This paper addresses how a data set conducive to efficient processing might increase the total RF bandwidth, and presents examples of how a fixed RF bandwidth might then limit SAR geometries.

The Georgia Tech Research Institute has designed a radar detector detector (RDD) capable of sensing the presence of a radar detector in a moving vehicle at a distance of up to several miles, depending on the terrain. The RDD was designed for use in a radar detector density survey as part an ongoing United States Department of Transportation project to measure the potential impact of the Safety Warning System^TM on motorists in a work zone. In Canada and the two U.S. states where radar detectors are outlawed, law enforcement uses VG-2 detectors able to sense the leakage of the radar detector's local oscillator (LO). Due to the radar detector industry's stance that a radar detector is simply a radio receiver, the industry responded by adding countermeasure features. One type of countermeasure turns off the radar detector LO when the leakage from the VG-2 LO is detected. Another method reduces the radar detector LO leakage to levels nearly impossible to detect using the VG-2.

GTRI is conducting research on the Safety Warning System (SWS), an off-the-shelf highway safety system that contains a 24 GHz motorist communications system and 24 GHz homodyne radar. This system is being evaluated to determine if it can reduce these types of farm equipment accidents. These research being conducted by GTRI on farm equipment accidents is part of a more comprehensive Federal Highway Administration research project being conducted on vehicular safety technology. The goal of this research, as it relates to farm equipment safety, is to determine if the SWS system can be used to warn both the approaching driver and farm equipment operator. Specifically, can the homodyne radar be used to warn the farm equipment driver of a motorist's approach and can the approaching driver equipped with an SWS receiver be warned of the farm equipment's presence in time to avoid a collision.

We analyzed the feasibility of an all-weather, low-cost millimeter wave sensor for the detection of obstacles on the paths of the trains. We formulated the requirements for such a sensor and experimentally demonstrated radar detection of vehicles and trains.

The purpose of this paper is to use various mathematical methods for processing precipitation data and presenting an independent ground-based measurement of rainfall rate and accumulation for calibration of TRMM Precipitation Radar (PR). Reflectivity data from the Melbourne National Weather Service WSR-88D Next generation Radar (NEXRAD) were collocated in time and space with gauge data located in Central and South Florida. The corresponding NEXRAD reflectivity (Z) and gauge rain states (R) were matched. Different regression methods were tested to find a suitable Z-R relationship for convective and stratiform rainfall for Melbourne NEXRAD. This Z-R relationship was used to produce instantaneous rain rate maps, and monthly rain accumulation maps. This resulted in a large coverage area for comparison with TRMM PR rain products; instantaneous rain rate, and average monthly rain rate. Statistics are presented for three months operations during August and September 1998, and June 1999.

A method of moment (MoM) analysis is developed for electromagnetic scattering from a generalized perfectly conducting target in the near field of a tree trunk in a layered medium environment. In this analysis, the tree trunk is modeled as a dielectric body of revolution and the layered medium electrical properties can be lossy and dispersive, of interest for simulating real soil. The MoM analysis employs the layered medium Green's function, which is evaluated efficiently via the method of complex images. To simplify the analysis, the conducting target is considered to be a flat plate. To rigorously account for the interaction between these disparate targets, the conducting target and tree trunk are modeled separately, with interactions handled via an efficient iterative procedure. In addition to yielding accurate results, this procedure has memory and run-time requirements that are significantly less than required of a straightforward brute force MoM approach. This latter issue is particularly important for the problem of interest here, since the tree trunk and conducting target are generally electrically large and because this work is ultimately directed towards modeling conducting targets in the near field of multiple tree trunks (that is, simulating targets concealed in tree foliage).

The multi-level fast multipole algorithm (MLFMA) is applied to the problem of scattering from surface and subsurface targets. In this paper we demonstrate how the MLFMA is modified to handle the half-space problem, and present example results for several scattering problems of interest. In particular, we present results for scattering from buried unexploded ordnance.

We present results of an unexploded ordnance (UXO) detection algorithm based on template matching in ultra-wideband (130 MHz to 1.2 GHz) synthetic aperture radar (SAR) data. We compute scattered fields of UXOs in different orientations, both on the surface and buried at different depths, using a physical optics (PO) approximation for perfectly conducting targets in a half space via the half-space Green's function. The PO code that we developed computes the scattered fields in a lossy and dispersive material. This permits simulation of targets in real soil. The frequency-domain scattered fields are transformed into time-domain. SAR images of the UXOs at different aspect angles are generated by a standard backprojection technique, with the same resolution as the ground-penetrating ultra-wideband SAR. These SAR images form the templates for detection of the UXOs.

The U.S. Army Research Laboratory is currently using an experimental radar with an impulse-based transmitter to gather data on subterranean targets. This synthetic aperture radar has an instantaneous bandwidth of 50 - 1100 MHz, with the upper frequency limit set by the bandwidth of the analog-to-digital (A/D) converter that acts as the baseband receiver for the impulse transmitter. The A/Ds are the biggest limiting factor for most impulse-based radars, as their analog bandwidth and high sampling rates push the state of the art, but effects of transmitter waveforms, antenna response, and preamplifier bandwidth must be considered as well. We discuss a number of design changes that will allow the use of existing A/D converters while expanding the frequency coverage by approximately a factor of two, with major changes to other hardware. The design models chosen are evaluated in a PV-WAVE framework that allows one to choose the transmit waveshape and assess the influences of phase stability, signal-to-noise ratio, and system and filter frequency response. A comparison of the model results is presented, along with tradeoffs and recommendations.

Polarimetric synthetic aperture radar (SAR) imagery is susceptible to degradation because of variations in transmit and receiver antenna patterns and hardware limitations on achievable polarimetric isolation. A number of polarimetric calibration procedures have been developed to compensate for these effects. However, most of these techniques were formulated with narrowband radar systems in mind. The few that are applicable to ultrawideband (UWB) SAR depend on very high quality antenna measurements or detailed electromagnetic scattering models of test targets. In this paper we develop and implement a blind polarimetric equalization (BPE) procedure suitable for UWB SAR. BPE uses second-order statistics from clutter and targets of opportunity to derive parameters describing the polarimetric response of the radar system. Knowledge of these parameters permits the removal of crosstalk and the correction of imbalances between the polarimetric channels. While not a complete calibration solution, BPE is shown to greatly simplify any subsequent calibration procedure. For UWB applications BPE is implemented as a series of frequency domain operations, a feature that sets this approach apart from previously reported clutter calibration algorithms. The performance of the technique was verified using narrowband inverse SAR measurements, UWB SAR simulations, and UWB SAR data collected by the Army Research Laboratory's BoomSAR system.

Urban-warfare specialists, law-enforcement officers, counter-drug agents, and counter-terrorism experts encounter operational situations where they must assault a target building and capture or rescue its occupants. To minimize potential casualties, the assault team needs a picture of the building's interior and a copy of its floor plan. With this need in mind, we constructed a scale model of a single- story house and imaged its interior using synthetic-aperture techniques. The interior and exterior walls nearest the radar set were imaged with good fidelity, but the distal ones appear poorly defined and surrounded by ghosts and artifacts. The latter defects are traceable to beam attenuation, wavefront distortion, multiple scattering, traveling waves, resonance phenomena, and other effects not accounted for in the traditional (noninteracting, isotropic point scatterer) model for radar imaging.

The wide waveform bandwidths desired for fine range resolution is current and future phased array radars have an impact on the performance of adaptive digital beamforming due to the inevitable frequency response mismatches that occur from channel to channel across the band. In previous work, analytical expressions for the covariance between channels have been developed for a simple two channel array with a sinusoidal ripple in either the amplitude or phase response in one of the channels. This paper extends those results to a more general model that allows larger arrays utilizing time delay steering to be modeled. Expressions are developed for the covariance between channels as a function of the parameters of the channel frequency responses. The performance of the adaptive beamformer is characterized with respect to SINR for waveforms that are flat over the bandwidth and for LFM waveforms processed using stretch processing.

The problem we are addressing is one of generalization: given training data characterizing a set of targets (in specific configurations), how can we design a classifier that is robust to changes in target configuration and can generalize to other targets of the same generic class? The specific problem is identifying land vehicles from an inverse synthetic aperture radar image of the target. Issues in data modeling, experimental design and exploratory data analysis are discussed. Two complementary approaches are described: one that seeks to capture structure in the high- dimensional data space by projecting the data nonlinearly to a reduced dimensional feature space prior to classification; and a second that models the data in the data space using a Bayesian mixture model approach. Preliminary results for the mixture model approach are presented.

A major multi-national measurement campaign was commissioned at Swynnerton in the UK; it was called MIMEX (Millimeter Wave Imaging Experiment). This activity produced a large volume of millimeter wave imagery both active and passive. In this paper the application of this imagery to surveillance and weapon systems will be discussed.

The present study looks at the problem of classifying ground targets based on the experimental signatures gathered with a millimeter-wave FMCW radar. The selection of features for robust target classification using circular polarization signatures is considered here. The technique relies on the concept of transient polarization and uses the circularly polarized responses of the targets to define the classification features. Each target can then be decomposed into substructures tracked during the measurements. Once the features of all the substructures are extracted, an artificial neural network is used to classify the signatures.

In the presentation it will be shown that the classification approach based on polarimetric features supplemented by geometrical information shows promising results using turntable measurements of one of the targets as reference data and airborne measurements of a set of targets as test data. Furtheron the results using the split measurement data sets (split into reference and test sets) will be discussed in detail. The two approaches are compared to each other and their respective advantages and disadvantages summarized.

Target detection and recognition using polarimetric SAR data has been studied by using PHARUS and RAMSES data collected during the MIMEX campaign. Additionally very high-resolution ISAR data was used. A basic detection and recognition scheme has been developed, which includes polarimetric speckle- filtering, CFAR detection and the extraction of geometrical, intensity and polarimetric features. From the SAR images we conclude that polarimetric features can be useful to discriminate targets from clutter. At resolutions of 1 meter or better, shape and orientation recognition can be obtained with these features. To classify the targets, other features or other techniques have to be used. Examples are polarimetric decomposition techniques, of which two have been explored using the ISAR data.

The paper consists of two parts. In the first part the main characteristics of heavy sea are discussed that are essential for radar observation at grazing angles, the sea surface spike characteristics are analyzed in the framework of the statistical theory of random process surges above some boundary, the sea surface shadowing function is considered. In the second part the radar spike characteristics of sea backscattering at frequencies of 10.0, 35.0, 75.0 and 140.0 GHz are analyzed.