Welcome to the AMTA paper archive. Select a category, publication date or search by author.
(Note: Papers will always be listed by categories. To see ALL of the papers meeting your search criteria select the "AMTA Paper Archive" category after performing your search.)
In this paper we present a recently-developed adaptive method for decomposing radar signals into a wide range of physically meaningful mechanisms. The signal decomposition we will pursue is based on a set of functions referred to as “atoms” in the signal processing literature. These atoms can be derived from disparate mechanisms such as scattering center responses given by the Geometric Theory of Diffraction (GTD), which are localized in range, and resonance phenomena, which are localized in frequency. The atoms populate a “dictionary” of waveforms which are used to decompose radar signals. Traditional Fourier methods have restricted the class of atoms solely to harmonic exponentials. In this paper, we consider a number of signal decomposition methods. We chose a method based on the Basis Pursuit procedure of Chen and Donoho [1],which uses an L1 norm as opposed to a least squares approach to search the dictionary. We chose this technique because of its sparsity of representation and convergence properties. We will show results of this technique applied to numerically simulated data and show that disparate scattering mechanisms can be isolated and identified in the radar data.
For the past seven years, ERIM has been studying RCS measurement error sources and processing methods by which these errors can be reduced. Image editing is an extension of range-gating where scattering measurements are improved by removing undesired scattering phenomena in the range-crossrange image domain. Conventional image editing methods rely on a user-supplied polygon to segment an image into desired and undesired scattering regions. However, the polygon method suffers from variability due to user and display characteristics, provides little hope for automation, and cannot be easily extended beyond two dimensions. An alterative approach based on peak region segmentation minimizes or eliminates these limitations and adds an element of optimally that can also improve the performance of image editing techniques. In this paper, we will discuss the application of peak region segmentation to the image editing problem and show examples that demonstrate some of the advantages of this approach.
ERIM has developed techniques, based on parametric spectral estimators, for removing additive target support contamination from narrowband RCS measurements [1]. These techniques allow target and support returns to be extracted from frequency sweep data with much greater accuracy and resolution than that afforded by conventional Fourier techniques. These algorithms have recently been enhanced to incorporate scattering mechanism frequency dependence in the underlying signal model. Specifically, damped exponential and power-of-frequency sweep data with much greater accuracy and resolution than that afforded by conventional Fourier techniques. These algorithms have recently been enhanced to incorporate scattering mechanism frequency dependence in the underlying signal model. Specifically, damped exponential and power-of-frequency signal models have been used. The modification substantially improves algorithm performance in measurement situations where there is small absolute bandwidth, but relatively large fractional bandwidth, which can lead to appreciable variation in scattering mechanism amplitude. The paper will demonstrate the technique’s ability to remove target support contamination using numerical simulations and compact range measurements of canonical targets mounted on pylon supports. It will be shown that the algorithm can remove the additive pylon contamination even for situations where the pylon return dominates the target return and cannot be resolved from the target in conventional Fourier range profiles.
A compact range antenna measurement error model is presented which shows that the ripple in the quiet zone can only be caused by stray radiation from the edges of the reflector, presuming a perfectly shaped (serrated) reflector. This is proven by defining an equivalent system which gives significant intuitive insight in the behavior of a compact range. For a simple example this model is shown to be consistent with PO. The model intuitively explans many antenna measurement accuracy observations made in a compact range without the need for extensive knowledge of antenna or diffraction theory. These observations include the relation between quiet zone ripple characteristics and antenna measurement accuracy, especially for boresight, narrow angle and wide angle measurements. It also explains why new correction techniques like AAPC work so well in spite of their presumable simplified modeling.
Compact Payload Test Ranges (CPTR) for test zones of 5 meters or larger can be used for both payload and advanced antenna testing. In both cases accurate calibration, including amplitude and phase characteristics across the test zone, is required. Accurate data analysis is needed in order to establish corresponding error budgets. In addition, boresight determination will be required in both measurement types for most applications. Since it may be difficult or even impossible to scan the test zone field using a (planar) scanner, application of a large reference target (a rectangular or circular flat plate) can be seen as in interesting alternative. RCS measurements are then performed and test-zone field characteristics are determined in both amplitude and phase. Time- and spectral domain techniques can provide valuable information as to the location of possible disturbances. The evaluations is complemented with the measurement of a VAlidation STandard (VAST) antenna in combinations with an advanced APC technique. These techniques have been demonstrated at the CPTR at ESTEC, Noordwijk, the Netherlands. Results and practical considerations are presented in this paper.
Nowadays, the standard facility for accurate satellite antenna testing is the Compensated Compact Range (CCR). In order to increase measurement accuracy several techniques can be applied, which are based on antenna pattern comparison. The theory of these techniques together with experimental results have been described in several papers in the past [1][2][3]. This paper presents how pattern comparison techniques are applied for space programs and is another step to official qualification of the Advanced Antenna Pattern Comparison (AAPC) method at Dornier Satellitensysteme (DSS).
For narrow beam antennas or track antennas some parameters like main beam or null position and 3 dB beamwidth need to be determined with an accuracy of less than a mill or mrad. With Near Field measurements, the Far Field is normally calculated by FFT-processing. This does, however, not provide the required accuracy. Nevertheless, the measured Near Field data contains information about any Far Field point. An iterative approach is presented to determine the Far Field antenna characteristics with high accuracy.
A new measurement control and data preprocessing system has been implemented at the TUD-ESA Spherical Near Field Antenna Test Facility. This facility is located at the Technical University of Denmark (TD) and operated in cooperation with European Space Agency (ESA). The measurement control system as well as a flexible information system is described. The data collected during measurements are passed through a preprocessor before the data are stored on disc. By taking advantage of the band-limited nature of the measured near field the preprocessor is able to detect RF leakages and to correct for the non-ideal sampling that is caused by non-zero integration time of the receiver.
A new strategy reducing the effect of the environmental noise in the evaluation of the radiated far field by means of a near-field far-field transformation technique is presented. A plane-polar scanning system is considered although the approach holds for general scanning geometries. Numerical and experimental results confirm the effectiveness of proposed the technique.
The techniques that are presently available for the measurement of complex permittivities of dielectric substrates are only applicable to single layer substrates. This paper presents a new technique which uses the Finite Element Method (FEM) to estimate the complex permittivities of individual layers from the measurement of the S-parameters of a rectangular waveguide holding a multilayer dielectric substrate sample. In this method, a network analyzer is used to measure reflection and transmission coefficients f a rectangular waveguide loaded with a layered sample. Using FEM, the reflection and transmission coefficients are determined as a function of the complex permittivities of the multilayer substrate. Measured and calculated values of the reflection and transmission coefficients are then matched using the Newton-Raphson Method to estimate the complex permittivities of the layers of the sample.
In this paper we present an improved theoretical modelling for the free space technique for measuring the complex permittivity of materials at microwave frequencies. The theory was developed for a transmission set-up with two identical pyramidal horn antennas. By performing a spectral decomposition of the aperture fields, the new model takes the effect of the non plane wave character into account when the sample is not placed in the far field of the transmitting antenna. With the use of the new theoretical model it becomes possible to place the sample much closer to the antennas without infringing the theoretical assumptions since no plane wave incidence is needed. In this way the transversal dimensions of the sample can be reduced significantly. The validity of the new theoretical model was verified by measurements on many dielectric (Plexiglas, polystyrene,…) and lossy materials. A comparison was made with the values obtained when the usual plane wave theory is used.
Multipactor breakdown is a possible payload failure for communication satellites. The multipactor phenomenon significantly increases noise levels interfering with signals relayed by a satellite; it can even damage or destroy RF components and transmission lines. In order to retire the risk of single point failure in a recent spacecraft C-band antenna suite, certain critical components were tested to determine the threshold of multipaction. The components tested were: waveguide isolators, a waveguide polarization switch, and a waveguide PIM filter. The need to test the components at higher power levels than had been attempted on previous multipaction tests required the development of a new test system. The system utilized ring resonant multiplication to develop pulsed power levels of +77 dBm. The multipaction global detection methods utilized a spectrum analyzer montoring the notched noise level, and peak power meters monitoring the arrier pulse shape for distortion. This paper briefly describes the test system developed.
In this paper we discuss proper RF system design for performing insertion-loss measurements using a microwave receiver and harmonic mixers. Specifically we will deal with problems caused by changing reflection coefficients of the devices which feed the mixer. When broadband mixers and coaxial isolators are used problems may be caused by the changing load seen by the local oscillator. This is due to local oscillator leakage through the mixer and isolator. We will elaborate on this problem, noting its impact on the measurement and suggest a procedure to properly minimize its effect.
To accurately measure static or dynamic Radar Cross Section (RCS), one must use precise measurement equipment and test procedures. Recently, several DoD RCS ranges, including the Advanced Compact RCS Measurement Range at Wright-Patterson AFB, established procedures to estimate measurement error. Working cooperatively with the National Institute of Standards and Technology (NIST), Wright Laboratory established a baseline error budget methodology in 1994. As insight was gained from the error budget process, we noted that many common RCS measurement calibration techniques are subject to a wide variety of potential error sources. This paper examines two common so-polarized calibration devices (sphere and squat cylinder), and discussed techniques for evaluating calibration induced errors. A rigorous “double calibration” methodology is offered to track calibration measurement error. These techniques should offer range owners fairly simple methods to monitor the quality of their primary calibration standards at all times.
The calibration of nonreciprocal radars has been studied extensively. A brief review of known calibration techniques points to the desirability of a simplified calibration procedure. Fourier analysis of scattering data from a rotating dihedral allows rejection of noise and background contributions. Here we derive a simple set of nonlinear equations in terms of the Fourier coefficients of the data that can be solved analytically without approximations or simplifying assumptions. We find that independent scattering data from an additional target such as a sphere is needed to accomplish this. We also derive mathematical conditions that allow us to check calibration data integrity and the correctness of the mathematical model of the scattering matrix of the target.
Scientific-Atlanta has developed a new algorithm for obtaining high accuracy cross-polarization measurements from prime focus, single reflector, compact ranges. The algorithm reduced cross-polarization extraneous signals to levels that rival or exceed much more expensive dual reflector systems, but with the associated cost and simplicity of a single reflector system. This paper provides an overview of the new algorithm. It explains the limitations on conventional polarization measurements in single reflector systems and the methods for overcoming these limitations without error correction for some antennas. A method for determining if error correction is needed for a particular antenna is reviewed and the fundamentals of the error correction algorithm are explained. Preliminary test results are provided.
This paper presents several enhancement tools that were developed to improve the Georgia Tech Spherical Far-Field/ Near-Field Antenna Measurement Range. Measurement amplitude and phase drift was quantified by sampling an antenna measurement signal over long time intervals while leaving the AUT rotation positioners fixed. A return-to-point drift correction tool was implemented to correct for the long-term drift component for spherical surface measurements. Temperature sensitive components of the receiver were moved from an area with severe temperature variations to a temperature stable area to reduce the phase variation. A software tool was developed to display a histogram of the variation in repeated spherical scan measurements. Histogram vales show that drift correction improves the repeatability of an antenna pattern measurement. The shapes of the histograms have been helpful in identifying random and deterministic variations.
A method to generate time and direction of arrival (TADOA) spectra of the quiet zone fields of a RCS/ antenna range is presented. The TADOA spectra is useful for locating the stray signal sources in the RCS/ antenna range. To generate the TADOA spectra, quiet zone fields along a linear scan over the desired frequency band are probed. The probed data are calibrated to remove the magnitude and non-linear phase variation versus frequency. A calibration technique is also proposed in the paper. The TADOA spectra for simulated probed data as well as experimental probed data are shown.
In the antenna measurement field, large investments are typically required for installations; however, these installations soon become obsolete due to advances in technology. In order to recover as much of the original investment as possible, upgrading the original installation becomes an attractive alternative. Here, we detail an ongoing upgrade of a planar near-field system. The original single-beam, single frequency system was installed approximately twenty years ago. By replacing the network analyzer, and installing a faster computer controller and accompanying software, the system was upgraded to a state-of-the-art measurement system. The upgraded system is capable of high-speed beam and frequency switching on the fly. Using the existing scanner, motor control hardware, and laser positioning sensors, the system was delivered at a significantly reduced cost.
The use of global positioning satellite (GPS) signals for automotive navigation and this on-vehicle GPS antennas has become more common recently. As the number of users increases the cost of the highly integrated receiver is predicted to come down to less than $50. It is possible to measure the antenna patterns of GPS antennas as installed on vehicles, but it is important to make sure that the parameters measured are valid for the GPS environment. In this case, sky coverage and polarization are more important than the directive pattern, for example. This paper shows a method of comparing a number of antennas by using the actual GPS satellite signals as test signals.