This PDF file contains the front matter associated with SPIE Proceedings Volume 8361, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

This paper presents an analysis of the measurement of the angular velocity of walking humans using a millimeter-wave correlation interferometer. Measurement of the angular velocity of moving objects is a desirable function in remote sensing applications. Doppler radar sensors are able to measure the signature of moving humans based on micro-Doppler analysis; however, a person moving with little to no radial velocity produces negligible Doppler returns. Measurement of the angular movement of humans can be done with traditional radar techniques, however the process involves either continuous tracking with narrow beamwidth or angle-of-arrival estimation algorithms. A new method of measuring the angular velocity of moving objects using interferometry has recently been developed which measures the angular velocity of an object without tracking or complex processing. The frequency of the interferometer signal response is proportional to the angular velocity of the object as it passes through the interferometer beam pattern. In this paper, the theory of the interferometric measurement of angular velocity is covered and simulations of the response of a walking human are presented. Simulations are produced using a model of a walking human to show the significant features associated with the interferometer response, which may be used in classification algorithms.

In this paper, the problem of adaptively selecting radar waveforms from a pre-dened library of waveforms is addressed from an information theoretic perspective. Typically, radars transmit specic waveforms periodically, to obtain for example, the range and Doppler of a target. Although modern radars are capable of transmitting dierent waveforms during each consecutive period of transmission, it is hitherto unclear as to how these waveforms must be scheduled to best understand the dynamic radar scene. In this paper, a new information theoretic metric - directed information - is employed for waveform scheduling, and is shown to incorporate the past radar returns to eectively schedule waveforms. We formulate this waveform scheduling problem in a Gaussian framework, derive the corresponding maximization problem, and illustrate several special cases.

Ground Penetrating Radars (GPR) process electromagnetic reflections from subsurface interfaces to characterize the subsurface and detect buried targets. Our objective is to test an inversion algorithm that calculates the intrinsic impedance of subsurface media when the signal transmitted is modeled as the first or second derivative of a large bandwidth Gaussian pulse. For this purpose we model the subsurface as a transmission line with multiple segments, each having different propagating velocities and characteristic impedances. We simulate the propagation and reflection of the pulse from multilayered lossless and lossy media, and process the received signal with a rectifier and filter subsystem to estimate the impulse response. We then run the impulse response through the inversion algorithm in order to calculate the relative permittivity of each subsurface layer. We show that the algorithm is able to detect targets using the primary reflections, even though secondary reflections are sometimes required to maintain inversion stability. We also demonstrate the importance of compensating for geometric spreading losses and conductivity losses to accurately characterize each substrate layer and target. Such compensation is not trivial in experimental data where electronic range delays can be arbitrary, transmitted pulses often deviate from the theoretical models, and limited resolution can cause ambiguity in the range of the targets.

In this paper, we consider resolving over-the-horizon radar (OTHR) Doppler returns. A high-resolution time-frequency (TF) representation of the received signal is obtained by using the local polynomial Fourier transform (LPFT). From the optimally concentrated LPFT, multicomponent Doppler signatures, which are only several frequency bins apart, are extracted using an instantaneous frequency estimation method based on the Viterbi algorithm. The performance of the proposed method is validated using real data.

In synthetic-aperture radar (SAR) returned signals, ground-target vibrations introduce a phase modulation that is linearly proportional to the vibration displacement. Such modulation, termed the micro-Doppler effect, introduces ghost targets along the azimuth direction in reconstructed SAR images that prevents SAR from forming focused images of the vibrating targets. Recently, a discrete fractional Fourier transform (DFrFT) based method was developed to estimate the vibration frequencies and instantaneous vibration accelerations of the vibrating targets from SAR returned signals. In this paper, a demodulation-based algorithm is proposed to reconstruct focused SAR images of vibrating targets by exploiting the estimation results of the DFrFT-based vibration estimation method. For a single-component harmonic vibration, the history of the vibration displacement is first estimated from the estimated vibration frequency and the instantaneous vibration accelerations. Then a reference signal whose phase is modulated by the estimated vibration displacement with a delay of 180 degree is constructed. After that, the SAR phase history from the vibration target is multiplied by the reference signal and the vibration-induced phase modulation is canceled. Finally, the SAR image containing the re-focused vibration target is obtained by applying the 2-D Fourier transform to the demodulated SAR phase history. This algorithm is applied to simulated SAR data and successfully reconstructs the SAR image containing the re-focused vibrating target.

This paper presents a 3D ltered inversion scheme for turntable inverse synthetic aperture radar (ISAR) data from a scalar wave equation model. The proposed inversion scheme targets at the use of ltered back projection (FBP) and convolution back projection (CBP) imaging algorithms. In the paper, we also provide a derivation of a general imaging lter for the near-elds FBP and CBP imaging algorithms.

We have previously shown that Stokes eigenvectors can be numerically extracted from the Kennaugh(Stokes) matrices of both single-look and multilook fully polarimetric SIR-C data. The extracted orientation and ellipticity parameters of the Stokes eigenvector were found to be related to the Huynen orientation and helicity parameters for single-look fully polarimetric SIR-C data. We formally show in this paper that these two parameters, which diagonalize the Sinclair matrices of the single-look data, belong to a set of parameters which diagonalize the Kennaugh matrices of single-look data. Along with the cross sections kS_vvk², kS_hvk², kS_hhk² and the Span, the eigenvalues of the Kennaugh matrix and the covariance matrix are used as input features in the development of a neural net landcover classifier for SIR-C data.

With the advent of a new sampling theory in recent years, compressed sensing (CS), it is possible to reconstruct signals from measurements far below the Nyquist rate. The CS theory assumes that signals are sparse and that measurement matrices satisfy certain conditions. Even though there have been many promising results, unfortunately there still exists a gap between the theory and actual real world applications. This is because of the fundamental problem that the CS formulation is discrete. We propose a sampling and reconstructing method for frequency-sparse signals that addresses this issue. The signals in our scenario are supported in a continuous sparsifying domain rather than discrete. This work focuses on a typical case in which the unknowns are frequencies and amplitudes. However, directly looking for the unknowns that best fit the measurements in the least-squares sense is a non-convex optimization problem, because sinusoids are oscillatory. Our approach extends the utility of CS to simplify this problem to a locally convex problem, hence making the solutions tractable. Direct measurements are taken from non-uniform time-samples, which is an extension of the CS problem with a subsampled Fourier matrix. The proposed reconstruction algorithm iteratively approximates the solutions using CS and then accurately solves for the frequencies with Newton's method and for the amplitudes with linear least squares solutions. Our simulations show success in accurate reconstruction of signals with arbitrary frequencies and significantly outperform current spectral compressed sensing methods in terms of reconstruction fidelity for both noise-free and noisy cases.

A multi-featured sensor solution has been developed that enhances the operational safety and functionality of small airborne platforms, representing an invaluable stride toward enabling higher-risk, tactical missions. This paper demonstrates results from a recently developed multi-functional sensor system that integrates a high performance millimeter-wave radar front end, an evidence grid-based integration processing scheme, and the incorporation into a 3D Synthetic Vision System (SVS) display. The front end architecture consists of a w-band real-beam scanning radar that generates a high resolution real-time radar map and operates with an adaptable antenna architecture currently configured with an interferometric capability for target height estimation. The raw sensor data is further processed within an evidence grid-based integration functionality that results in high-resolution maps in the region surrounding the platform. Lastly, the accumulated radar results are displayed in a fully rendered 3D SVS environment integrated with local database information to provide the best representation of the surrounding environment. The integrated system concept will be discussed and initial results from an experimental flight test of this developmental system will be presented. Specifically, the forward-looking operation of the system demonstrates the system's ability to produce high precision terrain mapping with obstacle detection and avoidance capability, showcasing the system's versatility in a true operational environment.

This paper presents ARTEMIS, Inc.'s approach to development of end-to-end synthetic aperture radar systems for multiple applications and platforms. The flexible design of the radar and the image processing tools facilitates their inclusion in a variety of application-specific end-to-end systems. Any given application comes with certain requirements that must be met in order to achieve success. A concept of operation is defined which states how the technology is used to meet the requirements of the application. This drives the design decisions. Key to adapting our system to multiple applications is the flexible SlimSAR radar system, which is programmable on-the-fly to meet the imaging requirements of a wide range of altitudes, swath-widths, and platform velocities. The processing software can be used for real-time imagery production or post-flight processing. The ground station is adaptable, and the radar controls can be run by an operator on the ground, on-board the aircraft, or even automated as part of the aircraft autopilot controls. System integration takes the whole operation into account, seeking to flawlessly work with data links and on-board data storage, aircraft and payload control systems, mission planning, and image processing and exploitation. Examples of applications are presented including using a small unmanned aircraft at low altitude with a line of sight data link, a long-endurance UAV maritime surveillance mission with on-board processing, and a manned ground moving target indicator application with the radar using multiple receive channels.

This paper presents a rotating W band radar performing a full body scan of persons which are moving with constant speed below the radar. The radar consists of a FMCW module sweeping the frequency between 96 GHz and 99 GHz by a varactor tuned VCO. The transmit and receive modules are fabricated in split-block technology using 100 nm metamorphic HEMT MMICs. The used 4 channel receiver operates between 84 GHz and 104 GHz and an average noise figure of 3.5 dB. Polarimetric measurements are carried out for the detection of oblong objects such as explosive tubes.

Radar can provide inexpensive wide-area surveillance of river and port traffic for both security and emergency response. We demonstrate the tracking of multiple vessels as well as the micro-Doppler signatures of different classes of small vessels, including kayaks and zodiacs. The pattern of life of a river is analyzed over several days and can be used to easily identify suspicious or unusual cases.

The use of High-Frequency MicroWaves (HFMW) for high-resolution imagery has gained interest over the last years. Very promising in-depth applications can be foreseen for composite non-metal, non-polarized materials, widely used in the aeronautic and aerospace industries. Most of these materials present a high transparency in the HFMW range and, therefore, defects, delaminations or occlusions within the material can be located. This property can be exploited by applying 3-D HFMW imaging where conventional focused imaging systems are typically used but a different approach such as Synthetic Aperture (SA) radar can be addressed. This paper will present an end-to-end 3-D imagery system for short-range, non-destructive testing based on a frequency-modulated continuous-wave HFMWsensor operating at 100 GHz, implying no health concerns to the human body as well as relatively low cost and limited power requirements. The sensor scans the material while moving sequentially in every elevation plane following a 2-D grid and uses a significantly wide beam antenna for data acquisition, in contrast to focused systems. Collected data must be coherently combined using a SA algorithm to form focused images. Range-independent, synthetically improved cross-range resolutions are remarkable added values of SA processing. Such algorithms can be found in the literature and operate in the time or frequency domains, being the former computationally impractical and the latter the best option for in-depth 3-D imaging. A balanced trade-off between performance and image focusing quality is investigated for several SA algorithms.

This paper describes the analysis of fully polarimetric measurements acquired by the space-borne SAR-systems Radarsat 2 and TerraSAR-X with respect to the application on reconnaissance and security purposes. Firstly the analysis is carried out for single pixels by using the method of calculating SAR signatures in dependence on the orientation and ellipticity angle of the polarization ellipse. This method is applied to canonical scattering objects as well as to real measurements. Secondly, as a collective method for analyzing the types of scatterers in a measured image the entropy-alpha-method is used. The method is deployed to measurements from the JPL AIRSAR, DLR FSAR, Radarsat 2, and TerraSAR-X. The goal is to cover the different frequencies at L-, C-, and X-band as well as different spatial resolutions from 30m down to 0,25m. The results show that the methods are very well applicable. The application to real measurements demonstrates that the polarimetric effects are mainly dependent on frequency, spatial resolution and especially speckle or thermal noise. Proposals are given for upcoming analysis.

Tracking the progress and impact of large scale projects in areas of active conflict is challenging. In early 2010, the Canadian International Development Agency (CIDA) broke ground on an ambitious project to rehabilitate a network of just under 600 km of canals that supply water from the Arghandab River throughout southern Kandahar Province thereby restoring a reliable and secure water supply and stimulating a once vibrant agricultural region. Monitoring the region for signs of renewal is difficult due to the large areal extent of the irrigated land and safety concerns. With the support of the Canadian Space Agency, polarimetric change detection techniques are applied to space-borne SAR data to safely monitor the area through a time-series of RADARSAT-2 images acquired during the rehabilitation ground work and subsequent growing seasons. Change detection maps delineating surface cover improvement will aid CIDA in demonstrating the positive value of Canada's investment in renovating Afghanistan's irrigation system to improve water distribution. This paper examines the use of value-added SAR imaging products to provide short- and long-term monitoring suitable for assessing the impact and benefit of large scale projects and discusses the challenges of integrating remote sensing products into a non-expert user community.

We have developed a new technology for detecting underground tunnels - the Tunnel Detection Focused-Source Electromagnetic (TD-FSEM) method. It uses four horizontal electric dipole transmitters and a five-electrode grounded quadrupole receiver to measure the transient EM field. Such a setup directs the current under the receiver vertically, increasing the sensitivity of the measurement system to a relatively narrow column of subsurface media directly below the receiver. Our previously published feasibility modeling results allowed us to prove the concept by showing that the method provides data sufficient for reliable detection of clandestine tunnels embedded in a homogeneous subsurface. In this paper, we present a 3D EM modeling results showing that our method can be efficiently used in presence of near-surface conductive or resistive obstructions. We present comparisons of GPR, conventional dipole-dipole Controlled-Source EM (CSEM), and our TD-FSEM methods and show that the TD-FSEM, unlike conventional GPR and CSEM, allows for removal of unwanted shallow/near-surface masking effects. The TD-FSEM acquisition and processing unit can be mounted on a vehicle performing large-scale regional and local operations.

This paper provides a summary of the development of a three state machine-based cooperative control algorithm that is applied to multiple Unmanned Aerial Vehicles (UAVs) or Micro-Aerial Vehicles (MAVs) control. We use MAVs for cooperative search of a hidden electromagnetic source (emitter) in a controlled environment. MAVs are equipped with wireless sensor nodes capable of sensing an electromagnetic (EM) field around them. Simultaneous control and sensing capabilities of these MAVs are presented. The algorithm uses a three-state machine to control the MAVs during the search process. The first state is a decentralized cooperative search that allows MAVs to obtain information about the environment and detect EM emissions from the target. The second state implements a gradient descent algorithm in which the MAVs converge towards the target based on the received signal strength, while still maintaining a proximal distance from each other. MAVs are positioned at the optimal distance of the detected EM source before fine-tuning of the emitter localization is carried out. The third state incorporates a technique called Position-Adaptive Direction Finding (PADF), where the MAVs adapt their positions in order to further improve localization of a hidden emitter using an estimated path loss exponent as a feedback. We present simulation and experimental data that illustrate the proposed approach.

This paper investigates the expected performance of a ground-based, multi-story building imaging radar system through far-field and near-field computer models. We created a 3-D computer-aided design model of a complex two-story building, simulated the radar response from this complex structure for various geometries and applied synthetic aperture radar image formation algorithms consistent with the simulation scenarios. In this study, we employed the Finite Difference Time Domain method and the Xpatch software to compute the radar signatures. The numerical results give a better understanding of the phenomenology of the scattering and imaging processes and show that relying solely on the far-field scattering data at one elevation angle is not sufficient to obtain the multi-story building layout. Multiple elevation angle views are required in order to determine the location of imaged objects in the vertical direction. Xpatch simulation results in a near-field strip-map configuration suggest a way to achieve this goal within the constraints of a ground-based radar system.

Through-wall radar imaging is an emerging technology with great interest to military and police forces operating in an urban environment. A through-wall imaging radar can potentially provide interior room layouts as well as detection and localization of targets of interest within a building. In this paper, we present our through-wall radar system mounted on the side of a vehicle and driven along a path in front of a building of interest. The vehicle is equipped with a LIDAR (Light Detection and Ranging) and motion sensors that provide auxiliary information. The radar uses an ultra wideband frequency-modulated continuous wave (FMCW) waveform to obtain high range resolution. Our system is composed of a vertical linear receive array to discriminate targets in elevation, and two transmit elements operated in a slow multiple-input multiple output (MIMO) configuration to increase the achievable elevation resolution. High resolution in the along-track direction is obtained through synthetic aperture radar (SAR) techniques. We present experimental results that demonstrate the 3-D capability of the radar. We further demonstrate target detection behind challenging walls, and imagery of internal wall features. Finally, we discuss future work.

In this paper, we describe an operational pulse Doppler radar imaging system for indoor target localization and classification, and show how a target's micro-Doppler signature (μDS) can be processed when ultra-wideband (UWB) waveforms are employed. Unlike narrowband radars where time-frequency signal representations can be applied to reveal the target time-Doppler frequency signatures, the UWB system permits joint range-time-frequency representation (JRTFR). JRTFR outputs the data in a 3D domain representing range, frequency, and time, allowing both the μDS and high range resolution (HRR) signatures to be observed. We delineate the relationship between the μDS and the HRR signature, showing how they would form a complimentary joint feature for classification. We use real-data to demonstrate the effectiveness of the UWB pulse-Doppler radar, combined with nonstationary signal analyses, in gaining valuable insights into human positioning and motions.

Detection of stationary targets in urban sensing and through-the-wall radar images using likelihood ratio test (LRT) detectors has recently been considered in the literature. A shortcoming of the LRT detectors is that appropriate probability density functions of target and clutter images need to be predefined. In most practical scenarios, this information is not available a priori, and the mismatch of the assumed distribution functions degrades the performance of the LRT. In this paper, we apply image segmentation techniques to radar images of scenes associated with urban sensing. More specifically, the Otsu's method and maximum entropy segmentation are considered to aid in removing the clutter, resulting in enhanced radar images with target regions only. Performance of the segmentation schemes is evaluated and compared to that of the assumed LRT detector using real-data collected with Defence Research and Development Canada's vehicle-borne through-the-wall radar imaging system. The results show that, although the principles of segmentation and detection are different and serve disparate objectives, the segmentation techniques outperform the LRT detector for the considered cases.

Fully polarimetric through-the-wall radar measurements with high spatial resolution have been attained by using the ISAR (Inverse Synthetic Aperture Radar) technique. Polarimetric methods may reduce the effects of the wall interaction and increase the contrast between humans and the background. The main scene in the measurements was a human sitting in a small wooden cabin. The cabin was placed on a turntable and rotated, to obtain ISAR imaging. By switching the transmitter and receiver antennas between horizontal and vertical polarizations, four polarization combinations were obtained. Phase coherence was maintained through a whole measurement series. This enabled co-processing of the whole collected data set with coherent methods. A statistical description of the measured data was used, with the polarimetric coherency matrix applied to the received signals. ISAR images produced for the TTW scenes show that the human can be discerned from the background. The contrast between the human and the background was found to be greater with vertical polarization at transmit and receive, with less contrast using cross-polarization or horizontal co-polarization, due to the horizontal wall grain orientation. A classification scheme based on the eigenparameters of the coherency matrix (entropy, anisotropy and alpha angle) and the backscatter power has been tested to discriminate between different target objects in the cabin. The method shows some promise, but a reliable classification has not yet been attained.

The reduction of size, weight, power, and cost (SWaP-C) of radio frequency (RF) components is becoming increasingly important to meet industry requirements. In meeting the SWaP-C objectives, RF components will be required to be smaller and more power efficient than the current state- of- the- art while sustaining high performance functionality. In compliance with SWaP-C and high performance functionality is a High Efficiency Switching Power Amplifier. This study focuses on the more efficient breed of switching power amplifiers (PAs), particularly the Class F PA with new techniques to operate broadband on multiple radar bands. Efficiencies in the range of 60% to 80% for Class F PAs have been reported in literature; however, this efficiency is only attainable over narrow bandwidths on the order of 10%. Several innovative techniques have been identified to increase the efficiency and operational bandwidth of RF power amplifiers (PAs) for radar applications. The amplifier design also incorporates fast turn on and turn off circuits to achieve switching times of less than one microsecond (μs). This enables the PA to be switched off during the receive period to prevent self-generated noise from corrupting the received signal. Also, high-power transmit and receive (T/R) switches at the antenna feed can be eliminated. A wideband PA enables the design of a multi-band radar, reducing the number of components needed for operation in the L and X bands. A high efficiency PA is also key to reducing battery size and cooling requirements in radar applications.

This presentation describes an active antenna array architecture designed specifically for achieving low transmit and receive sidelobe levels without having to use attenuators to create the necessary aperture taper. An "irregular" subarray approach is used to eliminate the need for tapered-attenuation within the array's aperture, thereby drastically reducing the DC supply power consumption of the active phased array. On many UAVs, especially the smaller models, onboard DC power can be extremely limited. The so-called "irregular" subarray approach not only determines the exact locations of the T/R modules, but it also allows for all of the low-noise amplifiers to share the same part number and for all of the power amplifiers to also share the same part number. All of the LNAs are biased exactly in the same manner as are all of the PAs. By keeping the part numbers and bias conditions of the amplifiers the same, large instantaneous operational bandwidths can be obtained. Thus, this paper illustrates an active antenna array topology that can achieve wideband performance and low sidelobe levels with minimal DC power consumption.

We investigate a vircator as an economical high power pulsed microwave source for radar. Because of the inconsistency of spark gaps in the driver and operation of the tube based vircator, the resulting ringing pulse has a different pulse shape each time a pulse is generated. Therefore every time we pulse the source we must remove the effects of the ringing source pulse from the data resulting from that pulse. Scattering from a scene is considered random (white noise) with a superimposed non-white component due to the pulse. We propose a whitening filter to remove the effects of the ringing pulse from the random data. This produces a similar result as spectral factorization in which we first determine the pulse from the power spectrum of the data and then deconvolve the ringing pulse out of the received data. The removal of pulse specific ringing increases range resolution and allows data from sequential pulses from a single vircator or pulses from separate vircators to be combined for joint processing in a synthetic aperture radar.

The progress in developing metallic metamaterial lenses founded on stacked subwavelength hole arrays is reported. Before, the lens was studied when it emulates a medium with effective index of refraction -1. Here, the lens is investigated at higher frequencies, where it behaves like a near-zero index of refraction. We show that exploiting both regimes, dual-band capabilities are attainable. Moreover, a zoning technique is applied to the initial design to reduce the lens in terms of volume and weight, while the performance is maintained.

This paper proposes an image-based automatic target detection algorithm to be used in clutter and sparse target environments. We intend to apply the algorithm to an ultra-wideband multispectral radar concept by means of employing multi-carrier waveforms based upon Orthogonal Frequency Division Multiplexing (OFDM) modulation. Individual sub-bands of an OFDM waveform can be processed separately to yield range and cross-range reconstruction of a target scene containing both targets and clutter. Target detection in resulting images will be performed and contrasted with the detection performance of a traditional fixed-waveform Synthetic Aperture Radar system. The target detection algorithm is implemented through the use of scalar and vector field operations performed on the images from the reconstructed target scene. We hypothesize that the use of vector operations and field analysis will allow for an adaptive approach to the detection of targets within clutter.

We have developed a method for radar/sonar target discrimination employing techniques from non-linear dynamics. We demonstrate our method by simulating radar scattering from four similar targets where the radar wavelength and bandwidth resolution is on the order of the target size. We find that this method results in a high probability of target discrimination even in the presence of large amplitude noise and spurious clutter. We also present experimental data for acoustic wave scattering from two similarly sized targets. This bench top data was taken in the presence of large random noise and clutter. The identification method is shown to work over a wide range of angles.

We consider a problem of detecting a random spatially distributed signal source by an array of sensors based on the generalized approach to signal processing in noise. We derive some generalized detector (GD) structures under several assumptions on the available statistics. The GD performance is evaluated and the effect of source angular spread is investigated. We notice the degrees of freedom of detection statistic distributions depend on both the signal angular spread and the number of data snapshots. At high signal-to-noise ratio and with small degrees of freedom, an increase of angular spread improves the detection performance. With large degrees of freedom the increase of angular spread reduces the detection performance. A comparison between GD and conventional beamformer is carried out by computer simulations. The results indicate a superiority of GD as the angular spread becomes large over the conventional beamformer detector.

The ability to identify human movements can be an important tool in many different applications such as surveillance, military combat situations, search and rescue operations, and patient monitoring in hospitals. This information can provide soldiers, security personnel, and search and rescue workers with critical knowledge that can be used to potentially save lives and/or avoid a dangerous situation. Most research involving human activity recognition is focused on using the Short-Time Fourier Transform (STFT) as a method of analyzing the micro-Doppler signatures. Because of the time-frequency resolution limitations of the STFT and because Fourier transform-based methods are not well-suited for use with non-stationary and nonlinear signals, we have chosen a different approach. Empirical Mode Decomposition (EMD) has been shown to be a valuable time-frequency method for processing non-stationary and nonlinear data such as micro-Doppler signatures and EMD readily provides a feature vector that can be utilized for classification. For classification, the method of a Support Vector Machine (SVMs) was chosen. SVMs have been widely used as a method of pattern recognition due to their ability to generalize well and also because of their moderately simple implementation. In this paper, we discuss the ability of these methods to accurately identify human movements based on their micro-Doppler signatures obtained from S-band and millimeter-wave radar systems. Comparisons will also be made based on experimental results from each of these radar systems. Furthermore, we will present simulations of micro-Doppler movements for stationary subjects that will enable us to compare our experimental Doppler data to what we would expect from an "ideal" movement.

Knowing the statistical characteristics of the radar cross-section (RCS) of man-made, or cultural clutter, is crucial to the success of clutter mitigation, radar target detection algorithms, and radar system requirements in urban environments. Open literature studies regarding the statistical nature of cultural clutter focus primarily on radar probability models or limited experimental data analysis of specific locations and frequencies. This paper seeks to expand the existing body of work on cultural clutter RCS statistics at Ku-band for ground moving target indication (GMTI) and synthetic aperture radar (SAR) applications. We examine the normalized RCS probability distributions of cultural clutter in several urban scenes, across aspect and elevation angle, for vertical transmit/receive (VV) polarizations, and at diverse resolutions, using experimental data collected at Ku-band. We further describe frequency and RCS strength statistics of clutter discretes per unit area to understand system demands on radars operating in urban environments in this band.

Classifying human signatures using radar requires a detailed understanding of the RF scattering phenomenology associated with humans as well as their motion. We model humans engaged in the activity of walking and analyze the separability of different body parts with frequency as well as lookdown angle. This work seeks to estimate the ability to classify the micro-Doppler signals generated by human motion, and especially arm motion, as a function of the radar frequency and other parameters. The simulations imply that for classification using arm motion, frequencies at Ku-band or higher are probably required, and that lookdown angle has a significant effect on the classification capability of the radar. Additionally, the sensitivity of the system required to isolate the motion of different body parts is estimated.

Electromagnetic (EM) simulations from UHF to X-band are undertaken to determine the feasibility of exploiting ground surface scattering for subsurface target detection. A ground surface containing a target emplacement-related disturbance is physically represented by a composite roughness profile with Gaussian statistics, in conjunction with a homogeneous dielectric model. Bistatic wideband imaging of the target/surface disturbance scene-generated using scattering data from forward-looking sensing simulations-indicates that at X-band, for the parameters defined in this work, the disturbed surface feature is observable with a scattering response that is distinct-both visually and statistically-from that of the undisturbed surface. At lower frequencies (UHF to C-band), the disturbance is far less apparent due to the lack of sufficient image resolution. The buried target itself, however, is visible only at lower frequency bands. Consequently, it is expected that a judicious synthesis of the scattering data from different frequency bands can lead to enhanced target detection performance.

Proper development of ground-penetrating radar (GPR) technology requires a unique understanding of the electromagnetic (EM) properties of targets and background media. Thus, electromagnetic characterization of targets and backgrounds is fundamental to the success or failure of UWB GPR as a threat detection technique. In many cases, threats are buried in soil. Soil properties directly affect the radar signature of targets and determine the depth at which they can be detected by radar. One such property is permittivity. A portable system recently developed at the U.S. Army Research Laboratory measures permittivity in-situ with minimal disturbance of the dielectric sample. The measurement technique uses ring resonators. Design equations and physical dimensions are presented for fabricating resonators at frequencies between 600 MHz and 2 GHz. Only a handheld vector network analyzer, coaxial cabling, and the ring resonators are necessary for each measurement. Lookup curves generated in simulation are referenced to calculate the complex permittivity of the sample. The permittivity measurement is explained step-by-step, and data is presented for samples of soils from Ft. Irwin, California and Yuma, Arizona.

An approach for creating a low-cost Chaos Pulsed-Doppler Radar is presented. The objective of this effort is to develop a practical realization of a Chaotic Radar with performance advantages over other approaches. Many groups have proposed that Chaotic Waveforms are an effective radar signal generator due to: the relatively low cost of producing complex wideband waveforms and the difficulty in detecting and spoofing inherently complex modulations. PRA and Duke University report on the development of a radar design that uses a novel high-speed chaotic waveform generator. Preliminary experimental results are presented that characterize the performance of a chaotic waveform generator. In addition, the radar architecture will be proposed, realistic radar design criterion will be set forth, and simulations of a complete radar will be used to compare the chaotic radar to more traditional radar approaches.

In this contribution we examine the propagation of an ultrawideband (UWB) random noise signal through dispersive media such as soil, vegetation, and water, using Fourier-based analysis. For such media, the propagated signal undergoes medium-specific impairments which degrade the received signal in a different way than the non-dispersive propagation media. Theoretically, larger penetration depths into a dispersive medium can be achieved by identifying and detecting the precursors, thereby offering significantly better signal-to-noise ratio and enhanced imaging. For a random noise signal, well defined precursors in term of peak-amplitude don't occur. The phenomenon must therefore be studied in terms of energy evolution. Additionally, the distortion undergone by the UWB random noise signal through a dispersive medium can introduce frequency-dependent uncertainty or noise in the received signal. This leads to larger degradation of the cross-correlation function (CCF), mainly in terms of sidelobe levels and main peak deformation, and consequently making the information retrieval difficult. We would further analyze one method to restore the shape and carrier frequency of the input UWB random noise signal, thereby, improving the CCF estimation.

In prior work, we showed that any one of the state variables of the Lorenz chaotic flow can be used effectively as the instantaneous frequency of an FM signal. We further investigated a method to improve chaotic-wideband FM signals for high resolution radar applications by introducing a compression factor to the Lorenz flow equations and by varying two control parameters, namely ρ and β, to substantially increase the bandwidth of the signal. In this paper, we obtain an empirical quadratic relationship between these two control parameters that yields a high Lyapunov exponent which allows the Lorenz flow to quickly diverge from its initial state. This, in turn, results in an FM signal with an agile center frequency that is also chaotic. A time-frequency analysis of the FM signal shows that variable time-bandwidth products of the order of 10⁵ and wide bandwidths of approximately 10 GHz are achievable over short segments of the signal. Next, we compute the average ambiguity function for a large number of short segments of the signal with positive range-Doppler coupling. The resulting ambiguity surface is shaped as a set of mountain ridges that align with multiple range-Doppler coupling lines with low self-noise surrounding the peak response. Similar results are achieved for segments of the signal with negative range-Doppler coupling. The characteristics of the ambiguity surface are directly attributed to the frequency agility of the FM signal which could be potentially used to counteract electronic counter measures aimed at traditional chirp radars.

Noise radar systems encounter target fluctuation behavior similar to that of conventional systems. For noise radar systems, however, the fluctuations are not only dictated by target composition and geometry, but also by the non-uniform power envelope of their random transmit signals. This third dependency is of interest and serves as the basis for the preliminary analysis conducted in this manuscript. General conclusions are drawn on the implications of having a random power envelope and the impacts it could have on both the transmit and receive processes. Using an advanced pulse compression noise (APCN) radar waveform as the constituent signal, a computer simulation aids in quantifying potential losses and the impacts they might have on the detection performance of a real radar system.

Radio Frequency (RF) tomography has been proposed for imaging dielectric and conducting anomalies above-ground. Accordingly, low-cost electromagnetic transmitters are placed arbitrarily above ground, surrounding a large area of interest. In a preliminary stage, sensors identify their position, orientation, and time reference. Subsequently, a transmitter radiates a known waveform. The probing wave impinges upon a target (represented in terms of dielectric or conducting anomaly), thus producing scattered elds. Spatially distributed receivers collect samples of the total electric eld, remove noise, clutter and the direct path, and store the information concerning only the scattered eld. In the next iteration, a dierent transmitter is activated, or dierent wave- forms are used. Then, the collected data is typically relayed to a centralized location for processing and imaging. To ensure persistent sensing, fast back-propagation algorithms are implemented (either involving correlation or multiplication by a hermitian matrix). Resolution using back-propagation is aected by the sidelobe structure of the ambiguity function of the wave. Clearly, Linearly Stepped Frequency (LSF) waveform requires the lowest instantaneous bandwidth, but produces poor correlation properties. On the converse, Noise waveforms exhibit the idealized thumb-tack ambiguity function but typically require large instantaneous bandwidths. In an eort to exploit the benets of both individual waveforms, a noisy LSF waveform is developed. The NLSF performance, limitation and spectral dominance in reference to RF Tomography, along with its theoretical bounds, will be provided. Reconstructed images from simulated and experimental data will be compared.

A microwave-photonic, ultra-wideband (UWB) noise radar system is proposed and demonstrated. The system brings together photonic generation of UWB waveforms and fiber-optic distribution. The use of UWB noise provides high ranging resolution and better immunity to interception and jamming. Distribution over fibers allows for the separation the radar-operating personnel and equipment from the location of the front-end. The noise waveforms are generated using the amplified spontaneous emission that is associated with stimulated Brillouin scattering in a standard optical fiber, or with an erbium-doped fiber amplifier. Our experiments demonstrate a proof of concept for an integrated radar system, driven by optically generated UWB noise waveforms of more than 1 GHz bandwidth that are distributed over 10 km distance. The detection of concealed metallic object and the resolving of two targets with the anticipated ranging resolution are reported.

An ultrawideband (UWB) random noise radar operating at S-Band has been developed for through-wall detection, ranging, tracking, and imaging of targets. The system transmits a bandlimited UWB noise signal and accomplishes detection by cross-correlating the reflected signal with a time-delayed replica of the transmit signal. Noise radars have been found eminently suitable for most though-wall radar applications. Yet, in such scenarios, the antennas and the barrier (i.e. the wall) cause distortions in the return signal due to their frequency-dependent radiation and loss characteristics, respectively. In this paper, we explore the feasibility of characterizing the impulse response of various barriers and obstructions via measurements with the S-Band noise radar. As is well known, the entire operation of a linear system (e.g., antenna or barrier) can be captured in its impulse response h(t), i.e. the output of the system when excited by an impulse function at its input, δ(t). Thus, impulse response testing, generally, provides a complete diagnosis of the system over its entire mode of operation. This paper will present results on our impulse response characterization of the propagation and scattering environment through a barrier by the atypical method of cross correlation of noise signals. In addition, we will introduce a simple electromagnetic forward model for wall propagation and accompanying simulations.

This paper describes a millimeter-wave (mm-wave) radar system that has been constructed to simultaneously range and detect humans at distances up to 82 meters. This is done by utilizing a composite signal consisting of two waveforms: a wideband noise waveform and a single tone. These waveforms are summed together and transmitted simultaneously. Matched filtering of the received and transmitted noise signals is performed to range targets with high resolution, while the received single tone signal is used for Doppler analysis. The Doppler measurements are used to distinguish between different human movements using characteristic micro-Doppler signals. Using hardware and software filters allows for simultaneous processing of both the noise and Doppler waveforms. Our measurements establish the mm-wave system's ability to detect humans up to and beyond 80 meters and distinguish between different human movements. In this paper, we describe the architecture of the multi-modal mm-wave radar system and present results on human target ranging and Doppler characterization of human movements. In addition, data are presented showing the differences in reflected signal strength between a human with and without a concealed metallic object.

Principles of Active Radiometry are presented. Noise radiators are used to generate the low-coherence microwave noise field, and radiometers to evaluate its intensity, polarization and coherence. Several types of noise radiators are described as well as radiometers and antennas. The following applications are introduced: Material evaluation where insertion loss and reflectivity of grainy, irregular and moving objects are preferable. Microwave Coherence Tomography allowing the depth irregularity to be detected in low-loss objects. Near-Field antenna testing, field coherence evaluation, and spatial combining of noise radiators.

Trihedral corner reflectors are the preferred canonical target for SAR performance evaluation for many radar development programs. The conventional trihedrals have problems with substantially reduced Radar Cross Section (RCS) at low grazing angles, unless they are tilted forward, but in which case other problems arise mainly due to multipath effects. Consequently there is a need for better low grazing angle performance for trihedrals. This is facilitated by extending the bottom plate of the trihedral reflector. A relevant analysis of RCS for an infinite ground plate is presented. Practical aspects are also discussed.

ISAR has enjoyed some success in imaging maritime targets, particularly ships. In fact, a number of maritime ISAR systems have been operational for a number of years. With ISAR, the ship's own motion is critical to forming well-resolved ISAR images. Seemingly important to accounting for ship motion is to first understand the nature of the ship motion that we are likely to encounter. Designing ships for specific motion characteristics is the domain of naval architecture. This paper presents some preliminary analysis of naval architecture principles, and typical ship designs' impact on the ISAR problem.

It is fairly common in radar signal processing that sampled data is not sampled precisely at the desired positions within a function. Resampling the data to more advantageous sample locations entails interpolation of the data. The radar engineer often picks a resampling or an interpolation technique that "is handy", or "seems to work", without any analytical justification for his choice. However, understanding the science and mathematics that underpin interpolation can avoid unexpected and undesirable side effects from a suboptimal choice. This paper details interpolation kernel characteristics, allowing intelligent choices for algorithm design, tailored for radar signal processing applications.

Today, 77 GHz FMCW (Frequency Modulation Continuous Wave) radar has strong advantages of range and velocity detection for automotive applications. However, FMCW radar brings out ghost targets and missed targets in multi-target situations. In this paper, in order to resolve these limitations, we propose an effective pairing algorithm, which consists of two steps. In the proposed method, a waveform with different slopes in two periods is used. In the 1st pairing processing, all combinations of range and velocity are obtained in each of two wave periods. In the 2nd pairing step, using the results of the 1st pairing processing, fine range and velocity are detected. In that case, we propose the range-velocity windowing technique in order to compensate for the non-ideal beat-frequency characteristic that arises due to the non-linearity of the RF module. Based on experimental results, the performance of the proposed algorithm is improved compared with that of the typical method.

The detection and monitoring of subsurface excavations has a variety of applications in both the civil and defense domains. We have developed a novel InSAR method (Homogenous Distributed Scatterer (HDS)-InSAR) that exploits both persistent point and coherent distributed scatterers by using adaptive multilooking of statistically homogenous pixel neighborhoods. In order to enhance the detection of small scale structures in low SNR environments a matched parametric spatio-temporal model is fit to the deformation signal. We illustrate the performance of our new method for the city of Vancouver over the last nine years using InSAR stacks of RADARSAT-1 and RADARSAT-2 data.

The paper presents the test results of a mobile system for the protection of large-area objects, which consists of a radar and thermal and visual cameras. Radar is used for early detection and localization of an intruder and the cameras with narrow field of view are used for identification and tracking of a moving object. The range evaluation of an integrated system is presented as well as the probability of human detection as a function of the distance from radar-camera unit.

Stereo matching is a technique of finding the disparity map or correspondence points between two images acquired from different sensor positions; it is a core process in stereoscopy. Automatic stereo processing, which involves stereo matching, is an important process in many applications including vision-based obstacle avoidance for unmanned aerial vehicles (UAVs), extraction of weak targets in clutter, and automatic target detection. Due to its high computational complexity, stereo matching algorithms are one of the most heavily investigated topics in computer vision. Stereo image pairs captured under real conditions, in contrast to those captured under controlled conditions are expected to be different from each other in aspects such as scale, rotation, radiometric differences, and noise. These factors contribute to and enhance the level of difficulty of efficient and accurate stereo matching. In this paper we evaluate the effectiveness of cost functions based on Normalized Cross Correlation (NCC) and Zero mean Normalized Cross Correlation (ZNCC) on images containing speckle noise, differences in level of illumination, and both of these. This is achieved via experiments in which these cost functions are employed by a fast version of an existing modern algorithm, the graph-cut algorithm, to perform stereo matching on 24 image pairs. Stereo matching performance is evaluated in terms of execution time and the quality of the generated output measured in terms of two types of Root Mean Square (RMS) error of the disparity maps generated.

Recently, frequency modulated continuous wave (FMCW) technique has drawn a lot of attention in various applications where the high resolution performance is needed due to its cost-eectiveness and low complexity as well as the high resolution performance. One of degradation factors in the technique is the characteristics of nonlinear phase distortion in transmitted waveform. The phase distortion degrades the resolution performance, that is, the contrast and resolution of the obtained range prole or image can be degraded. Especially, as the system with FMCW technique requires higher resolution performance and longer range coverage, the degradation problem becomes more severe hence it can be limited to be utilized for long-range applications like synthetic aperture radar (SAR). This paper proposes the novel algorithms to estimate the nonlinear phase distortion without any expensive devices but only one reference delay line and provides a favorable condition for parallel processing in the algorithms. The estimate of the distortion can be utilized for designing predistortion to compensate it. Simulation result obtained with an arbitrarily generated nonlinear phase distortion, demonstrates that the proposed scheme has outstanding estimation performance.

The objective of the present investigation is to use radar data to detect targets situated on or under a road surface, and, at the same time, minimize the number of false alarms. The data used here have been collected by the Army Research Laboratory (ARL) Synchronous Impulse Reconstruction (SIRE) Radar. These data have been processed at different ranges from the radar, at different depression angles, and with different resolution. This has been achieved by integrating the data collected during the forward motion of the radar along the road. As a result, it has been possible to produce a series of images of the road in front of the radar at progressively better resolution. We show how the exploitation of the different behavior of clutter and targets at different resolution allows higher rates of target detection at lower false alarm rate than otherwise possible.

The Army Research Laboratory (ARL) has, in the past, demonstrated the effectiveness of low frequency, ultrawideband radar for detection of slow-moving targets located behind walls. While these initial results were promising, they also indicated that sidelobe artifacts produced by moving target indication (MTI) processing could pose serious problems. Such artifacts induced false alarms and necessitated the introduction of a tracker stage to eliminate them. Of course, the tracker algorithm was also imperfect, and it tended to pass any persistent, nearly collocated false alarms. In this work we describe the incorporation of a sidelobe-reduction technique-the randomized linear receiver array (RA)-into our MTI processing chain. To perform this investigation, we leverage data collected by ARL's synchronous impulse reconstruction (SIRE) radar. We begin by calculating MTI imagery using both the non-random and randomized array methods. We then compare the sidelobe levels in each image and quantify the differences. Finally, we apply a local-contrast target detection algorithm based on constant false alarm rate (CFAR) principles, and we analyze probabilities of detection and false alarm for each MTI image.

General Atomics Aeronautical Systems Inc. (GA-ASI) participated in the joint naval experiment Trident Warrior 2011 at Combat Direction Systems Activity (CDSA), Dam Neck, Va., in July 2011. The goal was to introduce the Lynx® Multi-Mode Radar's new Maritime Wide Area Search (MWAS) mode and display a viable Unmanned Aircraft System (UAS) full kill chain solution for the naval environment. GA-ASI presented a manned platform, a Beechcraft Super King Air 200 modified with an operators console, Lynx Multi-mode Radar, FLIR Star SAFIRE 380-HD EO/IR camera system, and an L-3 TCDL (aircraft data link system) as a surrogate for the Predator® B/ MQ-9 UAS.

Synthetic aperture radar (SAR) images processed using the polar format algorithm (PFA) may exhibit distortion if the curvature of the spherical wavefronts are not accounted for. The distortion manifests in geometric shifts and defocusing of targets, and intensifies as distances between pixels and the scene reference position increase. In this work, we demonstrate a method to mitigate the effects of wavefront curvature by applying localized (space-variant) phase corrections to sub-regions selected from the polar format processed image. The modified sub-images are then reassembled into a full image. To minimize discontinuities in the reconstructed image, the spatially variant phase adjustments are made to regions larger than the sub-images, and pared down before being reinserted into the complete image. The result is a SAR process that retains the efficiency of the PFA, yet avoids scene size limitations due to wavefront curvature distortions. The method is illustrated and validated using simulations and real data collected by the General Atomics Aeronautical Systems, Inc. (GA-ASI) Lynx® Multi-mode Radar System.