Synthetic aperture radar (SAR) is an all weather sensor that has provided breakthrough remote sensing capabilities for both civilian and military applications. SAR differs from other real-aperture sensors in that it achieves fine resolution using signal processing techniques that are based on certain assumptions about the relative dynamics between the sensor and the scene. When these assumptions are violated, the quality of the SAR imagery degrades, impacting its interpretability. This paper describes the development of a simulation testbed for evaluating the effects of SAR-specific error sources on image quality, including effects that originate with the sensor (e.g. system noise, uncompensated motion), as well as effects that originate in the scene (e.g. target motion, wind-blown trees). The simulation generates synthetic video phase history and can accommodate a variety of sensor collection trajectories, acquisition geometries, and image formation options. The simulation approach will be described, example outputs will be shown, and initial results relating simulation inputs to image quality measures will be presented.

Sandia-developed SAR systems are well known for their real-time, high quality, high resolution imagery. Recently, a series of tests were completed with the sub-thirty pound, Ku-band miniSAR system. A large data set was collected including real-time images of a variety of target scenes with resolutions as fine as 4 inches. This paper offers a sampling of real-time, high quality, fine resolution images representative of the output of Sandia's miniSAR radar. Images will be annotated with descriptions of collection geometries and other relevant image parameters. The miniSAR system and Sandia DeHaviland DHC-6 Twin Otter test bed are also briefly discussed.

Previous papers have demonstrated that anomalies observed in synthetic aperture radar (SAR) images can be attributed to perturbations in the clear-air refractive index. The atmospheric effect is small, and is more frequent, or apparent, in long range imaging. A geometrical optics (ray-trace) analysis is applied to potential atmospheric perturbations. The results correlate with observations, and demonstrate that the effect is strongly dependent on the azimuth orientation of the atmospheric perturbation. The analysis also demonstrates that the magnitudes of the spatial size of the anomalies and index perturbations can be much smaller than expected.

The U.S. Army Research Laboratory (ARL) has recently evaluated a commercially available borehole radar to detect targets hidden underground. The goal of the experiment is to demonstrate the feasibility of the borehole radar coupled with ARL's signal and image processing techniques to penetrate various geophysical media for target detection. In this paper, we briefly describe the commercial ultra-wideband borehole radar used in the experiment. A conventional technique to provide the attenuation and velocity maps of the underground area between two holes is called tomography. It requires separate probes for the transmitter and the receiver for the measurement, and generally is more time consuming and laborious. Another technique known as reflection is also widely used. In this mode, the transmitter and receiver travel together as one single unit in one hole to measure the reflection data from surrounding clutter and underground targets. Although this mode is much simpler to operate than tomography, the resulting image has inferior resolution in the cross-range (depth) direction. In our experiment we employ this reflection mode, where a small cylindrical metal target is placed in one hole while the radar (both transmitter and receiver) travels in another hole to measure the target return. To improve the poor cross-range resolution associated with the reflection raw data image, we apply the backprojection image formation algorithm that is commonly used in synthetic aperture radar to form high resolution 2D images. We present the resulting images of background (without target) and with target, and show that the underground target can be easily detected using change detection technique. This paper also compares the measured data with the electromagnetic model prediction of the same target.

Researchers at Georgia Tech Research Institute have developed a high resolution imaging radar technique that allows large sections of a test wall to be scanned in X and Y dimensions. The resulting images that can be obtained provide information on what is inside the wall, if anything. The scanning homodyne radar operates at a frequency of 24.1 GHz at with an output power level of approximately 10 milliwatts. An imaging technique that has been developed is currently being used to study the detection of toxic mold on the back surface of wallboard using radar as a sensor. The moisture that is associated with the mold can easily be detected. In addition to mold, the technique will image objects as small as a 4 millimeter sphere on the front or rear of the wallboard and will penetrate both sides of a wall made of studs and wallboard. Signal processing is performed on the resulting data to further sharpen the image. Photos of the scanner and images produced by the scanner are presented. A discussion of the signal processing and technical challenges are also discussed.

Sandia-developed SAR systems are well known for their real-time, high quality, high resolution imagery. Recently, a series of tests were completed with the sub-thirty pound, Ku-band miniSAR system. A large data set was collected including real-time images of a variety of target scenes with resolutions as fine as 4 inches. This paper offers a sampling of real-time, high quality, fine resolution images representative of the output of Sandia's miniSAR radar. Images will be annotated with descriptions of collection geometries and other relevant image parameters. The miniSAR system and Sandia DeHaviland DHC-6 Twin Otter test bed are also briefly discussed.

Orthogonal frequency division-multiplexing (OFDM) is rapidly emerging as a preferred method of UWB signaling in commercial applications aimed mainly at low-power, high data-rate communications. This paper explores the possibility of applying OFDM to use in imaging radar technology. Ultra-wideband nature of the signal provides for high resolution of the radar, whereas usage of multi-sub-carrier method of modulation allows for dynamic spectrum allocation. Robust multi-path performance of OFDM signals and heavy reliance of transceiver design on digital processors easily implemented in modern VLSI technology make a number of possible applications viable, e.g.: portable high-resolution indoor radar/movement monitoring system; through-the-wall/foliage synthetic aperture imaging radar with a capability of image transmission/broadcasting, etc. Our work is aimed to provide a proof-of-concept simulation scenario to explore numerous aspects of UWB-OFDM radar imaging through evaluating range and cross-range imaging performance of such a system with an eventual goal of software-defined radio (SDR) implementation. Stripmap SAR topology was chosen for modeling purposes. Range/cross-range profiles were obtained along with full 2-D images for multi-target in noise scenarios. Model set-up and results of UWB-OFDM radar imaging simulation study using Matlab/Simulink modeling are presented and discussed in this paper.

The purpose of this project is threefold: To determine the optimum range-Doppler characteristics of an FM broadcast, to characterize a typical FM broadcast signal in terms of the properties of the ambiguity function, and to develop a model for the FM broadcast signal as a band-limited multi-tone FM stochastic process. The way to achieve these objectives is to analyze the response of a matched filter at the receiver by considering two modulating input signals. The inputs are a band-limited white Gaussian noise and a segment of a typical broadcast. The analysis of these signals is performed in the range-Doppler domain by observing the behavior of the ambiguity function. The results of the analysis show that the band-limited white Gaussian noise yields an optimum ambiguity function with a narrow mainlobe and low sidelobes evenly distributed over the range-Doppler plane. In contrast, the ambiguity function of a typical broadcast will exhibit a wider mainlobe and higher sidelobes. By adjusting parameters of the multi-tone FM stochastic process, a good match to the observed characteristics of an actual FM broadcast can be obtained. Thus, the signal model can be used to estimate the limitations in the detection of moving targets by means of a radar system that exploits FM signals of opportunity.

Ultra-wideband radar leads to optimum slant range resolution in radar imaging. There is a major problem in receiving short duration ultra-wideband radar signals and preserving their wave forms, because the conventional radar receivers are super heterodyne receivers. The present paper will show how optical fiber recirculation loop based interferoceiver will resolve the above problem.

Various approaches exist to enable target classification through a decomposition of the polarimetric scattering matrix. Specifically, the Euler decomposition attempts to express the target scattering properties through more physically relevant parameters. Target classification in general has been limited by signature variability and the saturation of images by non-persistent scatterers.¹ The Euler decomposition is sensitive to additional parameter ambiguities.² It will be demonstrated how undesirable ambiguities may be identified and mitigated. Through the analysis of polarimetric ISAR signatures obtained in compact radar ranges at the University of Massachusetts Lowell Submillimeter Technology Laboratory (STL)^3,4,5,6 and the U.S. Army National Ground Intelligence Center (NGIC), the cause of non-persistent scatters will be investigated. A proper characterization of non-persistence should lead to better optimization of the Euler decomposition, and thus improve target classification.

See Through The Wall (STTW) radar applications have become of high importance to Homeland Security and Defense needs. In this work surface penetrating radar is simulated using basic physical principles of radar propagation and polarimetric scattering. Wavenumber migration imaging is applied to simulated radar data to produce polarimetric imagery. A detection algorithm is used to identify dihedral scattering signatures for mapping inner building walls. The detector utilizes two polarimetric channels: HH and VV to classify objects as outer wall, inner wall, or object within room. The final product is a data generated building model that maps the interior walls of the building.

Our work on the solution to the Rayleigh problem has concentrated on developing the theoretical and mathematical aspects to determining the probability density function or characteristic function for superpositions of random phases. In this paper, we apply the methodology to solve a number of practical problems of interest to radar engineers. The determination of the proper design of a matched filter for phase noise for an arbitrary noise distribution for the phase noise is solved in this paper under very general circumstances. This solution can also be applied to determining the a detector for monopulse phase noise.

Advances in commercial off-the-shelf (COTS) embedded computing technologies have yielded impressive gains in computational throughput over the past 5-10 years. Adaptive sensor array systems utilizing real-time teraflop class machines are in wide deployment today. Gains in processor density have generally been achieved by steady improvements in semiconductor optical lithographic processes along with less frequent innovations in processor chip architectures. It is likely that the real-time embedded community is entering an era where processor architectural innovations will be bearing most of the burden for producing processing density gains as lithographic processes approach fundamental physical limitations. More and more 'systems on a chip' are emerging to address these trends. An exciting example of this technology trend is IBM's new Cell Broadband Engine (CBE) architecture which offers massive SIMD compute power on multiple computational units interconnected via a high bandwidth internal fabric. This paper explores the application of a computationally intensive adaptive nulling problem on the CBE architecture.

In previous work, we proposed to generate random samples using chaotic maps. More specifically, we demonstrated that Gaussian samples can be obtained via two random number generators that utilize first order or second order chaotic maps. In this paper we extend this work and propose to utilize the chaos-based random generators to develop a Gaussian FM signal that can be exploited for radar imaging. For this purpose, we fine-tune the chaotic map parameters of the Gaussian FM signal until we obtain a white wide-band spectrum, which is computed as an ensemble average, and analyze the corresponding ambiguity surface of the signal. We observe that the ensemble average of the ambiguity surface approaches an ideal two dimensional delta with uniformly distributed sidelobes on the range-Doppler plane. On average the sidelobes of the surface have intensity inversely proportional to the length of the processed echo. For completeness, we compare the variance of the Gaussian FM ambiguity function to that of a random binary phase code with comparable bandwidth. Furthermore, we show through simulations that Fourier processing of the Gaussian FM signal can yield a high-resolution range-Doppler image aircraft with prominent point scattering points over a substantial SNR dynamic range.

Standard radar systems commonly use a first-order phase compensation to account for the Doppler effect. As target speed increases, higher-order phase terms are needed to compensate a large time-bandwidth product signal. For extremely high speeds, a relativistic scheme based on the Lorentz Transformation of the wave 4-vector is more desirable since it provides correct results. Since the echo location problem involves transmission of the signal from a stationary frame and signal reflection from a moving frame, the wave 4-vector must be transformed twice to simulate a round trip. We show that for relative motion in one direction, the round-trip Lorentz transformation is equivalent to compressing the instantaneous frequency of the signal. The frequency compression factor is a nonlinear function of speed v. The nonlinearity is apparent only for relativistic speeds. In this paper, we analyze the ambiguity surface of the linear FM (chirp) signal to compare first-order and relativistic (full) compensation, and demonstrate that at relativistic speeds the ambiguity surface of the fully compensated linear FM compensated shows a linear delay-Doppler coupling.

The uniqueness of the Bernoulli frequency modulated signal, and other chaos-based FM signals, can be exploited to improve the performance of the Synthetic Aperture Radar systems. Recent work suggests that the Bernoulli map has an unusual behavior compared to other one dimensional discrete maps, such as Logistic or Tent maps. Additional work indicates that the sum of consecutive Bernoulli samples is generally non-Gaussian, except when the map parameters A= 0.5 and B = 1.8. This motivates us to analyze the behavioral differences of the maps for various parameters using the Lyapunov exponent, pseudo-phase spatial trajectory and neighbor samples correlation. Specifically, the correlation of Bernoulli samples is analyzed in terms of the probability density function which is derived from experimental data. Some of statistical tools used include the Forbenius-Perron Operator, and the correlation properties of chaotic sequences. In addition, other measurements of chaos derived from nonlinear dynamical modeling will be used such as: the Lyapunov exponent and the bifurcation diagram. Results show differences between the calculated features; for example, the Lyapunov exponent is bigger for Bernoulli FM than Logistic or Tent FM. In summary, we determined that Bernoulli FM is more chaotic than Logistic or Tent FM. We have also found another singularity in the correlation of sequence samples for the Bernoulli map.

Distributed airborne sensor geometries are considered that are comprised of multiple radar/comm transmit and receive nodes. Under this distributed robotic sensor concept, each of these radar transmit/receive nodes position-adaptively converge to the vicinity of a signal leakage point. A number of signal leakage point geometries are investigated that conform to geometries for typical building-type structures. The results include a set of electromagnetic computations that simulate the signal interaction and signal propagation between multiple leakage points. 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 scattered from embedded objects within the indoor environment back to the simulated sensor-nodes in the outdoor environment, are presented. Interpretations of these results are included from a signal analysis perspective. These results also include approximate preliminary systems-type calculations with regard to this distributed position-adaptive UAV radar system concept. Potential applications are outdoor-to-indoor detection of objects-of-interest that are within a building via implementation of a intelligent multi-static sensor network.

AVTIS is a compact portable remote sensing instrument originally designed for ground based surveying of active volcanic lava domes. Its primary mode of operation is active, using a monostatic 200mW 94GHz FMCW radar. The 94GHz signal is provided via a multiplied 7GHz source. Careful choice of a low-noise, highly linear 7GHz source has extended the range of the radar to at least 7km whilst retaining a range resolution of 1m. We will present results showing the range resolution and discrimination of separated targets for both natural topography at far ranges (>1km) and man made targets at close ranges (<1km). For close range imaging, the signal bandwidth can be increased to improve the range resolution allowing finer quality imagery to be retrieved.

S-band, be-polarization, combined, short pulse (~25ns) scatterometer-radiometer system is described, for water surface, bare soil, land snow and vegetation covers short range (~5m) remote sensing.

Sandia National Laboratories has teamed with General Atomics and Sierra Monolithics to develop the Athena tag for the Army's Radar Tag Engagement (RaTE) program. The radar-responsive Athena tag can be used for Blue Force tracking and Combat Identification (CID) as well as data collection, identification, and geolocation applications. The Athena tag is small (~4.5" x 2.4" x 4.2"), battery-powered, and has an integral antenna. Once remotely activated by a Synthetic Aperture Radar (SAR) or Moving Target Indicator (MTI) radar, the tag transponds modulated pulses to the radar at a low transmit power. The Athena tag can operate Ku-band and X-band airborne SAR and MTI radars. This paper presents results from current tag development testing activities. Topics covered include recent field tests results from the AN/APY-8 Lynx, F16/APG-66, and F15E/APG-63 V(1) radars and other Fire Control radars. Results show that the Athena tag successfully works with multiple radar platforms, in multiple radar modes, and for multiple applications. Radar-responsive tags such as Athena have numerous applications in military and government arenas. Military applications include battlefield situational awareness, combat identification, targeting, personnel recovery, and unattended ground sensors. Government applications exist in nonproliferation, counter-drug, search-and-rescue, and land-mapping activities.

Micro-Doppler is generated from targets with simple harmonic motions, characterized by a sinusoidal instantaneous frequency in the time-frequency plane. This type of micro-Doppler arises from vibrating or rotating targets, which are commonly present in indoor settings. It is shown that the use of basis functions matched to the sinusoidal micro-Doppler signatures proves effective in identifying the micro-Doppler components in indoor imaging. These functions are optimum in the maximum likelihood (ML) sense. Asymptotic properties of the proposed linear decomposition are derived. The basis decomposition provides enhanced phase and frequency resolutions and is robust to noise. It is strongly dependent on the Bessel function (of the first kind) characteristics. Simulation results are presented to demonstrate the effect of non-orthogonality of the basis functions and the respective frequency and phase resolution properties of the decomposition.

The Microwave Sensors Branch of the Army Research Laboratory (ARL) recently evaluated the potential of a commercially available borehole radar system for an underground target detection application. We used this ground-penetrating system, which is capable of operation at either 100 or 250 MHz, to conduct experiments at a locally constructed test site. Since the site's soil characteristics would severely impact conclusions drawn from the collected data, we also obtained and analyzed soil samples in order to determine the electrical properties of the earth in the vicinity of the boreholes. In addition, we modeled and then built a canonical target, using this canonical target as an input to electromagnetic simulations. The outputs from these simulations guided us in the analysis and interpretation of the collected radar data. In this paper, we present a description of both the data collection itself and the results of a posteriori analysis of the collected data. We begin by describing the test site along with the procedures that we followed when conducting the experiments. Next, we present a soil analysis and the expected target radar cross section (RCS) obtained from the electromagnetic modeling simulations. We then discuss the implications of these results for system performance. Finally, we present an analysis of real data from the collection and compare it to what we expect based on the soil analysis and the output of the electromagnetic models. Collectively, these analyses provide an indication of the borehole radar's true potential for detecting underground targets.

The ability to make radar signature databases portable for use within similar sensor systems is critical to the affordability of airborne signature exploitation systems. The capability to hybridize measured and synthetic signature database components as well as integrate signature components from different radar sensors will maximize the investment required to build complex radar signature databases. Radar target scattering response signatures are analyzed as random processes. Radar signature analysis techniques using information theory are introduced. Methods for using the target scattering response signal subspace within limited aspect viewing sectors are developed.

The goal of our research is to assess the capability of ultra-wide-band (UWB) radar for detection of roadside improvised explosive devices (IEDs). Radar scattering signatures of artillery shells over a broadband frequency range, with different Tx/Rx polarizations, and at various aspect angles have been explored based on simulation and indoor measurement. Characteristics of IEDs versus clutter, wave penetration at different frequencies are also investigated. Finally, visibility of IED targets is tested on a moving cart in outdoor settings, with IED targets on ground surface, recessed, and buried underground at different distances away from the radar.

One of the challenges of using synthetic aperture radar (SAR) to detect a man behind a wall is determining the amount of signal attenuation introduced by the signal's propagation through the wall. This attenuation is difficult to determine because the thickness and the electromagnetic properties of the wall are normally not known a priori. We describe a procedure for estimating the relative permittivity, conductivity, and thickness of the wall that minimize the error between physics-based predicted values of wall return and the corresponding values of the SAR image. The accuracy of the prediction is a function of the resolution of the SAR image relative to the thickness of the wall-the SAR image must have sufficient resolution such that the locations of the front and rear surfaces of a uniform wall can be estimated from the SAR image. The signal level behind the wall, or equivalently the signal attenuation by the wall is then determined from the estimates of the thickness and electromagnetic parameters. We demonstrate the effectiveness of this identification procedure using data generated by XPATCH simulations of three different wall materials.

Traditional approaches for locating and characterizing contaminated sites rely on invasive techniques which require drilling, testing, and sampling. These techniques provide the most direct access to the subsurface, but they are generally expensive and only provide measurements at points in a three dimensional surface. Furthermore, invasive techniques in polluted areas can promote further spread of contaminants. Development of non-invasive techniques that offer rapid and relatively inexpensive characterization is, therefore, necessary to detect and monitor plumes and sources of contaminants. Non-invasive techniques are also required for locating buried objects, such as landmines and unexploded ordnances. The use of cross well radar (CWR) as a non-invasive technique that has proven to be a reliable technology for detection of target objects that exhibit significant contrast of dielectric properties in saturated soils. Its application to detection of heterogeneously distributed phases in unsaturated soils under variable flow conditions has yet to be developed. This paper addresses the development of 2D flow and electromagnetic (EM) soilBed setup to further assess and enhance CWR technology for the detection of Dense Non-Aqueous Phase Liquids (DNAPLs) contamination and other target elements in variably-saturated soils subjected to transient flow conditions. Loop antennas have been developed and tested for this purpose. Transmission and reflection measurements are evaluated to determine the antenna's reliability and optimize their performance in the 2D electromagnetic field. The measurements indicate that a 2D EM boundary condition may be imposed by placing two parallel perfectly-reflecting metal plates along one of the dimensions of the soilBed setup. Transmission and reflection characteristics of the antennas vary with their method of construction. Results show a reliable and reproducible response from the loop antennas, but suggest some wave leakage and indicate that their design must be optimized. Measured variations in the transmission, reflection and impedance in the presence and absence of a buried object suggest that the 2D EM soilBed setup using loop antennas can be aplied to detect target elements in subsurface environments subjected to flow conditions. Future work addresses the assessment of CWR technology as a non-invasive method for detection and monitoring of heterogeneously-distributed target objects in subsurface environments.