Synthetic Aperture Radar (SAR) was invented by Carl Wiley at Goodyear Aircraft Company in Goodyear, Arizona, in 1951. From that time forward, as the company became Goodyear Aerospace Corporation, Loral Corporation, and finally Lockheed Martin Corporation, the Arizona employees past and present played a long and storied role in numerous SAR firsts. These include the original SAR patent (known as Simultaneous Doppler Buildup), the first demonstration SAR and flight test, the first operational SAR system, the first operational SAR data link, the first 5-foot resolution operational SAR system, the first 1-foot resolution SAR system, and the first large scale SAR digital processor. The company has installed and flown over five hundred SAR systems on more than thirty different types of aircraft for numerous countries throughout the world. The company designed and produced all of the evolving high performance SAR systems for the U. S. Air Force SR-71 “Blackbird” spy plane throughout its entire operational history, spanning some twenty-nine years. Recent SAR accomplishments include long-range standoff high performance SAR systems, smaller high resolution podded SAR systems for fighter aircraft, and foliage penetration (FOPEN) SAR. The company is currently developing the high performance SAR/MTI (Moving Target Indication) radar for the Army Aerial Common Sensor (ACS) system.

Sandia National Laboratories designs and builds Synthetic Aperture Radar (SAR) systems capable of forming high-quality exceptionally fine resolution images. During the spring of 2004 a series of test flights were completed with a Ka-band testbed SAR on Sandia’s DeHavilland DHC-6 Twin Otter aircraft. A large data set was collected including real-time fine-resolution images of a variety of target scenes. This paper offers a sampling of high quality images representative of the output of Sandia’s Ka-band testbed radar with resolutions as fine as 4 inches. Images will be annotated with descriptions of collection geometries and other relevant image parameters.

Airborne synthetic aperture radar (SAR) imaging systems have reached a degree of accuracy and sophistication that requires the validity of the free-space approximation for radio-wave propagation to be questioned. Based on the thin-lens approximation, a closed-form model for the focal length of a gravity wave-modulated refractive-index interface in the lower troposphere is developed. The model corroborates the suggestion that mesoscale, quasi-deterministic variations of the clear-air radio refractive-index field can cause diffraction patterns on the ground that are consistent with reflectivity artifacts occasionally seen in SAR images, particularly in those collected at long ranges, short wavelengths, and small grazing angles.

Relatively small motion measurement errors manifest themselves principally as a phase error in Synthetic Aperture Radar (SAR) complex data samples, and if large enough become observable as a smearing, blurring, or other degradation in the image. The phase error function can be measured and then deconvolved from the original data to compensate for the presumed motion error, ultimately resulting in a well-focused image. Techniques that do this are termed “autofocus” algorithms. A very popular autofocus algorithm is the Phase Gradient Autofocus (PGA) algorithm. The nearly universal, and typically reasonable, assumption is that the motion errors are less than the range resolution of the radar, allowing solely a phase correction to suffice. Very large relative motion measurement errors manifest themselves as an unexpected additional shifting or migration of target locations beyond any deterministic migration during the course of the synthetic aperture. Degradation in images from data exhibiting errors of this magnitude are substantial, often rendering the image completely useless. When residual range migration due to either real or apparent motion errors exceeds the range resolution, conventional autofocus algorithms fail. Excessive residual migration is increasingly encountered as resolutions become finer, less expensive inertial sensors are used, and operating ranges become longer (due to atmospheric phenomena). A new migration-correction autofocus algorithm has been developed that estimates the excessive residual migration and applies phase and frequency corrections to properly focus the image. This overcomes the conventional constraint that motion errors not exceed the SAR range resolution.

Individuals who carry bombs on their bodies and detonate those bombs in public places are a security problem. There is belief that suicide bombings currently used in the mid-east may spread to the United States if the organized terrorist groups operating in the United States are not identified and the cell members arrested. While bombs in vehicles are the primary method currently used to spread terror in Iraq, U. S. warfighters are starting to face suicide bombers. This may become more of the situation if a stand-off detection capability is developed for the vehicle bomb case. This paper presents a concept, that if developed and commercialized, could provide an inexpensive suicide bomber screening system that could be used to screen individuals approaching a checkpoint while the individual is still 500 to 1,000 feet from the checkpoint. The proposed system measures both the radar cross-section of the individual and the radar derived gait characteristics that are associated with individuals carrying a bomb on their body. GTRI researchers propose to use human gait characteristics, as detected by radar, to determine if a human subject who is carrying no visible load on the body is actually carrying a concealed load under their clothes. The use of radar gait as a metric for the detection (as opposed to a video system) of a suicide bomber is being proposed because detection of gait characteristics are thought to be less sensitive to where the bomb is located on the body, lighting conditions, and the fact that the legs may be shrouded in a robe. The detection of a bomb using radar gait analysis may also prove to be less sensitive to changing tactics regarding where the bomb is placed on the body. An inert suicide bomb vest was constructed using water pipes to simulate the explosive devices. Wiring was added to simulated detonators. The vest weighs approximately 35 pounds. Radar data was taken on the volunteer subject wearing the vest that simulated the suicide bomb. This paper discusses the findings after the data was analyzed. A Provisional Patent has been filed by Georgia Tech Research Corporation on the subject matter that is discussed in this paper.

We investigate the use of a low-cost, two-element receiving array for tracking human movements in indoor surveillance applications. Conventional direction of arrival (DOA) detection requires the use of an antenna array with multiple elements. Here we investigate the use of only two elements in the receiver array. The concept entails simultaneously resolving the Doppler frequencies of the returned signals from the moving targets and the DOA of the targets. Simulation is performed to demonstrate the concept. Both the monostatic and the bistatic scenario where the transmitter and the receiving array are placed at different locations are investigated. DOA errors and tolerances are analyzed for each scenario. An experimental system is constructed to test the concept. The system consists of a two-element receiver array operating at 2.4 GHz. Measurement results of various collection scenarios are presented.

The results of numerical electromagnetic simulation and analysis of a set of positive-adaptive UAV radar signals are presented. These signals are simulated via the modeling of materials that enclose “building-type” structures with a series of connected dielectric materials. For example, windows, walls, and doors are each modeled separately by a combination of suitable material properties. Signals from objects that are embedded within these “building-type” structures are also simulated via the development and application of appropriate geometrical and materials models. Analysis of the resulting simulated “leakage signals” that penetrate the surfaces of these “building-type” structures and are backscattered from embedded objects within the indoor environment back to the simulated outdoor environment are presented. The results of a signal analysis are presented in two categories. The first set of results illustrates signal trends that can be exploited by “position-adaptive” mini-UAV's to isolate effective “leakage points” in “building-type” structures. The second set of results illustrate signal trends from embedded objects after a particular “position-adaptive” mini-UAV has converged to a “leakage point.”

Random noise radar has recently been used in a variety of imaging and surveillance applications. These systems can be made phase coherent using the technique of heterodyne correlation. Phase coherence has been exploited to measure Doppler and thereby the velocity of moving targets. The Doppler visibility, i.e., the ability to extract Doppler information over the inherent clutter spectra, is constrained by system parameters, especially the phase noise generated by microwave components. Our paper proposes a new phase noise model for the heterodyne mixer as applicable for ultrawideband (UWB) random noise radar and for the local oscillator in the time domain. The Doppler spectra are simulated by including phase noise contamination effects and compared to our previous experimental results. A Genetic Algorithm (GA) optimization routine is applied to synthesize the effects of a variety of parameter combinations to derive a suitable empirical formula for estimating the Doppler visibility in dB. According to the phase noise analysis and the simulation results, the Doppler visibility of UWB random noise radar depends primarily on the following parameters: (a) the local oscillator (LO) drive level of the receiver heterodyne mixer; (b) the saturation current in the receiver heterodyne mixer; (c) the bandwidth of the transmit noise source, and; (d) the target velocity. Other parameters such as the carrier frequency of the receiver LO and the loaded quality factor of the LO have a small effect over the range of applicability of the model and are therefore neglected in the model formulation. The Doppler visibility curves generated from this formula match the simulation results very well over the applicable parameter range within 1 dB. Our model may therefore be used to quickly estimate the Doppler visibility of random noise radars for trade-off analysis.

Ground penetrating radar investigations of buildings and other man-made structures often require data collection along curvilinear interfaces having various radii of curvature and geometry. Data collected along non-planar interfaces and containing multiple targets are often difficult to interpret without the application of migration. Migration is not commonly applied to surveys along non-planar interfaces because most published algorithms and available software assume data collection along planar interfaces. We provide a versatile imaging strategy that migrates data by using a kernel derived from the forward model. Modeled and measured data from columns are used to illustrate the utility of our imaging approach.

In this note we show how to arrive at the characteristic function for sum of N sinusoidal random variables under the conditions that a characteristic function for the individual component can be calculated. Using this formalism, it is possible to generalize Huygen's principle to random variables with each point of space to be an antenna with a gain pattern that is a superposition of random variables. Applications of this principle to several problems of interest is then briefly discussed.

We evaluated two random number generator algorithms using first-order and second-order chaotic maps. The first algorithm, which is based on the central limit theorem, allows us to approximate a Gaussian random variable as the sum of a given chaotic sequence. We considered two first-order maps (Bernoulli, Tent) and two second-order maps (Logistic, and Quadratic). In each instance, we verified that the sequence of random numbers had kurtosis of 3. In the case of the Bernoulli map, we determined that the statistical independence of samples is dependent on the map parameter B. The second algorithm, which is based on Von Neumann's Method, allowed us to reject samples from a chaotic sequence with uniform distribution to obtain a Gaussian distribution within a specific range (U, V). For the first-order maps, we estimated their probability density function in this range and computed deviations from the theoretical Gaussian density. In summary, we determined that samples generated via these two algorithms satisfied statistical tests for normal distributions, thus demonstrating that chaotic maps can be effectively to generate Gaussian samples.

Efforts are being made to exploit the full-polarimetric radar scattering nature of ground targets to extract maximum information, enabling target identification and classification. These efforts have taken varied approaches to decomposing the polarimetric scattering matrix into more meaningful, phenomenological parameter spaces. The Euler parameters have potential value in target classification but have historically met with limited success due to ambiguities that arise in the decomposition as well as the parameters sensitivity to noise and target movement. Using polarimetric ISAR signatures obtained from stationary targets in compact radar ranges at the University of Massachusetts Lowell Submillimeter Technology Laboratory (STL)1,2,3,4 and the U.S. Army National Ground Intelligence Center (NGIC), correlation studies were performed in the Euler parameter space to assess to its impact on improving target classification. Methods for deriving explicit transform equations that minimize ambiguities will be presented, as well as the results of the correlation studies.

The main focus of this paper is the development of fusion strategies for multiple location synthetic aperture radar (SAR), and inverse synthetic aperture radar (ISAR) images. The techniques being developed are to be used in conjunction with super-resolution and target identification strategies for non-cooperative target recognition (NCTR). Multiple location processing has the ability to provide improved image quality as well as target detection and classification capabilities since the different aspects or “looks” can provide additional clues about the shape, dimensions, and special features of a target. Many traditional SAR/ISAR processing techniques seek to maximize the instantaneous SNR for a signal in the presence of additive noise. Unfortunately, these techniques do not directly address the recreation of an image with minimum mean squared error between the reconstructed SAR image and the reflectivity map of the actual scene. This paper examines techniques capable of improving the probability of object detection within an image generated via spatial fusion. The strategies focus on image level fusion of the SAR/ISAR data. Canonical SAR/ISAR data is used to validate and compare the fusion results. Preliminary results using DARPA's MSTAR database are also presented.

We present a procedure wherein unattended ground sensors (UGSs) that are not equipped the GPS can locate their own positions by transmitting pulses and receiving retransmitted pulses from UGSs that are equipped the GPS. T The payoff of this approach is reduced cost for the network of UGSs. We show through simulation that the implementation of this procedure locates the sensors that do not have GPS with sufficient accuracy for the network of UGS to detect and locate moving targets.

Traditionally, ultra-wideband radars increase bandwidth by using shorter pulses. However, by decreasing the pulse width, the power on target decreases and radar detection probabilities decrease. Therefore, new approaches to increase bandwidth are needed that still have adequate power on target. One radical new approach is to use a Transform Domain Communication System (TDCS) as an ultra-wideband radar. The primary advantage of this technique is that it has properties similar to Gaussian noise meaning that the radar would improve bandwidth similar to pseudo-noise (PN) sequences. Also, based on the number of carriers used to generate the TDCS code, multiple pulses can be made mutually orthogonal. This orthogonality can be exploited to effectively increase the maximum unambiguous range for pulse-Doppler radars up to the range of the horizon. In essence, TDCS radar possesses a high pulse repetition frequency (PRF) for velocity estimation and large unambiguous range.

This study investigates the modeling of through-wall sensing using ultra-wideband (UWB) signals. The combined method of ray tracing and diffraction (CMRTD) is employed to model and study the interaction between the UWB signal and the target. The result is obtained in frequency domain, and then transformed into time domain by use of inverse Fourier transform. Numerical results of scattering from a two-dimensional (2D) perfectly conducting circular cylinder are obtained and compared with those from the eigenfunction expansion method. Good agreements between the results are achieved. In addition, the attenuating effects of walls are considered and numerical result of scattering from a 2D perfectly conducting circular cylinder behind a homogeneous, single-layered dry wall is presented. The model can be easily extended to handle the dielectric target and the multiple-layered walls.

Effectiveness of modern weapon systems demands that it is important to distinguish between friends and foes as early as possible. Radar signal based helicopter categorization is a challenging task for all types of radars. Airborne pulse Doppler radar with an appropriate digital signal processing unit has a good potential to perform categorization or even classification, providing that radar parameters are carefully fixed. Moreover, some information about the main rotor parameters of interesting helicopter types must also be known in advance. The idea of this paper is to present a helicopter categorization method, which is based on estimates of the main rotor blade tip velocity and the time interval between successive main rotor blade flashes. Both incoherent integration and conventional coherent integration play an important role in the new method. Moreover, a new edge detection algorithm is applied to coherently integrated signal. Simulations are performed in order to show the effectiveness of the method.

In this paper Ka-band (37GHz), dual polarization, combined short-pulse scatterometer-radiometer is described, for short distance remote sensing of bare soil and land snow cover and for simultaneous and coincident measurements of observed media microwave reflective and emissive characteristics, under laboratory-control conditions. Developed system is set on a mobile bogie moving on the height of 6.5m along a stationary platform of 26m of length. It allows carry out polarimetric (vv, vh, hh, hv), simultaneous and coincident microwave active-passive measurements of observed surface (soil, soil vegetation, snow and water surface) parameters at angles of incidence from the while of 0-60^o.

Sandia National Laboratories designs and builds Synthetic Aperture Radar (SAR) systems capable of forming high-quality exceptionally fine resolution images. During the spring of 2004 a series of test flights were completed with a Ka-band testbed SAR on Sandia’s DeHavilland DHC-6 Twin Otter aircraft. A large data set was collected including real-time fine-resolution images of a variety of target scenes. This paper offers a sampling of high quality images representative of the output of Sandia’s Ka-band testbed radar with resolutions as fine as 4 inches. Images will be annotated with descriptions of collection geometries and other relevant image parameters.

The technique of bistatic RCS estimate is proposed for sea and terrain clutter with application of monostatic RCS measured data for small grazing angles that can be applied in centimeter and millimeter bands of radio waves. The models of bistatic scattering are analyzed including the model of small perturbations, facet model and theorem of equivalence; the limitations of these models are discussed. The features of bistatic scattering near the specular points (the forward scattering) are analyzed and estimates are carried out the RCS values for different terrain types. In directions differing from quasi-specular one, the normalized bistatic RCS dependences on grazing angles and bistatic angle are obtained for different types of terrain and for sea at frequencies of 10.0 and 35.0 GHz. These data are founded at results of RCS measurements carried out by author for monostatic regime. The comparison of calculated results with experimental data available for bistatic radar was carried out. It is shown the good coincidence of these results. Thus, the proposed technique permits to evaluate the bistatic RCS using the data of monostatic measurements for small grazing angles at wide band of frequencies up to 95 GHz.

This paper presents real time optical positioning technology developed by ENSCO, Inc. and a system-level concept for its application to wall penetration and wall interior imaging. The concept merges this positioning technology, commercial Ground Penetrating Radar (GPR) systems, and software imaging techniques. We demonstrate existing real-time optical positioning hardware and present the concept for the next generation. This hardware integrates with a commercial GPR system, triggering it in real time. Software techniques, such as 3D migration, process the data and make them suitable for display and interpretation by inexperienced operators in near real time.

In this paper C-band (~5.75GHz), dual polarization, Doppler scatterometer is developed, for short distance remote sensing of water surface microwave reflective and spectrum characteristics simultaneous and coincident measurements, under laboratory-control conditions. Developed system will be set on a mobile bogie moving on the height of 6.5m along a stationary platform of 32m of length. It will allow carry out polarimetric (vv, vh, hh, hv), simultaneous and coincident microwave active measurements of pool water surface parameters at angles of incidence from the while of 0-40^o.

A concept of combining data of altimeter and slight tilted radiometer observations and a microwave active-passive, combined method of detection and identification of sea surface signatures are presented. Developed method allows detect the sea surface microwave slight-contrast signatures and identify precisely the origins of their formation.

A need exists for greater situational awareness at the lower echelons of the Army. Radar Frequency (RF) sensors on small, lightweight Unmanned Aerial Vehicles (UAV) could provide lower echelon commanders with all-weather reconnaissance, early warning, and target acquisition; however, the designs of these RF sensors are limited by the projected size and weight restrictions on the payload for a class II UAV. Consequently, these designs may favor combining simple RF sensor hardware with digital-signal processing (DSP) solutions over more sophisticated radar hardware. In this paper, we show the potential of simple, low cost RF sensors with hemispherical antenna coverage to overcome these limitations. The proposed RF sensor system used DSP and pre-defined UAV flight pattern to detect and track moving targets from range and Doppler information. Our objective is to conceive and model a suite of software options that, by combining UAV flight patterns and processing algorithms, will be able to detect and track moving targets. In order to accomplish this, we are building a simulation that uses sensor models, target models, and battlefield dynamics to predict the targeting capabilities of the RF sensor system. We will use this simulation (1) to determine the tradeoffs between sensor complexity (and cost) and the military significance of the information gathered, and (2) to describe sensor error budgets for endgame lethality models