis demonstrate the correlator’s ability to reduce carrier frequency offset (CFO) and local oscillator (LO) phase noise to offer the fidelity of test signal necessary to achieve extremely high dynamic ranges of up to 135dB.

Indoor RCS measurement facilities are usually dedicated to the characterization of only one azimuth cut and one elevation cut of the full spherical RCS target pattern.  In order to perform more complete characterizations, a spherical experimental layout has been developed at CEA for indoor Near Field monostatic RCS assessment. This experimental layout is composed of a 4 meters radius motorized rotating arch (horizontal axis) holding the measurement antennas while the target is located on a polystyrene mast mounted on a rotating positioning system (vertical axis). The combination of the two rotation capabilities allows full 3D near field monostatic RCS characterization. This paper details a RCS measurement technique and the associated-post processing of raw data dedicated to the localization of the scatterers of a target under test. A specific 3D radar imaging method was developed and applied to the fast 3D spherical near field scans. Compared to classical radar images, the main issue is linked with the variation of polarization induced by the near-field 3D RCS facility. This method is based on a fast and efficient regularized inversion that reconstructs simultaneously HH, VV and HV 3-D scatterer maps. The approach stands on a simple but original extension of the standard multiple scatterer point model, closely related to HR polarimetric characterization. This algorithm is tested on simulated and measured data from a metallic target. Results are analyzed and compared in order to study the 3D radar imaging technique performances.

For near-field antenna measurement it is sometimes desirable or necessary to measure only the magnitude of the near-field - to perform so-called phaseless (or amplitude-only or magnitude-only) near-field antenna measurements [1]. It is desirable when the phase measurements are unreliable due to probe positioning inaccuracy or measurement equipment inaccuracy, and it is necessary when the phase reference of the source is not available or the measurement equipment cannot provide phase. In particular, as the frequency increases near-field phase measurements become increasingly inaccurate or even impossible. However, for the near-field to far-field transformation it is necessary to obtain the missing phase information in some other way than through direct measurement; this process is generally referred to as the phase retrieval. The combined process of first measuring the magnitudes of the field and subsequently retrieving the phase is referred to as a phaseless near-field antenna measurement technique. Phaseless near-field antenna measurements have been the subject of significant research interest for many years and numerous reports are found in the literature. Today, there is still no single generally accepted and valid phaseless measurement technique, but several different techniques have been suggested and tested to different extents. These can be divided into three categories: Category 1 – Four magnitudes techniques, Category 2 – Indirect holography techniques, and Category 3 -Two scans techniques. This paper provides an overview of the different phaseless near-field antenna measurement techniques and their respective advantages and disadvantages for different near-field measurement setups. In particular, it will address new aspects such as probe correction and determination of cross-polarization in phaseless near-field antenna measurements. [1] OM. Bucci et al. “Far-field pattern determination by amplitude only near-field measurements”, Proceedings of the 11’th ESTEC Workshop on Antenna Measurements, Gothenburg, Sweden, June 1988.

A novel system architecture has been developed which makes measurements at N times the analyzer’s frequency, yet requires no communication with the analyzer.  Millitech’s Spartan Test Modules, STMs, splits the input signal from an analyzer, multiplies this by N for the source, and by N-1 for the LO of the receiver mixer.  The mixer downconverts to the original signal, while maintaining its phase integrity, and sends this back to the analyzer.  This scheme is straightforward for narrow bandwidth requirements, but becomes more difficult for wideband ones.  The filtering and temperature compensation requirements are high, but have been solved for these bands resulting in a dynamic range of 70 to 80 dB across 54-69 GHz for V-Band and across 69-90 GHz for E-Band, which directly relates to the side lobe resolution in an antenna pattern measurement.  The wide dynamic range doesn’t come at a cost of slowing the sweep, as in other frequency extension solutions.  This puts the Spartan system performance at the same or higher level as other mixer based systems that have much higher hardware requirements.  STMs can be used to convert any make, model or vintage of vector network, scalar network or spectrum analyzer into a millimeter-wave test station.  The small size of the STMs allows them to be mounted directly onto the back of the antennas.  Therefore, readily available, < 10 GHz cables can be used for the long run back to the analyzer.  The Spartan enables state-of-the-art antenna measurements either directly, in compact ranges, or in near-field ranges, examples will be shown.

We have developed a 60 GHz chip antenna designed for use as a gain and pattern verification tool in the calibration process of a millimeter wave antenna test chamber. The antenna is designed to interface with ground-signal-ground (GSG) micro-probes that have a probe pitch of 150 um to 250 um.  This low temperature cofired ceramic (LTCC) chip antenna is fabricated using DuPont’s 9K7 GreenTapeTM material system with gold conductors.  Features include a wafer-probe transition, a shielded stripline corporate feed network, aperture coupled patch elements, and an integrated Sievenpiper electromagnetic bandgap (EBG) structure for surface wave mode suppression.  The use of the EBG structure enables main beam gain enhancement and side lobe level suppression.  This 2x2 antenna array is directive such that it offers a nominal gain of 12 dBi at broadside over 58-62 GHz with an antenna efficiency of at least 60%.  The entire antenna package has a nominal size of only 10.9 mm x 12.2 mm x 0.71 mm.  Since this antenna package material is hermetic, it has stable performance under varying humidity and temperature which is highly desirable as a reference antenna.

Spherical near-field (SNF) antenna test systems offer unique advantages over other types of measurement configurations and have become increasingly popular over the years as a result. To yield high accuracy far-field radiation patterns, it is critical that the rotators of the SNF scanner are properly aligned. Many techniques using optical instruments, laser trackers, low cost devices or even electrical measurements [1 - 3] have been developed to align these systems. While these alignment procedures have been used in practice with great success, some residual alignment errors always remain. These errors can sometimes be quantified with high accuracy and low uncertainty (known error) or with large uncertainties (unknown error). In both cases, it is important to understand the impact that these SNF alignment errors will have on the far-field pattern calculated using near-field data acquired on an SNF scanner. The sensitivity to various alignment errors has been studied in the past [4 - 6]. These investigations concluded that altering the spherical acquisition sampling grid can drastically change the sensitivity to certain alignment errors. However, these investigations were limited in scope to a single type of measurement system. This paper will expand upon this work by analyzing the effects of spherical alignment errors for a variety of different measurement grids and for different SNF implementations (phi-over-theta, theta-over-phi) [7]. Results will be presented using a combination of physical alignment perturbations and errors induced via simulation in an attempt to better understand the sensitivity to SNF alignment errors for a variety of antenna types and orientations within the measurement sphere. Keywords: Spherical Near-Field, Alignment, Uncertainty, Errors. References [1]     J. Demas, “Low cost and high accuracy alignment methods for cylindrical and spherical near-field measurement systems”,  in the proceedings of the 27th annual Meeting and Symposium, Newport, RI, USA, 2005. [2]     S. W. Zieg, “A precision optical range alignment tecnique”, in the proceedings of the 4th annual AMTA meeting and symposium, 1982. [3]     A.C. Newell and G. Hindman, “The alignment of a spherical near-field rotator using electrical measurements”,  in the proceedings of the 19th annual AMTA meeting and symposium, Boston, MA, USA, 1997. [4]     A.C. Newell and G. Hindman, “Quantifying the effect of position errors in spherical near-field measurements”,  in the proceedings of the 20th annual AMTA meeting and symposium, Montreal, Canada, 1998. [5]     A.C. Newell, G. Hindman and C. Stubenrauch, “The effect of measurement geometry on alignment errors in spherical near-field measurements”,  in the proceedings of the 21st annual AMTA meeting and symposium, Monterey, CA, USA, 1999. [6]     G. Hindman, P. Pelland and G. Masters, “Spherical geometry selection used for error evaluation”,  in the proceedings of the 37th annual AMTA meeting and symposium, Long Beach, CA, USA, 2015. [7]     C. Parini, S. Gregson, J. McCormick and D. Janse van Rensburg, Theory and Practice of Modern Antenna Range Measurements. London, UK: The Institute of Engineering and Technology, 2015

At AMTA 2006, we introduced the world to a system and method for over-the-air (OTA) testing of MIMO wireless devices with the concept of the boundary array technique, whereby the far-field over the air RF propagation environment is emulated to produce the realistic near field multi-path propagation conditions necessary for MIMO communication.  Last year, the CTIA released Version 1.0 of their "Test Plan for 2x2 Downlink MIMO and Transmit Diversity Over-the-Air Performance," which standardizes on the boundary array technique (commonly referred to as the Multi-Probe Anechoic Chamber technique to differentiate it from the use of a reverberation chamber)  for MIMO OTA testing.  As the wireless industry just now prepares to perform certification testing for MIMO OTA performance for existing 4G LTE devices, the rest of the community is looking forward to the development of 5G.  The corresponding future releases of the 3GPP wireless standard are expected to standardize the use of Massive MIMO in existing cellular communication bands.  Massive MIMO is similar to the concept of mulit-user MIMO in IEEE 802.11ac Wi-Fi radios, but is taken to the extreme, with potentially hundreds of antennas and radios per cellular base station.  This high level of radio to antenna integration at the base station will for the first time drive the industry beyond just antenna pattern measurements of base stations and OTA performance testing of handsets to full OTA performance testing of these integrated systems.  At the same time, handset design is evolving to use adaptive antenna systems that will pose additional testing challenges.  Likewise, manufacturers are looking to evaluate real-world usage scenarios that aren't necessarily represented by the test cases used for mobile device certification testing.  This paper will discuss a number of these advances and illustrate ways that the MIMO OTA test systems must evolve to address them.

At AMTA 2006, we introduced the world to a system and method for over-the-air (OTA) testing of MIMO wireless devices with the concept of the boundary array technique, whereby the far-field over the air RF propagation environment is emulated to produce the realistic near field multi-path propagation conditions necessary for MIMO communication.  Last year, the CTIA released Version 1.0 of their "Test Plan for 2x2 Downlink MIMO and Transmit Diversity Over-the-Air Performance," which standardizes on the boundary array technique (commonly referred to as the Multi-Probe Anechoic Chamber technique to differentiate it from the use of a reverberation chamber)  for MIMO OTA testing.  As the wireless industry just now prepares to perform certification testing for MIMO OTA performance for existing 4G LTE devices, the rest of the community is looking forward to the development of 5G.  In the search for ever more communication bandwidth, the wireless industry has set its sights on broad swaths of unused spectrum in the millimeter wave (mmWave) region above 20 GHz.  The first steps into this area have already been standardized as 802.11ad by the members of the WiGig Alliance for short range communication applications in the unlicensed 60 GHz band, with four 2.16 GHz wide channels defined from 58.32-65.88 GHz.  With the potential for phenomenal bandwidths like this, the entire telecommunications industry is looking at the potential of using portions of this spectrum for both cellular backhaul (mmWave links from tower to tower) as well as with the hopes of developing the necessary technology for mobile communication with handsets.  The complexity of these new radio systems and differences in the OTA channel model at these frequencies, not to mention limitations in both the frequency capabilities and resolution requirements involved, imply the need for a considerably different environment simulation and testing scenarios to those used for current OTA testing below 6 GHz.  The traditional antenna pattern measurement techniques used for existing cellular radios are already deemed insufficient for evaluating modern device performance, and will be even less suitable for the adaptive beamforming arrays envisioned for mmWave wireless devices.  Likewise, the array resolution and path loss limitations required for a boundary array system to function at these frequencies make the idea of traditional OTA spatial channel emulation impractical.  However, as we move to technologies that will have the radio so heavily integrated with the antenna system that the two cannot be tested separately, the importance of OTA testing cannot be understated.  This paper will discuss the potential pitfalls we face and introduce some concepts to attempt to address some of the concerns noted here.

In near-field antenna measurements, the pattern effect of the measuring probe represents a systematic error and thus probe pattern correction is a constitutive part of the existing processing algorithms. However, as it was shown in [1], in spherical near-field measurements, for typically used measurement distances, not exceeding two to four diameters of the measured antenna, the probe pattern effect is relatively small, and in many situations the probe pattern can be taken as that of a Hertzian dipole with the resulting effect on the measured antenna pattern being either very small or even negligible. On the other hand, for shorter measurement distances, the probe pattern effect becomes significant and omitting the probe pattern causes noticeable changes in the measured antenna pattern. It was shown in [2] by approximate simulations that in these cases applying the correction using the probe pattern not at the measured frequency, but at the center frequency of a standard waveguide band provides negligible error for even very small measurement distances, not exceeding one or two diameters of the measured antenna, depending on the probe type. Since an approximate model was used for the simulations, the obtained results show only preliminary picture and can only be used as tentative guidelines. In this paper, in order to prove the results of the simulations and the derived conclusions, experimental validation of the simplified probe pattern correction was carried out by processing measured results of several electrically large antennas including probe pattern correction at the measured frequency and at the center frequency of the waveguide band, and comparing the difference. The measured results of a center-fed parabolic reflector, an offset reflector, and a base-station antenna were used for the validation. The obtained results generally confirm the simulations and prove the conclusions that just a single probe pattern can be used for all frequencies over a standard waveguide band for majority of spherical near-field measurement scenarios. [1] S. Pivnenko, J.L. Besada, A. Ruiz, C. Rizzo: On the probe pattern correction in spherical near-field antenna measurements. Proc. 37th AMTA Symposium, Long Beach, CA, USA, October 2015 [2] S. Pivnenko, E. Venero, C. Rizzo: Application of single probe correction file for multi-frequency spherical near-field antenna measurements. Proc. 10th EuCAP, Davos, Switzerland, April 2016