We propose an airborne ground penetration radar that detects small buried objects. Earth electrical characteristics are discussed. Radar frequency is considered to penetrate 30 cm into typical ground and bandwidth is selected to achieve 5 cm range resolution in typical ground. A tunable free-electron maser allows adjustment to cope with earth variability. Frequency is selected to provide narrow enough beams so that clutter no longer dominates. The effects of clutter is reduced by beamforming with an array along the wing and by using a synthetic aperture antenna in the flight direction. The wiggler of the free-electron maser is modified to operate at low enough frequencies to provide adequate earth penetration. Pulse shaping or signal chirp provides the bandwidth at the frequency selected. We make an approximate prediction of signal to noise to show feasibility. Finally we discuss post processing to distinguish objects of interest from clutter.

This paper describes a study of the operation of a long range CWFM radar using "System View" software for modeling and simulation. The System View software is currently offered by Agilent. The models that were studied include: a model illustrating the basic principle of operation of the CWFM radar, the range resolution of the radar, the effect of long range processing and the resultant approach with the tradeoff of detected range resolution due to Doppler frequency shift as a function of range distance. The study was performed as part of the design of an airborne CWFM radar. The radar can be designed with a single antenna or a dual antenna. The dual antenna approach is presented in this paper.

The U.S. Army Research Laboratory (ARL) has been investigating the utility of ultra-wideband (UWB) synthetic aperture radar (SAR) technology for detecting concealed targets in various applications. We have designed and built a vehicle-based, low-frequency UWB SAR radar for proof-of-concept demonstration in detecting obstacles for autonomous navigation, detecting concealed targets (mines, etc.), and mapping internal building structures to locate enemy activity. Although the low-frequency UWB radar technology offers valuable information to complement other technologies due to its penetration capability, it is very difficult to comprehend the radar imagery and correlate the detection list from the radar with the objects in the real world. Using augmented reality (AR) technology, we can superimpose the information from the radar onto the video image of the real world in real-time. Using this, Soldiers would view the environment and the superimposed graphics (SAR imagery, detection locations, digital map, etc.) via a standard display or a head-mounted display. The superimposed information would be constantly changed and adjusted for every perspective and movement of the user. ARL has been collaborating with ITT Industries to implement an AR system that integrates the video data captured from the real world and the information from the UWB radar. ARL conducted an experiment and demonstrated the real-time geo-registration of the two independent data streams. The integration of the AR sub-system into the radar system is underway. This paper presents the integration of the AR and SAR systems. It shows results that include the real-time embedding of the SAR imagery and other information into the video data stream.

Modern radars can pick up target motions other than just the principle target Doppler; they pick out the small micro-Doppler variations as well. These can be used to visually identify both the target type as well as the target activity. We model and measure some of the micro-Doppler motions that are amenable to polarimetric measurement. Understanding the capabilities and limitations of radar systems that utilize micro-Doppler to measure human characteristics is important for improving the effectiveness of these systems at securing areas. In security applications one would like to observe humans unobtrusively and without privacy issues, which make radar an effective approach. In this paper we focus on the characteristics of radar systems designed for the estimation of human motion for the determination of whether someone is loaded. Radar can be used to measure the direction, distance, and radial velocity of a walking person as a function of time. Detailed radar processing can reveal more characteristics of the walking human. The parts of the human body do not move with constant radial velocity; the small micro-Doppler signatures are timevarying and therefore analysis techniques can be used to obtain more characteristics. Looking for modulations of the radar return from arms, legs, and even body sway are being assessed by researchers. We analyze these techniques and focus on the improved performance that fully polarimetric radar techniques can add. We perform simulations and fully polarimetric measurements of the varying micro-Doppler signatures of humans as a function of elevation angle and azimuthal angle in order to try to optimize this type of system for the detection of arm motion, especially for the determination of whether someone is carrying something in their arms. The arm is often bent at the elbow, providing a surface similar to a dihedral. This is distinct from the more planar surfaces of the body and allows us to separate the signals from the arm (and knee) motion from the rest of the body. The double-bounce can be measured in polarimetric radar data by measuring the phase difference between HH and VV. Additionally, the cross-pol and co-pol Doppler signatures are analyzed, showing that the HH polarization may perform better on dismounts in open grass.

In this paper we investigate the impact of polarization dynamics in interference analysis in urban areas; furthermore we connect the polarization dynamics with different scattering environments and frequencies which introduces more flexibility for diversity schemes and the implementation of concepts such as polarization MiMO etc. The key idea is that once a non-monochromatic wave impinges upon buildings, the spectral components of the wave are variably depolarized upon scattering. In wideband signals we show that polarization dynamics are different in various frequencies and different environments and that introduces another degree of freedom to reduce interference, to add diversity and therefore improve capacity.

We have previously reported on the analysis of fully polarimetric single look and multilook SIR-C data. We have reported that the Stokes(Kennaugh) matrices for each pixel have one and only one eigenvector that satisfies the property of a Stokes Vector. We now report on new analysis of fully polarimetric SIR-C data and ISAR data from the Submillimeter-Wave Technology Laboratory at the University of Massachussetts Lowell which shows that the remaining three eigenvectors of the Stokes matrix are quaternions which represent rotations. Furthermore, the three direction vectors of these quaternions form an orthogonal cartesian set of axes. We also discuss relationships between the angles of the Stokes Vector with the Euler parameters initially proposed by Huynen.

A time-integrated range-Doppler map shows the micro-Doppler characteristics of targets in radar images that enable an operator to classify different target types and to classify different activities being done by the targets. A time-integrated range-Doppler map is a compilation of range-Doppler maps over time that results in a spectrogram-like characterization of Doppler while maintaining the range information as well. These are compiled from the range-Doppler maps by taking the maximum value for each pixel over a time range. The time resolution is overlapped onto the range resolution, which is in effect a rotation of the traditional spectrogram which compresses range. This type of radar imaging also allows multiple subjects to be viewed simultaneously and avoids tracking issues in spectrogram creation. The display of range-Doppler movies or spectrograms with range extent is also demonstrated.

Alternative approaches and models continue to be investigated and evolved to correctly locate and identify a buried mine with minimum risk. Though microwave remote sensing based detection of shallow buried landmines provides such a risk free alternative, it is a highly complex and computationally intensive task involving several parameters. The present paper deals with the use of data obtained in multiple polarizations and their transforms approximating rough surface conditions in sand for landmine detection. Data in both HH and VV polarizations in microwave X-band frequency (10 GHz, 3cm) was generated using a live landmine (with explosives less fuze) for the present study under field conditions. Various transforms such as image differencing, image ratioing and polarization discriminant ratio (PDR) were studied for its effect on landmine detection. However, it was found that most of the clutter and noise gets suppressed on using a transform obtained by subtracting the difference of data in two polarizations from its sum. The surface roughness conditions have been approximated as available in western parts of India and which are suitable for application of microwave radar remote sensing for detection of minefields. With the advent of satellites providing data in various polarizations, it has now become relevant to investigate methods which can be used for landmine detection using polarization techniques. The proposed analysis is expected to be useful in future in detection of landmines using multi-polarization satellite data in microwave X-band in deserts such as those existing in the western borders of India.

Knowing the statistical characteristics of a target's radar cross-section (RCS) is crucial to the success of radar target detection algorithms. A wide range of applications currently exist for dismount (i.e. human body) detection and monitoring using ground-moving target indication (GMTI) radar systems. Dismounts are particularly challenging to detect. Their RCS is orders of magnitude lower than traditional GMTI targets, such as vehicles. Their velocity of about 0 to 1.5 m/s is also much slower than vehicular targets. Studies regarding the statistical nature of the RCS of dismounts focus primarily on simulations or very limited empirical data at specific frequencies. This paper seeks to enhance the existing body of work on dismount RCS statistics at Ku-band, which is currently lacking, and has become an important band for such remote sensing applications. We examine the RCS probability distributions of different sized humans in various stances, across aspect and elevation angle, for horizontal (HH) and vertical (VV) transmit/receive polarizations, and at diverse resolutions, using experimental data collected at Ku-band. We further fit Swerling target models to the RCS distributions and suggest appropriate detection thresholds for dismounts in this band.

Knowing the statistical characteristics of a target's radar cross-section (RCS) is crucial to the success of radar target detection algorithms. Open literature studies regarding the statistical nature of the RCS of ground vehicles focus primarily on simulations, scale model chamber measurements, or limited experimental data analysis of specific vehicles at certain frequencies. This paper seeks to expand the existing body of work on ground vehicle RCS statistics at Ku-band for ground moving target indication (GMTI) applications. We examine the RCS probability distributions of civilian and military vehicles, across aspect and elevation angle, for HH and VV polarizations, and at diverse resolutions, using experimental data collected at Ku-band. We further fit Swerling target models to the distributions and suggest appropriate detection thresholds for ground vehicles in this band.

The automatic identification of human activities has become an area of interest in recent years. Identifying human activities is useful in various applications, such as through-barrier identification of intruders and non-contact monitoring of patients in hospitals. Numerous methods of human activity classification have been proposed in the past, including the use of Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs). Most research in this area thus far has utilized the Short-Time Fourier Transform (STFT) as a method of obtaining the feature vectors necessary for classification. In this paper, we propose the use of the Empirical Mode Decomposition (EMD) algorithm as an alternative approach for obtaining feature vectors from human micro-Doppler signals and utilize an SVM for classification. Since the micro-Doppler signature is unique to a specific activity, the EMD outputs can be utilized as feature vectors. By utilizing the EMD algorithm in conjunction with an SVM, binary classification of human activities have shown to yield accurate results. Because SVMs were originally developed to solve the binary classification problem, additional steps must be taken in order to extend the problem to identify multiple classes. In this paper, two methods for multi-class classification will be demonstrated and compared. The first method is the one-against-all approach and the second is a decision tree based approach. In both cases, a high degree of accuracy is achieved.

In this paper, we summarize our efforts of using three different radars (impulse radar, swept frequency radar, and continuous-wave radar) for through-the-wall sensing. The purpose is to understand the pros and cons of each of the three radars. Through extensive experiments, it was found that the radars are complementary and multiple radars are needed for different scenarios of through-the-wall target detection and tracking.

High-resolution through-the-wall radar imaging (TWRI) systems can provide a high degree of situational awareness in urban sensing applications. However, such systems generate huge amounts of data, owing to the use of wideband signals and large arrays to achieve high resolutions in range and crossrange. This makes both data acquisition and processing challenging. In this paper, we present fast data acquisition and processing schemes for TWRI. We use compressive sensing and novel concepts of microwave tomography to establish a reduced-redundancy spatial and frequency measurement configuration, which provides clear advantages in terms of measurement time and algorithm complexity. Performance validation of the proposed strategy is provided using laboratory experiments.

In urban sensing and through-the-wall radar, the existence of targets in proximity to walls or buildings results in multipath returns. In this paper, we exploit the multipath from the walls to achieve target localization with a single sensor. We deal with sparse scenes of single targets. A time-of-arrival wall association algorithm is derived to relate target multipath returns to the respective walls, followed by a nonlinear least squares optimization to determine the target location. Simulated and experimental data are used to validate the proposed algorithms.

For through-the-wall radar imaging (TWRI), an accurate characterization of the wall is important for the enhancement of imaging of the target behind the wall. In this paper we cast the two-dimensional (2D) wall interior structure imaging as a subsurface imaging problem. The region between the front and back walls is imaged using a novel linear inverse scattering algorithm for 2D subsurface imaging. The imaging algorithm is based on first order Born approximation and exploiting halfspace Green's function. The exploding reflection model is employed and then the Green's function is expanded in the spectral domain to formulate a novel real time intra-wall imaging algorithm. The linearization of the inversion scheme and the employment of FFT/IFFT in the imaging formula make the imaging algorithm suitable in several applications concerning the diagnostics of large probed domain and allow real time processing. A numerical result is presented to show the effectiveness and efficiency of the proposed algorithm for real time intra-wall characterization.

Analytical expressions for Fresnel reflection and transmission coefficients have been extensively used in ray-tracing simulation. Although these tools accurately predict the field for simple homogeneous wall structures, it is difficult, if not impossible, to extend such an analysis to find reflection and transmission coefficients for walls composed of dielectric and imperfectly conducting materials or complex, inhomogeneous structures. In principle, Fresnel theory is considered a high-frequency method, but in practical problems (such as walls with metallic rebars and similar applications), transmission does not monotonically decrease with incidence angle, and Fresnel theory does not apply. In this paper, we use the FDTD method to extend the theoretical Fresnel formulation to certain types of problems where Fresnel theory does not apply. We find that the presence of rebar affects transmission characteristics much more significantly than permittivity or wall depth. We initially verify the FDTD method with simple theoretical applications, and then we go further in more complicated cases; we furthermore extend our analysis to polarization effects that occur from such inhomogeneities.

Novel antennas exhibiting directivity enhancement by using a short focal length plano-concave lens engineered by stacked subwavelength hole arrays in such a way that an effective negative index of refraction is obtained. An additional unexpected property of this design is that it opens the possibility to achieve an index close to zero, n → 0, arisen from ε- and μ-near-zero extreme values. Our original design works with evanescent modes in comparison with the well known classical metallic lenses operating with propagating modes. In our case, this leads to a negative index of refraction, whereas metallic lenses exhibit a positive but less than one index of refraction. It is demonstrated by means of a simple design based on dispersion diagram and ray tracing an easy and correct method for rather accurate results. Also, an optimization of the hole diameter or longitudinal lattice constant to achieve not only n = -1, but also free space matching is possible simultaneously. A power enhancement up to 24 dB with cross-polarization below -30 dB with regards to co-polar, when the lens is applied as antenna radiation beamforming has been measured. For the case of index close to zero, n → 0, the power enhancement is 27 dB whereas the cross-polarization remains -17 dB with regards to co-polar. New improvements are under analysis in order to determine if this technology could be competitive with current state of the art of waveguide lenses and Fresnel zone plate lenses.

The solutions to the Rotman lens design equations constrain the minimum size of the device. Here we use Transformation Optics to compress a transmission line based Rotman lens by 27 percent along the optical axis while maintaining the beam steering range, gain and side lobe amplitudes over the full frequency range of the original lens. The transformation applied requires an anisotropic magnetic response, which is achieved in the transmission line context using complementary electric dipole structures patterned into the top conductor of the lens. The non-resonant complementary metamaterial elements provide an anisotropic, eective magnetic permeability with values that can be varied across a spatial region by varying the geometry of each element.

This paper discusses the general concept of using metamaterials in microwave lenses. The different optics afforded by the inclusion of metamaterials in the lens structure produce new features such as reduced size and new beam formations. The use of negative refractive index materials is discussed in reference to the original concept of the perfect lens, leading to the Rotman lens and the Luneburg lens. In Rotman lens, negative refractions help reducing the lens size and a broadband electromagnetic band gap (EBG) surface is used to prevent reflections off the sidewalls. Verification of negative refraction and simulation of isotropic material performance are presented, as well as an example of broadening the band of an EBG surface.

The sense and avoidance (SAA) and due-regard radar systems have strict requirements on size, weight and power (SWaP) and target localization accuracies. Also, the multi-mission capabilities with both weather and hard targets are critical to the survivability of unmanned aerial vehicles (UAV) in the next generation national airspace. The aperture limitations of the aircraft sensor installation, however, have prevented large antennas/arrays to be used. The tradeoffs among frequencies, resolutions and detection range/accuracies have not been fully addressed. Innovative concepts of overcoming the aperture limitation by using a special type of super-resolution technology are introduced. The first technique is based on a combination of thinned antenna array, an extension to the traditional Multiple Signal Classification (MUSIC) technique, and applying a two-dimensional sidelobe mitigation technique. To overcome the degradation of MUSIC-type of approach due to coherent radar signals, a special waveform optimization procedure is used. The techniques for mitigating artifacts due to "thinned" array are also introduced. Simulated results of super-resolution techniques are discussed and evaluated, and the capability of separating multiple targets within aperture-constrained beamwidth is demonstrated. Moreover, the potential capabilities of autonomous weather hazard avoidance are also analyzed.

In synthetic-aperture radar (SAR), ground-target vibrations introduce a phase modulation in the returned signals, a phenomenon often referred to as the micro-Doppler effect. Earlier work has shown that the problem of estimating common ground-target vibrations can be transformed into the problem of successively estimating chirp parameters of the returned signal in properly sized subapertures. Recently, a method based on the discrete fractional Fourier transform (DFRFT) was proposed, in conjunction with the subaperture framework, to estimate target vibrations in the absence of noise. In this paper a pseudo-subspace approach is employed to extend the applicability of the DFRFT-based vibration-estimation method to signals that are corrupted by white noise. The new algorithm first calculates the inverse discrete Fourier transform of row and column projections of the magnitude of the DFRFT spectrum of the SAR returned signal to obtain two vectors. Next, covariance matrices are estimated from the sample covariance matrices of the two vectors. A pseudo-subspace approach is then applied to the covariance matrices to yield the pseudo-spectra. The chirp rate of the signal is estimated by finding the principle frequency component in the corresponding pseudo-spectrum. Monte-Carlo simulations demonstrate that the proposed method generally offers improved mean-square-error performance in the presence of noise compared to the direct DFRFT-based method.

In previous work, we constructed wideband FM signals for high range resolution applications using the non-linear Lorenz system, which has a set of three state variables and three control parameters. The FM signals were generated using any one of the three state variables as the instantaneous frequency which was then controlled by adjusting the values of the parameters in the chaotic regime. We now determine the spectral characteristics of the Lorenz FM signal and compare the spectral characteristics to those of a similar FM signal based on the Lang-Kobayashi system. We show that for either chaotic system, the local linearity of the attractor yields an FM signal with a distinct chirp behavior. Irrespective of the statistical independence of the chaotic flow samples, we show that the chaotic FM signal follows Woodward's theorem in the sense that the spectrum of the FM signal follows the shape of the probability density function of the state variable. The chirp rate of the FM signal can be controlled through a time-scale parameter that compresses or expands the chaotic flow. As the chaotic flow evolves in time, so does the spectrum of the corresponding FM signal, which experiences changes in center frequency and bandwidth. We show that segments of the signal with a high chirp rate can be significantly compressed to achieve high range-Doppler resolution. The ability to change the center frequency and the shape of the spectrum is interpreted as added frequency agility.

This paper provides a summary of recent results on a novel multi-platform RF emitter localization technique denoted as Position-Adaptive RF Direction Finding (PADF). This basic PADF formulation is based on the investigation of iterative path-loss based (i.e. path loss exponent) metrics estimates that are measured across multiple platforms in order to robotically/intelligently adapt (i.e. self-adjust) the location of each distributed/cooperative platform. Recent results at the AFRL indicate that this position-adaptive approach shows potential for accurate emitter localization in challenging embedded multipath environments (i.e., urban environments). As part of a general introductory discussion on PADF techniques, this paper provides a summary of our recent results on PADF and includes a discussion on the underlying and enabling concepts that provide potential enhancements in RF localization accuracy in challenging environments. Also, an outline of recent results that incorporate sample approaches to real-time multi-platform data pruning is included as part of a discussion on potential approaches to refining a basic PADF technique in order to integrate and perform distributed self-sensitivity and self-consistency analysis as part of a PADF technique with distributed robotic/intelligent features. The focus of this paper is on the experimental performance analysis of hardware-simulated PADF environments that generate multiple simultaneous mode-adaptive scattering trends. We cite approaches to addressing PADF localization performance challenges in these multi-modal complex laboratory simulated environments via providing analysis of our multimodal experiment design together with analysis of the resulting hardware-simulated PADF data.

The clutter locus is an important concept in space-time adaptive processing (STAP) for ground moving target indicator (GMTI) radar systems. The clutter locus defines the expected ground clutter location in the angle-Doppler domain. Typically in literature, the clutter locus is presented as a line, or even a set of ellipsoids, under certain assumptions about the geometry of the array. Most often, the array is assumed to be in the horizontal plane containing the velocity vector. This paper will give a more general 3-dimensional interpretation of the clutter locus for a general linear array orientation.

A new scheme for sampling and detecting as well as reconstructing analog signals residing in a wide spectrum band through compressive sensing is proposed. By applying compressive sensing techniques, this scheme is able to detect signals quickly over very wide bandwidth. Unlike existing compressive sensing approaches which carry out the sampling and detection/reconstruction procedures separately, in this scheme, the sampling process and the detection/reconstruction process has close interaction. Once a new signal is detected, resources are allocated for further processing. Previously detected signals are then removed as interference to facilitate new signal detection. Hence, this scheme can quickly adapt to the variation of the environment. Moreover, it avoids the complex iterative optimization procedure in reconstruction and instead uses one step detection procedure to reaches real-time handling of signals. Simulation results show that it can closely track the spectrum change even when the signal is weak.

In this paper, we address an adaptive detection of range-spread targets or targets embedded in Gaussian noise with unknown covariance matrix by the generalized detector (GD) based on the generalized approach to signal processing (GASP) in noise. We assume that cells or secondary data that are free of signal components are available. Those secondary data are supposed to process either the same covariance matrix or the same structure of the covariance matrix of the cells under test. In this context, under designing GD we use a two-step procedure. The criteria lead to receivers ensuring the constant false alarm rate (CFAR) property with respect to unknown quantities. A thorough performance assessment of the proposed detection strategies highlights that the two-step design procedure of decision-making rule in accordance with GASP is to be preferred with respect to the plain one. In fact, the proposed design procedure leads to GD that achieves significant improvement in detection performance under several situation of practical interest. For estimation purposes, we resort to a set of secondary data. In addition to the classical homogeneous scenario, we consider the case wherein the power value of primary and secondary data vectors is not the same. The design of adaptive detection algorithms based on GASP in the case of mismatch is a problem of primary concern for radar applications. We demonstrate that two-step design procedure based on GASP ensures minimal loss.

A new design of a noise radar system is proposed with capabilities to measure and acquire the radar signature of various targets. The proposed system can cover a noise bandwidth of near DC to 30 GHz. The noise radar signature measurements were conducted for selective targets like spheres and carpenter squares with and without dielectric bodies for a noise band of 400MHz-3000MHz. The bandwidth of operation was limited by the multiplier and the antennae used. The measured results of the target signatures were verified with the simulation results.

Conventional digital signal processing scheme in noise radars has some limitations related to combination of high resolution and high dynamic range. Those limitations are caused by a tradeoff in performance of currently available ADCs: the faster is ADC the smaller is its depth (number of bits) available. Depth of the ADC determines relation between the smallest and highest observable signals and thus limits its dynamic range. In noise radar with conventional processing the sounding and reference signals are to be digitized at intermediate frequency band and to be processed digitally. The power spectrum bandwidth of noise signal which can be digitized with ADC depends on its sampling rate. The bandwidth of radar signal defines range resolution of any radar: the wider the spectrum the better the resolution. Actually this is the main bottleneck of high resolution Noise Radars: conventional processing doesn't enable to get both high range resolution and high dynamic range. In the paper we present a way to go around this drawback by changing signal processing ideology in noise radar. We present results of our consideration and design of high resolution Noise Radar which uses slow ADCs. The design is based upon generation of both probing and reference signals digitally and realization of their cross-correlation in an analog correlator. The output of the correlator is a narrowband signal that requires rather slow ADC to be sampled which nowadays may give up to 130 dB dynamic range.

Frequency spectrum responses of targets are of importance in UWB radar for target identification and recognition. As technology's digitization rate of analog sources increases, direct acquisition of wider bandwidths is becoming possible. Through conversion to the frequency domain, wider bandwidth spectral responses for targets can be produced. However, to directly digitize higher frequencies with UWB signals directly (i.e., ≥ 4 GHz), the technology is somewhat limited. This paper will present a technique which utilizes both hardware and software to produce a lower bandwidth signal (e.g., 1.5 GHz), which contains larger spectral bandwidth information (e.g., 6 GHz). The technique utilizes a double band folding methodology implemented in hardware, or software, to translate larger bandwidths into lower bandwidths for direct digitization. The generated lower bandwidth will have a unique spectral response containing the superimposed amplitudes of the larger bandwidth transmitted signal. This folded spectrum can then be used in applications such as target recognition and identification. Simulated and experimental results will be presented to evaluate the advantages and disadvantages of such an approach.

Correlation detection is an essential ingredient in noise radar. Such detection is achieved via coherent signal processing, which, conceivably, gives the best enhancement in the signal-to-noise ratio. Over the years, much research and progress has been made on the use of noise radar systems as means for effective through-wall detection. Information about a particular target's range and/or velocity are often acquired by comparing and analyzing both transmit and received waveforms. One of the widely used techniques employed to measure the degree of similarity between the two signals is correlation. The aforementioned methodology determines to what extent two waveforms match by multiplying and shifting one signal with respect to a time-lagged version of the second signal. This feature of correlation is very applicable to radar signals since a received signal from a target is delayed on the path of return to the receiving antenna. Transmission and reflection impairments will distort the propagating signals and degrade the correlation. Thus, it is essential that we try to study the effects that such degradations can have on the signals that will be used in the correlation process. This paper presents some concepts of a noise radar system, simulation studies, and an analysis of the results ascertained.

The Doppler spread pertaining to the ultrawideband (UWB) radar signals from moving target is directly proportional to the bandwidth of the transmitted signal and the target velocity. Using typical FFT-based methods, the estimation of true velocities pertaining to two targets moving with relatively close velocities within a radar range bin is problematic. In this paper, we extend the Multiple Signal Classification (MUSIC) algorithm to resolve targets moving velocities closer to each other within a given range bin for UWB random noise radar waveforms. Simulated and experimental results are compared for various target velocities using both narrowband (200MHz) and wideband (1GHz) noise radar signals, clearly establishing the unbiased and unambiguous velocity estimations using the MUSIC algorithm.

Noise excitation sources in radar systems have become increasingly useful in applications requiring wideband spectral responses and covertness. However, in applications requiring spectral controllability, traditional analog noise sources prove troublesome and require additional hardware such as sets of digital filters whose own spectral characteristics must also be accounted for. In an effort to reduce these issues and increase the applications of noise waveforms, a technique for generating a fully controllable pseudo-noise waveform is presented. This pseudo-noise waveform will be generated through the use of a multi-tone waveform. By randomizing the phase angles and setting the appropriate amplitudes to the individual tones, the result is a waveform whose temporal pattern resembles noise and frequency response is broadband. The capabilities of this digitally produced pseudo-noise multi-tone waveform is presented by optimization via a water-filling technique, thereby producing a flat spectral response for a user defined amplitude, effectively removing the spectral effects of the radar components. This optimized waveform is used to present methods for increasing signal to noise ratio (SNR) of cross-correlated responses of the waveform through the application of window functions to the waveform. As a whole, this paper showcases the ability to use this pseudo-noise multi-tone waveform for complete ultra-wideband (UWB) spectral control through water-filling and a method for increasing SNR of the cross correlated response of the transmitted and received radar waveform for a bandwidth of 2.5 GHz ranging from 2 to 4.5 GHz.

The electromagnetic (EM) waves propagating through causal, linear, and lossy dispersive media (soil, foliage, plasma, water, biological tissue, etc.), experience frequency-dependent attenuation and phase distortion. This has assumed significant importance for systems operating with ultrawideband (UWB) spectrum. This paper analyzes the dynamical evolution of UWB noise radar signals through dispersive media. The effects on the signal propagation due to the evolution of the Brillouin precursor through dispersive media are discussed. The evolving waveforms are then compared with the Brillouin precursor due to rectangular sine-modulated deterministic signals. The advantages of random noise waveforms through dispersive media are also discussed.

Ultra-wideband (UWB) excitation sources in radar systems have allowed for enhancement in capabilities such as target spectral response, clutter suppression, and range resolution. While generation of generic UWB signals has become easily achievable, direct acquisition, or digitization, of these bandwidths (≥ 4 GHz) is not. To account for this, many UWB radar systems implement a single or multi-stage band folding technique in the receiver hardware chain which allows for the direct digitization of the UWB waveform at a smaller bandwidth (e.g., 4 GHz into 1 GHz). While the lower bandwidth allows for larger than narrowband capabilities, it reduces desired features such as range resolution (e.g., 3.75 cm to 15 cm). In an effort to address this problem, and allow for utilization of full bandwidth of an UWB waveform, this paper presents a signal processing technique which utilizes hardware band folding to wrap a spectrally unique UWB multi-tone waveform into a lower frequency, lower bandwidth signal allowing for both direct digitization and conservation of UWB features. The signal processing technique utilizes the multi-tone waveform to generate an UWB signal composed of sections whose separate spectral peaks fold into the inner ΔF regions of the previous band. It will be shown, that through reassignment of these peaks, as well as the phase, to the individual frequencies, the intended UWB capabilities can be restored.

Synthetic aperture radar (SAR) systems are getting more and more applications in both civilian and military remote sensing missions. With the increasing deployment of electronic countermeasures (ECM) on modern battlefields, SAR encounters more and more interference jamming signals. The ECM jamming signals cause the SAR system to receive and process erroneous information which results in severe degradations in the output SAR images and/or formation of phony images of nonexistent targets. As a consequence, development of the electronic counter-countermeasures (ECCM) capability becomes one of the key problems in SAR system design. This paper develops radar signaling strategies and algorithms that enhance the ability of synthetic aperture radar to image targets under conditions of electronic jamming. The concept of SAR using chaotic carrier frequency agility pulses (CCFAP-SAR) is first proposed. Then the imaging procedure for CCFAP-SAR is discussed in detail. The ECCM performance of CCFAP-SAR for both depressive noise jamming and deceptive repeat jamming is analyzed. The impact of the carrier frequency agility range on the image quality of CCFAP-SAR is also studied. Simulation results demonstrate that, with adequate agility range of the carrier frequency, the proposed CCFAP-SAR performs as well as conventional radar with linear frequency modulation (LFM) waveform in image quality and slightly better in anti-noise depressive jamming; while performs very well in anti-deception jamming which cannot be rejected by LFM-SAR.

Noise waveforms generated using low cost diodes are a simple way for radars to transmit a wideband (> 4 GHz) multi-bit pseudorandom code for use in a cross correlation receiver. This type of waveform also has the advantage of being difficult to intercept and is less prone to interfere with adjacent systems. Radar designed to operate over this wide frequency range can take advantage of unique target Radar Cross Section (RCS) ripple versus frequency for objects of different materials and sizes. Specifically the periodicity and amplitude of the ripple is dependent on the shape and size of a target. Since background clutter does not display this variation, RCS variation determines whether a known target is present in a return. This paper will present the radar hardware and signal processing techniques used to maximize a target's unique spectral response against a cluttered background. The system operates CW over a 4-8 GHz bandwidth requiring the need to address issues regarding range resolution and far out undesired returns. Lessons learned from field observations and mitigation techniques incorporated in the system are included. This paper also deals with the signal processing technique used for detection, then discrimination. Detection thresholds are set and triggered by a simple correlation peak level. Discrimination involves inspection of the spectral return. A comparison performed in real time to a stored library value determines the presence of known objects. Measured data provided demonstrates the ability of the radar to discriminate multiple targets against multiple backgrounds.

A correlation processing algorithm in the spectral domain is proposed for detecting moving targets with random noise radar. AD converted reference and Rx signals are passed through FFT block, and they are multiplied after the reference signal is complex conjugated. Now inverse FFT yields the sub-correlation results, and range and velocity information can be accurately extracted by an additional FFT processing. In this design procedure, specific considerations have to be made for correlation length, averaging number, and number of sub-correlation data for Doppler processing. The proposed algorithm was verified by Simulink (Mathworks) simulation, and its logic was implemented with Xilinx FPGA device (Vertex5 series) by System Generator block sets (Xilinx) in the Simulink environment. A CW X-band random-FM noise radar prototype with an instantaneous bandwidth of 100 MHz was designed and implemented, and laboratory and field tests were conducted to detect moving targets, and the observed results showed the validity of the proposed algorithm and the operation of implemented FPGA logics.

We present a novel passive radar imaging method for moving targets using distributed apertures. We develop a passive measurement model that relates measurements at a given receiver to measurements at other receivers. We formulate the passive imaging problem as a Generalized likelihood ratio test (GLRT) for a hypothetical target located at an unknown position, moving with an unknown velocity. We design a linear discriminant functional by maximizing the signal-to-noise ratio (SNR) of the test-statistic, and use the resulting position- and velocity-resolved test-statistic to form an image of the scene of interest. We present numerical experiments to demonstrate the performance of our imaging method.

We describe a new approach to random-signal radar based on the recent discovery of analytically solvable chaotic oscillators. These surprising nonlinear systems generate random, aperiodic waveforms that offer an exact analytic representation, allowing the implementation of simple matched filters and coherent reception. Notably, this approach enables nearly optimal detection of noise-like waveforms without need for expensive variable delay lines to store wideband waveforms for correlation. Mathematically, the waveform is expressed as a linear convolution of a bit sequence with a fixed basis function. We realize a simple matched filter for the waveform using a linear filter whose impulse response function is the time reverse of the basis function. Importantly, linear filters matched to finite bit sequences can be defined, enabling pulse compression and spread spectrum radar. We present an example oscillator, its matched filter, and simulation results demonstrating the pulse compression radar concept.

One may describe the effect of a radar or sonar target on an incoming signal as a filter which produces a scattered signal. Chaotic signals are very sensitive to the effect of filters, so a radar or sonar target imposes a unique signature on a scattered chaotic signal. In this paper we describe a method that uses the concept of phase space dimension to create a reference from a scattered chaotic signal. This reference becomes part of a library, and comparing an unknown scattered signal to this library can reveal which target caused a particular scattered signal. Because we are not imaging the target, this method can use signal with low range resolution.

Conventional linear frequency modulation (LFM) synthetic aperture radar (SAR) is incapable of countering deceptive repeat jamming. In this paper, a new SAR signal based on chaotic coded orthogonal frequency division multiplexing (COFDM) is studied. The fact that chaotic codes are sensitive to the initial values allows generating a large number of different chaotic sequences to form SAR transmitting waveforms, where all the signal sequences are orthogonal to each other, enabling COFDM-SAR countering not only active noise but also deceptive repeat jamming. The procedures for COFDM waveform generation and SAR anti-jamming processing are discussed. Comparative studies of the electronic counter-countermeasure performance (ECCM) between COFDM-SAR and conventional LFM-SAR are made. Simulation results are presented to demonstrate the superior performance of COFDM-SAR in countering repeat deception as well as active noise jamming.

A principle of quadrature correlation detection of noise signals using an analog broadband microwave correlator is presented in the paper. Measurement results for the correlation function of noise signals are shown and application of such solution in the noise radar for precise determination of distance changes and velocity of these changes is also presented. Results for short range noise radar operation are presented both for static and moving objects. Experimental results using 2,6 - 3,6 GHz noise like waveform for the signal from a breathing human is presented. Conclusions and future plans for applications of presented detection technique in broadband noise radars bring the paper to an end.

This paper addresses the issue of interference suppression in noise radars. The proposed methods can be divided into non-parametric and parametric ones. The considered non-parametric methods are based on linear time-frequency (TF) tools, namely the short-time Fourier transform (STFT) and local polynomial Fourier transform (LPFT). The STFT is the simplest TF method, but, due to the resolution problem, it performs poorly with highly nonstationary interferences. The LPFT resolves the resolution problem, however at the cost of increased complexity. In parametric methods, the phase of interference is locally approximated by a polynomial, which is motivated by the Weierstrass's theorem. Using the phase approximation, the corrupted received signal is demodulated and successively filtered. Two methods for polynomial phase approximation are considered, the high-order ambiguity function (HAF) and product high-order ambiguity function (PHAF). The method based on the HAF is computationally efficient; however, it suffers from the identifiability problem when multicomponent signals are considered. The identifiability problem can be resolved using the PHAF.

The detection and identification of concealed weapons is an extremely hard problem due to the weak signature of the target buried within the much stronger signal from the human body. This paper furthers the automatic detection and identification of concealed weapons by proposing the use of an effective approach to obtain the resonant frequencies in a measurement. The technique, based on Matrix Pencil, a scheme for model based parameter estimation also provides amplitude information, hence providing a level of confidence in the results. Of specific interest is the fact that Matrix Pencil is based on a singular value decomposition, making the scheme robust against noise.

An approach toward real-time for radar through-the-wall (TTW) sensing is presented. The aim is to detect and classify human motions behind walls. In the future a system could support an operator with information on the activity in a closed room, e.g., in a hostage situation. To meet this objective, radar TTW measurements have been performed on moving person(s) inside a closed room and an MTI-based signal processing algorithm, using coherent subtraction between different frequency sweeps, has been developed. The radar was equipped with ridge horn antennas which were directed toward an outer door. For each measurement the radar frequency was swept between 5 and 10 GHz 1893 times, with a sweep sampling rate of ~94 Hz. A crucial algorithm parameter is the time distance between two subtracting sweeps, or the sweep difference. Fast and slow motions are captured by using separate sweep differences. Applying the algorithm to the TTW data, we find that human motions behind a door can be detected with the background well suppressed. Even a person who is standing still without breathing is fairly easy to detect. The results are promising with low false alarm rates and fast signal processing rates, enabling real-time operation capability.

A typical Ground Moving Target Indicator (GMTI) radar specification includes the parameters Probability of Detection (P_D) - typically on the order of 0.85, and False Alarm Rate (FAR) - typically on the order of 0.1 Hz. The PD is normally associated with a particular target 'size', such as Radar Cross Section (RCS) with perhaps some statistical description (e.g. Swerling number). However, the concept of FAR is embodied at a fundamental level in the detection process, which traditionally employs a Constant-FAR (CFAR) detector to set thresholds for initial decisions on whether a target is present or not. While useful, such a metric for radar specification and system comparison is not without some serious shortcomings. In particular, when comparing FAR across various radar systems, some degree of normalization needs to occur to account for perhaps swath width and scan rates. This in turn suggests some useful testing strategies.

The classical rule-of-thumb for Synthetic Aperture Radar (SAR) is that a uniformly illuminated antenna aperture may allow continuous stripmap imaging to a resolution of half its azimuth dimension. This is applied to classical line-by-line processing as well as mosaicked image patches, that is, a stripmap formed from mosaicked spotlight images; often the more efficient technique often used in real-time systems. However, as with all rules-of-thumb, a close inspection reveals some flaws. In particular, with mosaicked patches there is significant Signal to Noise ratio (SNR) degradation at the edges of the patches due to antenna beam roll-off. We present in this paper a calculation for the optimum antenna beamwidth as a function of resolution that maximizes SNR at patch edges. This leads to a wider desired beamwidth than the classical calculation.

Synthetic Aperture Radar (SAR) is well known to afford imaging in darkness and through clouds, smoke, and other obscurants. As such, it is particularly useful for mapping and monitoring a variety of natural and man-made disasters. A portfolio of SAR image examples has been collected using General Atomics Aeronautical Systems, Inc.'s (GA-ASI's) Lynx^® family of Ku-Band SAR systems, flown on both operational and test-bed aircraft. Images are provided that include scenes of flooding, ice jams in North Dakota, agricultural field fires in southern California, and ocean oil slicks from seeps off the coast of southern California.

In this paper, we design and implement a digital impulse generator using a DCM block and an OSERDES block for a 24GHz UWB impulse-Doppler radar. The Federal Communications Commission (FCC) has confirmed the spectrum from 22 to 29GHz for UWB radar with a limit power of -41.3dBm/MHz. UWB signal possesses an absolute bandwidth larger than 500MHz or a relative bandwidth up to 20%. The vehicle radar is the key technology with the inherent advantage detected the distance and the velocity regardless of weather. Radar has a role to measure the distance and the velocity of long-distance vehicle. But, the radar with 1m resolution is difficult to satisfy the detection performance in the blind spot zone because the blind spot zone needs high resolution. So, UWB impulse-Doppler radar with 30cm resolution is suitable for the blind spot zone. The designed impulse generator has a 2ns pulse width and 100us PRI. We perform simulations through Xilinx ISE; experiments use a spectrum analyzer and a digital oscilloscope. For UWB radar, we use an AD9779 DAC module with a 1Gsps maximum sampling rate. For equipment, we use a TDS5104B oscilloscope of Tektronix with 3dB bandwidth at 1GHz for the analysis of the time domain and an E4448A spectrum analyzer of Agilent with a 50GHz spectrum for the analysis of the frequency domain. The results of the digital impulse measurement show a 2ns pulse width in the time domain, a 500MHz bandwidth, and a 10KHz spectrum peak in the frequency domain.

To improve road safety and realize intelligent transportation, Ultra-Wideband (UWB) radars sensor in the 24 GHz domain are currently under development for many automotive applications. Automotive UWB radar sensor must be small, require low power and inexpensive. By employing a direct conversion receiver, automotive UWB radar sensor is able to meet size and cost reduction requirements. We developed Automotive UWB radar sensor for automotive applications. The developed receiver of the automotive radar sensor is direct conversion architecture. Direct conversion architecture poses a dc-offset problem. In automotive UWB radar, Doppler frequency is used to extract velocity. The Doppler frequency of a vehicle can be detected using zero-padding Fast Fourier Transform (FFT). However, a zero-padding FFT error is occurs due to DC-offset problem in automotive UWB radar sensor using a direct conversion receiver. Therefore, dc-offset problem corrupts velocity ambiguity. In this paper we proposed a mean-padding method to reduce zero-padding FFT error due to DC-offset in automotive UWB radar using direct conversion receiver, and verify our proposed method with computer simulation and experiment using developed automotive UWB radar sensor. We present the simulation results and experiment result to compare velocity measurement probability of the zero-padding FFT and the mean-padding FFT. The proposed algorithm simulated using Matlab and experimented using designed the automotive UWB radar sensor in a real road environment. The proposed method improved velocity measurement probability.

The nature of the recent military conflicts and terrorist attacks along with the necessity to protect bases, convoys and patrols have made a serious impact on the development of more effective security systems. Current widely-used perimeter protection systems with zone sensors will soon be replaced with multi-sensor systems. Multi-sensor systems can utilize day/night cameras, IR uncooled thermal cameras, and millimeter-wave radars which detect radiation reflected from targets. Ranges of detection, recognition and identification for all targets depend on the parameters of the sensors used and of the observed scene itself. In this paper two essential issues connected with multispectral systems are described. We will focus on describing the autonomous method of the system regarding object detection, tracking, identification, localization and alarm notifications. We will also present the possibility of configuring the system as a stationary, mobile or portable device as in our experimental results.

In this paper, we analyze the resolution of bistatic synthetic aperture radar (BISAR) imaging for stationary objects. In particular, we analyze the resolution of images reconstructed by the method of a filtered backprojection inversion, an inversion method which is derived from a scalar wave equation model. In this context we are able to account for the effects of antenna beam patterns and arbitrary flight trajectories. The analysis is done by examining the data collection manifold for different experiment geometries and system parameters.

Researchers at the U.S. Army Research Laboratory (ARL) designed and fabricated the Synchronous Impulse REconstruction (SIRE) radar system in an effort to address fundamental questions about the utilization of low frequency, ultrawideband (UWB) radar. The SIRE system includes a receive array comprising 16 receive channels, and it is capable of operating in either a forward-looking or a side-looking mode. When operated in side-looking mode, it is capable of producing high-resolution Synthetic Aperture Radar (SAR) data. The SAR imaging algorithms, however, initially operated under the assumption that the vehicle followed a nearly linear trajectory throughout the data collection. Under this assumption, the introduction of vehicle path nonlinearities distorted the processed SAR imagery. In an effort to mitigate these effects, we first incorporated segmentation routines to eliminate highly non-linear portions of the path. We then enhanced the image formation algorithm, enabling it to process data collected from a non-linear vehicle trajectory. We describe the incorporated segmentation approaches and compare the imagery created before and after their incorporation. Next, we describe the modified image formation algorithm and present examples of output imagery produced by it. Finally, we compare imagery produced by the initial segmentation algorithm to imagery produced by the modified image-formation algorithm, highlighting the effects of segmentation parameter variation on the final SAR image.

A wideband photonic RF vector modulator with novel architecture is presented and demonstrated with capability of continuous amplitude modulation and 0°-360° phase shifting. In-phase and quadrature-phase components of the output signal are used to produce 360° continuous phase shifting and optical attenuation is used to control the signal amplitude. A novel 8-tap all-optical transverse-filter implementing a Hilbert transform is proposed and demonstrated to produce 90° quadrature phase-shift for the broadband RF signal. Experimental apparatus and results for continuous vector modulation will be presented for the frequency range of 1-6 GHz.

The expression for the spectral degree of coherence of a far field scattered from a collection of particles, in the polar spherical coordinate system, is derived. Such an expression may be reduced to the one in the Cartesian coordinate system in the special case when the two points considered are in the same radial direction.

Techniques such as clinometry, stereoscopy, interferometry, and polarimetry are used for Digital Elevation Model (DEM) generation from Synthetic Aperture Radar (SAR) images. The choice of technique depends on the SAR configuration, the means used for image acquisition, and the relief type. The most popular techniques are interferometry for regions of high coherence and stereoscopy for regions such as steep forested mountain slopes. Stereo matching, which is finds the disparity map or correspondence points between two images acquired from different sensor positions, is a core process in stereoscopy. Additionally, automatic stereo processing, which involves stereo matching, is an important process in other applications including vision-based obstacle avoidance for unmanned air vehicles (UAVs), extraction of weak targets in clutter, and automatic target detection. Due to its high computational complexity, stereo matching has traditionally been, and continues to be, one of the most heavily investigated topics in computer vision. A stereo matching algorithm performs a subset of the following four steps: cost computation, cost (support) aggregation, disparity computation/optimization, and disparity refinement. Based on the method used for cost computation, the algorithms are classified into feature-, phase-, and area-based algorithms; and they are classified as local or global based on how they perform disparity computation/optimization. We present a comparative performance study of two pairs, i.e., four versions, of global stereo matching codes. Each pair uses a different minimization technique: a simulated annealing or graph cut algorithm. And, the codes of a pair differ in terms of the employed global cost function: absolute difference (AD) or a variation of normalized cross correlation (NCC). The performance comparison is in terms of execution time, the global minimum cost achieved, power and energy consumption, and the quality of generated output. The results of this preliminary study provide insights into the suitability and relative merits of these algorithms and cost functions for execution on field-deployable and on-board computer systems with size, weight, and power (SWaP) constraints. The results show that for 12 out of 14 instances the graph cut codes, compared to their simulated annealing counterparts provided a 35-85% improvement in energy consumption and, therefore, are promising candidates for use in field-deployable and on-board systems.

This paper proposes two studies on a Shark antenna array, working in the frequency band [800MHz - 8GHz], in a configuration including N generators and N antennas. The first study deals with the evaluation of the performances of the array from the analyze of the transient performances of the elementary system "generator + antenna". The second study concerns the comparison of two arrays having the same surface area, but a different number of antennas thanks to a scaling method on the dimensions of the elementary antenna.

Pulse reflection between front-end components is a common problem for impulse radar systems. Such reflections arise because radio frequency components are rarely impedance-matched over an ultra-wide bandwidth. Any mismatch between components causes a portion of the impulse to reflect within the radar front-end. If the reflection couples into the transmit antenna, the radar emits an unintended, delayed and distorted replica of the intended radar transmission. These undesired transmissions reflect from the radar environment, produce echoes in the radar image, and generate false alarms in the vicinity of actual targets. The proposed solution for eliminating these echoes, without redesigning the transmit antenna, is to dissipate pulse reflections in a matched load before they are emitted. A high-speed switch directs the desired pulse to the antenna and redirects the undesired reflection from the antenna to a matched load. The Synchronous Impulse Reconstruction (SIRE) radar developed by the Army Research Laboratory (ARL) is the case-study. This paper reviews the current front-end design, provides a recent radar image which displays the aforementioned echoes, and describes the switch-cable-load circuit solution for eliminating the echoes. The consequences of inserting each portion of the new hardware into the radar front-end are explained. Measurements on the front-end with the high-speed switch show an attenuation of the undesired pulse transmissions of more than 18 dB and an attenuation in the desired pulse transmission of less than 3 dB.

MIMO systems have revolutionized wireless communications resulting in unprecedented channel capacity. This breakthrough led researchers in radar as well as wireless communications communities to investigate the applicability of MIMO systems to radar. Preliminary research is showing that the full benefits of MIMO technology is realized when antenna spacing results in a decorrelated target scattering matrix. This requires antenna placement such that each receiver is observing an independent view of the target. Research is also showing that suboptimal improvements can be attained when the scattering matrix is partially correlated. This situation arises when antennas are collocated. In this work, we investigate the feasibility of MIMO Radar technology when antenna placement is quite restricted, such as in phased-array antennas. We extend the theoretical results for the correlation coefficients derived for statistical MIMO radar. We apply these results to assess the degree of decorrelation that can be achieved with Phase-Array antennas. We quantify our results as a function of antenna element spacing, frequency band and target RCS. In addition, we quantify the degree of decorrelation that is achievable by antennas that are typical in a tactical missile environment. Our results show that even when the antennas are quite small, it is possible to achieve a significant degree of decorrelation for a certain class of targets and certain frequency bands.

Moving target indication (MTI) algorithms often operate within a relatively narrow frequency band relying on Doppler processing to detect moving targets at long standoff ranges. At these standoff ranges, received wavefronts impinging on a linear array can be considered planar, enabling implementation of a variety of phase-based beam-forming techniques. At near ranges, however, the plane-wave assumption no longer holds. We describe enhancements to an impulse-based, low-frequency, ultra-wideband, moving-target imaging system for near-range, through-the-wall MTI. All MTI image processing is performed in the time domain using a change detection (CD) paradigm. We discuss how MTI image quality can be increased through the introduction of randomized linear arrays. After describing the process in detail, we present results obtained using data collected by an impulse-based, low frequency, ultra-wideband system.