AMTA Paper Archive

A technology for target identification has been developed that is directly applicable to the analysis of the backscattering behavior of targets. For the latter purpose the target is placed on a turntable, and amplitude/phase data are collected over the aspect angle sector of interest, using a radar with sufficient bandwidth to resolve the target in range. For ground vehicles and small aircraft a range resolution of about 1 ft is sufficient. Standard processing is used to form an ISAR image over the appropriate aspect angle sector. The difference relative to the more conventional procedures is that the complex ISAR image, intensity and phase, is analyzed rather than only the intensity. This allows us to identify spurious responses that are generated by certain features on the target, but appear in locations other than those of the features. The analysis of the complex image permits us to associate the genuine image responses with the features responsible for the responses, so that the strength and type of backscattering can be determined for the target features. With respect to the type of backscattering, we can determine whether the effective location of the feature is stable, or whether it drifts with aspect angle or frequency. We. can also determine the effective crossrange and range widths of the various features. The features that can be analyzed are those with responses sufficiently strong to exceed the general background. This is typically a fairly large number.

For frequencies above 30 GHz, RCS reference target method is, in general, more accurate than scanning the field by a probe. Application of mechanically calibrated targets with a surface accuracy of 0.01 mm means that the phase distribution can be reconstructed accurately within approximately 1.2 degrees across the entire test zone at 100 GHz. Furthermore, since the same result can be obtained for both azimuth and elevation patterns, all data is available for the characterization of the entire test zone. In fact, due to the fact that the reference target has a well known radar cross-section, important indication of errors in positioning can be obtained directly from angular data as well. In the first place the data can be used in order to recognize the first order effects (+/- 5 degrees in all directions). Applying this data, defocussing of the system reflector or transverse and longitudinal CATR feed alignment can be recognized directly. Furthermore, mutual coupling can be measured and all other unwanted stray radiation incident from larger angles can be recognized and localized directly (using time­domain transformation techniques). Inmost cases even a limited rotation of +/- 25 degrees in azimuth and +/- 10 degrees in elevation will provide sufficient data for analysis of the range characteristics. Finally, it will be shown that sufficient accuracy can be realized for frequencies above 100 GHz with this method.

A method is presented that gives a 3D ISAR image from a 2D measurement. An ordinary 2D image is created. An extra receive channel is used to give height information in every pixel. This channel gets its signal from two extra receive antennas with different elevation lobes. The antennas feed a hybrid which creates a difference signal that goes to the extra receive channel. Height information is derived like in an amplitude comparison monopulse radar. This way a height number is assigned to every pixel in the 2D image. Thus a 3D image is created. It is required that in a 2D resolution cell the reflexes come from only one height. If not, the height information given by the difference signal will be a weighted average of the heights of the reflexes. The method is applied to a ground plane RCS range. No measurements have yet been performed.

A waterline bistatic algorithm, based on the exact near field to far field transformation (NFFFT) and previously exercised on numerical data, is here applied to actual measured data taken at a traditional RCS range reconfigured for near field measurements. The resulting far field predictions for a lOA and 20A conesphere were initially worse than expected. Further examination of the data yielded two important observations. First, the data were found to have relative alignment errors from set to set, leading to a significant broadening of the predicted far field peaks. Second, a few data sets exhibited a constant phase offset inconsistent with the other measured data. This paper discusses the detection of the data misregistration issues highlighted above, along with their ad hoc correction. Predictions are give for the waterline bistatic NFFFT algorithm applied to the measured near field data, both before and after the corrections have been applied. The results are compared with analogous results for numerical input data.

This paper deals with High Resolution (HR) Filtering. Extracting the frequency dependence of radar scatterers is a common task in Radar Cross Section Analysis (RCS). This is usually achieved with signal processing tools like finite impulse response filters allowing filtering in the range domain. However, when range resolution is poor, it becomes impossible to extract the exact feature since it is not deconvolved in the range domain. We thus propose to use HR methods to overcome these difficulties. These methods are applied to estimate the frequency response of the creeping wave of a small sphere. The results show good agreement with the theoretical response.

Instrumentation Radar systems evolution includes improved stability. Metrologists know frequency within Hertz. Amplitude and Phase variations are low. Ranges check drift with reference systems. Still, with increased capability, expectations of accuracy have increased. Todays instrumentation makes analysis of stability factors practical. This study analyzes Radar Cross Section (RCS) return of a stable target under controlled conditions. Methodology will be an analysis of a constant RCS target return. The target is a stable object at a typical measurement site. Data points are at several discrete frequencies in bands between S and Ku. This study sample is a set of data taken over a 87 hour span with several duty factors. Duty factors will range from minimal 0.1% to 1.5%, near the 2% maximum for the output amplifiers. Acquisition times for data sets are chosen for outdoor temperatures ranging from hot -- desert afternoon -- through cool in the early morning. This data will be analyzed statistically. If statistical correlations exist, analysis will quantify factor contributions with multiple linear regression. Hypothesis: Drift does not correlate to variables such as duty factor, & temperature.

Recently, we proposed a new method for the testing of antennas or the measurement of RCS in submillimeter wave region. A specific feature of this technique resides in that investigated object or its scaled model is mounted inside a quasi-optical waveguide in the form of a circular hollow dielectric waveguide (HDW) so as to determine the scattering parameter of the waveguide dominant HE11 mode which is certainly related to the wanted RCS of the object under study. In this paper, we intend to theoretically substantiate the proposed method for measuring RCS inside a circular HDW by using geometrical optical ray representation of guided modes and "virtual" waveguide concept. Then, a correspondence between RCS of an object inside a HDW and in a free space is established. Also, RCS of reference objects such as a perfectly conducting square flat plate inside a circular HDW are measured and compared with predicted returns in free space.

ORBIT/FR has recently installed and qualified a combined microwave (2-18 GHz) and millimeter wave (92.5-95.5 GHz) RCS system in an existing compact range based chamber. The facility is used for scale model reflectivity measurements on a wide variety of targets. The system features a unique, high power hardware gating system at the millimeter wave band that contains an integrated compact range feed assembly specifically designed to optimize RCS performance. Changeover between the microwave and millimeter wave bands is possible by placement of the appropriate compact range feed assembly on the feed stand, with locating pins being utilized to assure repeatable performance of the feeds in the compact range system. The system utilizes the FR959 RCS Measurement Workstation and HP 8530/85330 "turbo" based receiver system. Appropriate upconversion and downconversion hardware is integrated into the millimeter wave gating system to allow a common set of HP 8360 series sources and the HP 8530 IF receiver to be utilized for operation in both bands. The system is capable of producing high quality ISAR images at the millimeter wave frequencies, as well as in the microwave band.

ORBIT/FR has recently delivered an integrated facility capable of being used for Antenna, Radar Cross Section (RCS), and EMI measurements to the Naval Underwater Warfare Center in Newport, RI. The facility includes a shielded anechoic chamber, a compact range system capable of producing a 6 foot diameter quiet zone, multi-axis positioning equipment, and a complete complement of Antenna, RCS, and EMI measurement instrumentation and data collection hardware/software. The facility is capable of operation over a frequency range of 100 MHz to 50 GHz, with compact range operation feasible above 2 GHz. The facility can be reconfigured to go between antenna and RCS measurements in any band using both frequency band and antenna/RCS mode switching. In addition, automatic positioning of the appropriate compact range feed to the reflector focal point is available. EMI measurements require minimal relocation of absorber in an isolated area of the chamber floor. Performance of the system is optimized by location of critical RF equipment on the compact range feed carousel or on the positioning system rail carriage. This system offers a unique combination of performance and convenience for making all three types of measurements.

This paper discusses Radar Cross Section (RCS) measurement capability at Very High Frequencies (VHF) in the Boeing 9-77 Range in Seattle, Washington. This indoor facility provides a unique asset to the RCS measurement community. Initially operational in 1989, the 9-77 Range was upgraded in 1995 to include a VHF measurement capability. This was achieved using a 56 foot square array of 256 elements, for RCS measurements at frequencies from approximately 140 to 220 MHz, with a 40 foot quiet zone. In this paper, we discuss results from the characterization process used to verify the initial capability and ongoing operation of the RCS measurement system at VHF. We include data demonstrating the sensitivity, stability and dynamic range of the system. We also present samples of recent field probes, and background subtraction and stability measurements. A comparison is made between calculated and measured canonical target signatures.

Recently, several deployable, ground-to-ground col­ lection systems have been developed for the assessment of aircraft RCS on the flight-line. The majority of these systems require bulky rail or scanning hardware in order to collect diagnostic imaging data. The measurement technique described in this paper, while not a "cure-all", does eliminate the need for bulky hardware by allowing the collection system to move freely around the target while collecting radar backscattering data. In addition, a nearfield-to-farfield transformation (NFFFT) algorithm is incorporated in the process to allow the collection of scattering data collected in the near field to be processed and evaluated in the far field. The techniques described in this paper are a part of a data conditioning process which improves the data quality and utility for subsequent analysis by an automated diagnostic system described elsewhere in this proceedings [1]. The techniques are described and demonstrated on numerically simulated and experimentally measured data.

The RATSCAT radar cross section (RCS) measurement facility at Holloman AFB, NM is working to satisfy DoD and customer desires for certified RCS data. This paper discusses the low frequency characterization of the RATSCAT VHF/UHF Measurement System (RVUMS). The characterization was conducted on a portable pit with a 30' foam column at the RAMS site. System noise, clutter, backgrounds and generic target measurements are presented and discussed. Potential error sources are examined. The use of background subtraction and full polarimetric calibration are presented. Potential errors, which can occur from using certain cross-pol calibration techniques, are discussed. The phase relationship between each polarization components of the scattering matrix and cross-pol validation techniques are considered.

Over the past few years several flightline RCS measurement systems have been developed. Some of these systems use a linear rail to collect aspect angle data and other systems use no rails or "free path" circular collections. A comparison of these two different collection methods have never been publicly presented. In this paper, a discussion of the differences between the Linear-SAR and Circular-SAR collection methodologies will be presented. Specifically, issues associated with field quality, nearfield effects, and processing requirements will be covered in the discussion. Linear-SAR has the advantage of being more easily controlled and therefore not requiring motion compensation. Linear-SAR systems generally do not have mechanisms to point the antenna toward the target, thus limiting the angle extent of the collection aperture. In contrast, the Circular-SAR can theoretically collect data 360 degrees around the target at a constant range. However, the free-path Circular­SAR requires some form of motion compensation of the data for image formation processing.

Target support and clutter contamination can be a limiting factor in radar cross section (RCS) measurements of signature­ controlled targets. Conventional ISAR image editing methods can be used to remove contamination, but their performance degrades rapidly when the available resolution is insufficient to identify and separate the support returns from those of the target. ERIM International, Inc. (EI) has developed and successfully demonstrated data post-processing techniques based on 1-D parametric spectral estimators for removing additive contamination from low resolution swept frequency measurements [1, 2]. To further enhance performance and take advantage of the cross-range resolution afforded by target aspect information, EI has investigated the use of coherent 2-D spectral estimation techniques for improved identification and mitigation of measurement contamination in frequency and angle diverse data. In particular, parametric signal history editing (PSHE) algorithms based on 2-D TLS-Prony [3] and 2-D MEMP [4] have been developed and exercised on numerical simulations and measured data. The paper demonstrates 2-D spectral estimation in representative measurement situations, identifies strengths and limitations, and quantifies mitigation algorithm performance. In addition, automated filtering of spectral representations using energy level ordering, Cramer Rao Bounds (CRBs), and spatial filtering are discussed.

The close proximity of the ground to the radar antenna and the target under test is often hard to avoid at an outdoor RCS measurement range. Ground reflection of energy from the antenna leads to target illumination errors, and target-ground interactions lead to multipath errors. By proper positioning of the antenna and target, ground reflections of the antenna illumination can be exploited to increase overall system sensitivity by concentrating more energy on the target; however, this is only effectivefor narrowband measurements over a limited target region [1]. Reducing target-ground interactions by increasing the target height above the ground generally has limits due to mechanical restrictions on both the radar antennas and the target. This paper will present a model-based data post-processing technique to mitigate illumination errors and target-ground interactions in ground plane range RCS measurements. The algorithm is an extension of the network model multipath mitigation technique previously developed for indoor RCS measurement ranges [2,3,4]. The technique will be described and demonstrated using a numerical simulation of the RCS measurement of a canonical target over a ground plane.

The Seeker Test and Evaluation Facility (STEF) located on Range C-52A at Eglin AFB FL. is used to perform high-resolution multispectral (EO-IR-RF-MMW) signature measurements of US and foreign ground vehicles primarily to support the Research, Development, Test and Evaluation (RDT&E) of smart weapons (seekers, sensors and Countermeasure techniques). In order to support two major DOD signature measurement programs in 1997 this facility required significant range upgrades and enhancements to realize reduced background levels, increase measurement accuracy and improve radar system reliability. These modifications include the addition of a 350'X 120' asphalt ground plane, a new secure target support facility, a redesigned low RCS shroud for the target turntable and a new core radar system (Lintek elan) and data acquisition/analysis capability for the existing radars Millimeter-Wave Instrumentation, High Resolution, Imaging Radar System - MIHRIRS). This paper describes the performance increase gained as a result of this effort and provides information on site characterization and radar instrumentation improvements as well as examples of measured RCS of typical ground vehicle signatures and ISAR imagery

The scattered field from an arbitrary target may include a variety of scattering mechanisms such as specular and diffraction terms, creeping waves and resonant phenomena. In addition, buried within such data are target-mount interactions and clutter terms associated with the test environment. This research presents a method for decomposing a broadband complex signal into its constituent mechanisms. The method makes use of basis functions (words) which best describe the physics of the scattered fields. The MUSIC algorithm is used to estimate the time delay of each word. A constrained optimization refines the estimate and determines the energy for each. The method is tested using two far-field radar cross section (RCS) measurements. The first example identifies target­mount interactions for a common calibration sphere. The second example applies the method to a low observable (LO) ogive target.

Low dielectric permittivity columns are often used in RCS measurements to support targets. Electromagnetic and mechanical interactions between target and mast occur and subtraction cannot eliminate them. In this paper we will study mechanical phenomena such as bending and compression (up to buckling) under load and we will calculate the dielectric mast backscattering level owing to these two deformations. It appears that compression effects are usually negligible compared to bending effects. Finally, we will propose some rules on mast design. More specifically, a cubic section can perform the support of the targets and so decrease the spurious backscattering.

Full polarimetric scattering measurements are increasingly being required for radar cross-section (RCS) tests. Conventional co-and cross-polarization calibrations fail to take into account the small amount of antenna cross-polarization that will be present for any practical antenna. In contrast, full polarimetric calibrations take into account and compensate for the cross-polarization the calibration process. We present a full polarimetric calibration procedure and a simulation-based performance study quantifying how well the procedure improves measurement accuracy over conventional independent channel calibration.

Recently, RCS measurements were made of several common calibration objects of various sizes in the Boeing 9-77 Range. A study was conducted to examine the accuracy and errors induced by using each as a calibration target with a string support system. This paper presents the results of the study. Two of the objects, i.e., the 14"-ultrasphere and the 4.5"-dia. cylinder, are found to perform the best in that they exhibit the least departures (error) from theory. The measured departures of 0.2 to 0.3 dB are consistent with the temporal drift of the radar in several hours.