This PDF file contains the front matter associated with SPIE Proceedings Volume 9077, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Real-time digital implementation of three-dimensional (3-D) infinite impulse response (IIR) beam filters are discussed. The 3-D IIR filter building blocks have filter coefficients, which are defined using algebraic closed-form expressions that are functions of desired beam personalities, such as the look-direction of the aperture, the bandwidth and sampling frequency of interest, inter antenna spacing, and 3dB beam size. Real-time steering of such 3-D beam filters are obtained by proposed calculation of filter coefficients. Application specific computing units for rapidly calculating the 3-D IIR filter coefficients at nanosecond speed potentially allows fast real-time tracking of low radar cross section (RCS) objects at close range. Proposed design consists of 3-D IIR beam filter with 4 4 antenna grid and the filter coefficient generation block in separate FPGAs. The hardware is designed and co-simulated using a Xilinx Virtex-6 XC6VLX240T FPGA. The 3-D filter operates over 90 MHz and filter coefficient computing structure can operate at up to 145 MHz.

Microwave power amplifiers often operate in the nonlinear region to maximize efficiency. However, such operation inevitably produces significant harmonics at the output, thereby degrading the performance of the microwave systems. An automated method for canceling harmonics generated by a power amplifier is presented in this paper. Automated tuning is demonstrated over 400 MHz of bandwidth with a minimum cancellation of 110 dB. The intended application for the harmonic cancellation is to create a linear radar transmitter for the remote detection of non-linear targets. The signal emitted from the non-linear targets is often very weak. High transmitter linearization is required to prevent the harmonics generated by the radar itself from masking this weak signal.

Carbon Fiber Composite (CFC) materials have been used for decades in the aerospace, automotive, and naval industries. They have often been used because of their mechanical advantages. These advantageous characteristics have typically included low weight and high strength. It is also a benefit that CFC materials can be made into nearly any shape or size. With the abundant use of CFC materials, it seems desirable to better under- stand the electromagnetic applications of these materials. CFC materials consist of a non-conductive resin or epoxy in addition to conductive carbon fibers. The carbon fibers can be oriented and layered in many different configurations. The specific orientation and layering of the carbon fibers has a direct impact on its electrical characteristics. One specific characteristic of interest is the conductivity of CFC materials. The work in this paper deals with probing the conductivity characteristics of CFC materials for applications in antenna and radar design. Multiple layouts of carbon fiber are investigated. The DC conductivity was measured by applying a conductive epoxy to sample edges and using a milliohm meter. Shielding effectiveness was then predicted based on fundamental electromagnetics for conducting media. Finally, prototype dipole antennas made from CFC materials were investigated.

Vehicular applications of UWB GPR demand multiple GPR sensors operating in a harsh environment. One of the key elements of in the sensor is its UWB antenna which has minimal inter-element coupling, low group delay, high directivity and less prone to environmental conditions. Tapered slot antennas (TSA's) provide good impedance match, but they need to be modified for above specifications. Parasitic slot loaded TSA with balanced feed is proposed and a multi-channel antenna array structure is formed. Structural parameters are numerically analyzed and a prototype is built. Measurements show good performance for UWB GPR applications.

Unmanned aerial systems (UASs) have become a critical asset in current battlespaces and continue to play an increasing role for intelligence, surveillance and reconnaissance (ISR) missions. With the development of medium-to-low altitude, rapidly deployable aircraft platforms, the ISR community has seen an increasing push to develop ISR sensors and systems with real-time mission support capabilities. This paper describes recent flight demonstrations and test results of the RASAR (Real-time, Autonomous, Synthetic Aperture Radar) sensor system. RASAR is a modular, multi-band (L and X) synthetic aperture radar (SAR) imaging sensor designed for self-contained, autonomous, real-time operation with mission flexibility to support a wide range of ISR needs within the size, weight and power constraints of Group III UASs. The sensor command and control and real-time image formation processing are designed to allow integration of RASAR into a larger, multi-intelligence system of systems. The multi-intelligence architecture and a demonstration of real-time autonomous cross-cueing of a separate optical sensor will be presented.

Compressive sensing has emerged as a topic of great interest for radar applications requiring large amounts of data storage. Typically, full sets of data are collected at the Nyquist rate only to be compressed at some later point, where information-bearing data are retained and inconsequential data are discarded. However, under sparse conditions, it is possible to collect data at random sampling intervals less than the Nyquist rate and still gather enough meaningful data for accurate signal reconstruction. In this paper, we employ sparse sampling techniques in the recording of digital microwave holograms over a two-dimensional scanning aperture. Using a simple and fast non-linear interpolation scheme prior to image reconstruction, we show that the reconstituted image quality is well-retained with limited perceptual loss.

The Army Research Laboratory (ARL) has developed an impulse-based vehicle-mounted forward-looking ultra- wideband (UWB) radar for imaging buried landmines and improvised explosive devices (IEDs). However, there is no control of the radiated spectrum in this system. As part of ARL's Partnerships in Research Transition (PIRT) program, the above deficiency is addressed by the design of a Stepped-Frequency Radar (SFR) which allows for precise control over the radiated spectrum, while still maintaining an effective ultra-wide bandwidth. The SFR utilizes a frequency synthesizer which can be configured to excise prohibited and interfering frequency bands and also implement frequency-hopping capabilities. The SFR is designed to be a forward-looking ground- penetrating (FLGPR) Radar utilizing a uniform linear array of sixteen (16) Vivaldi notch receive antennas and two (2) Quad-ridge horn transmit antennas. While a preliminary SFR consisting of four (4) receive channels has been designed, this paper describes major improvements to the system, and an analysis of expected system performance. The 4-channel system will be used to validate the SFR design which will eventually be augmented in to the full 16-channel system. The SFR has an operating frequency band which ranges from 300 - 2000 MHz, and a minimum frequency step-size of 1 MHz. The radar system is capable of illuminating range swaths that have maximum extents of 30 to 150 meters (programmable). The transmitter has the ability to produce approximately -2 dBm/MHz average power over the entire operating frequency range. The SFR will be used to determine the practicality of detecting and classifying buried and concealed landmines and IEDs from safe stand-off distances.

We propose a new laboratory method for characterizing synthetic aperture radar (SAR) systems through the use of a synthetic scene generator. Flight tests are the only definitive way to characterize the system level performance of airborne synthetic aperture radar systems. However, due to the expense of flights tests it is beneficial to complete as much testing as possible in a laboratory environment before flight testing is performed. There are many existing tests that are employed to measure the performance of various subsystems in a SAR system, find defective hardware, and indicate design problems that need to be mitigated. However, certain issues can only be found on an integrated system, and laboratory testing at a system level is typically confined to characterizing the impulse response (IPR) of a single point target through the use of an optical delay line. While useful, delay line testing requires running a modified version of real-time image formation code as the delay line does not completely mimic a real target. Ideally, system level tests are performed on unmodified code. On modern SAR systems many algorithms are data driven (e.g., autofocus) and require a substantially more sophisticated data model for testing. We desire to create a complete system test by combining an arbitrary number of point targets and clutter patterns to mimic radar responses from a real scene. This capability enables complete testing of radar systems in a laboratory environment according to prescribed terrain/scene characteristics. This paper presents an overview of the system requirements for a synthetic scene generator. The analysis is limited to SAR systems utilizing chirp waveforms and stretch processing. Furthermore, we derive relationships between IF bandwidth, target position, and the phase history model. A technique to properly compensate for motion pulse to pulse is presented. Finally, our concept is demonstrated with simulation data.

This paper presents a study on the target position and velocity estimation for air-to-air radar. Two types of post-Kalman filtering (post-KF) processing methods, weighted least-square (WLS) estimators and tracking filters, are investigated to improve the filtering accuracy and thus to improve the target position and velocity estimation accuracy. The WLS estimators are considered up to the second order in order to effectively handle target acceleration. For this type of post- KF processing, the KF outputs within a sliding window are passed through a WLS estimator to refine the estimates of current target position and velocity and their variances. For the tracking filter method, an alpha-beta (α-β) filter and an alpha-beta-gamma (α-β-γ) filter are utilized; both dynamically smooth the KF outputs. For the linear motion scenarios where the target flies with constant velocity, the first-order WLS estimator and the α-β filter are expected to be a good fit, which permits accurate projection of the target position to a future time of interest. The second-order WLS estimator and the α-β-γ filter are capable of handling more general scenarios where the targets may be accelerating. The performance and effectiveness of these proposed post-KF processing methods are demonstrated by using Monte Carlo simulations.

In general, interpretation of signatures from synthetic aperture radar (SAR) data is a challenging task even for the expert image analyst. For the most part, this is caused by radar specific imaging effects, e.g. layover, multi-path propagation or speckle noise. Specifically for the application in maritime security, ship signatures exhibit additional defocusing effects due to the ship’s movement even when they are anchored. Focusing on object recognition, the detection of target signatures can be done with a pretty good chance of success, but the identification is often impossible. To assist image analysts in their recognition tasks, a SAR simulation tool has been developed recently. It is very simple to operate, by simulating available 3D model data of ships and test the resulting simulated signatures with their real counterpart from SAR images. This is a very robust way to identify larger vessels out of current one meter resolution space borne SAR data. Nevertheless, for smaller vessels this can be still very challenging, because the resolution is too coarse. Recently, TerraSAR-X initiated a new staring spotlight imaging mode that enhances cross-range resolution significantly and therefore also improves the chance for the identification of smaller vessels. This paper demonstrates the capabilities of the developed simulation tool in assisted target recognition specifically on ship signatures. The improvement of recognition performance will be studied by comparing results for TerraSAR-X sliding spotlight mode and staring spotlight mode data.

In this paper, we apply a time and frequency analysis method based on the complete ensemble empirical mode decomposition (CEEMD) in GPR signal processing. It decomposes the GPR signal into a sum of oscillatory components, with guaranteed positive and smoothly varying instantaneous frequencies. The key idea of this method relies on averaging the modes obtained by EMD applied to several realizations of Gaussian white noise added to the original signal. It can solve the mode mixing problem in empirical mode decomposition (EMD) method and improve the resolution of ensemble empirical mode decomposition (EEMD) when the signal has low signal noise ratio (SNR). First, we analyze the difference between the basic theory of EMD, EEMD and CEEMD. Then, we compare the time and frequency analysis results of different methods. The synthetic and real GPR data demonstrate that CEEMD promises higher spectral-spatial resolution than the other two EMDs method. Its decomposition is complete, with a numerically negligible error.

In this paper, spectrum sensing techniques are explored for nonlinear radar. These techniques use energy detection to identify an unoccupied receive frequency for nonlinear radar. A frequency is considered unoccupied if it satisfies the following criteria: 1) for a given frequency of interest, its energy must be below a predetermined threshold; 2) the surrounding energy of this frequency must also be below a predetermined threshold. Two energy detection techniques are used to select an unoccupied frequency. The first technique requires the fast Fourier transform and a weighting function to test the energy in neighboring frequency bins; both of these procedures may require a high degree of computational resources. The second technique uses multirate digital signal processing and the fast binary search techniques to lower the overall computational complexity while satisfying the requirements for an unoccupied frequency.

In through-the-wall radar (TTWR) applications, scattering by indoor clutter elements can greatly confound the detection of humans. This paper analyzes the spectral and azimuthal scattering characteristics of various types of individual furniture targets and compares these to humans. Radar cross section (RCS) values of furniture and humans are obtained using the finite difference time domain (FDTD) technique over the 1-5 GHz frequency range and the 0-360 azimuth angle range for both co- and cross-polarized scattering. In the case of furniture, RCS responses show to the highest returns when viewing the planar surfaces of the clutter objects. Objects consisting primarily of smaller planar surfaces with more complex geometrical features showed a more uniform response with no preferred orientation showing a larger response. Human RCS produced from the biological models showed a more constant RCS when viewing the co-polarized response, where the back produced the highest response due to the more planar surface. The cross-polarized response was more varied providing for a wider range of RCS values.

Feature-based methods have been recently considered in the literature for detection of stationary human targets in through-the-wall radar imagery. Specifically, textural features, such as contrast, correlation, energy, entropy, and homogeneity, have been extracted from gray-level co-occurrence matrices (GLCMs) to aid in discriminating the true targets from multipath ghosts and clutter that closely mimic the target in size and intensity. In this paper, we address the task of feature selection to identify the relevant subset of features in the GLCM domain, while discarding those that are either redundant or confusing, thereby improving the performance of feature-based scheme to distinguish between targets and ghosts/clutter. We apply a Decision Tree algorithm to find the optimal combination of co-occurrence based textural features for the problem at hand. We employ a K-Nearest Neighbor classifier to evaluate the performance of the optimal textural feature based scheme in terms of its target and ghost/clutter discrimination capability and use real-data collected with the vehicle-borne multi-channel through-the-wall radar imaging system by Defence Research and Development Canada. For the specific data analyzed, it is shown that the identified dominant features yield a higher classification accuracy, with lower number of false alarms and missed detections, compared to the full GLCM based feature set.

We develop an efficient iterative approach to the solution of the discrete three-dimensional Helmholtz equation with variable coefficients and PML boundary conditions based on compact fourth and sixth order approximation schemes. The coefficient matrices of the resulting systems are not Hermitian and possess positive as well as negative eigenvalues so represent a significant challenge for constructing an efficient iterative solver. In our approach these systems are solved by a combination of a Krylov subspace-type method with a matching high order approximation preconditioner with coefficients depending only on one spatial variable. In the algorithms considered, the direct solution of high order preconditioning system is based on a combination of the separation of variables technique and Fast Fourier Transform (FFT) type methods. The resulting numerical methods allow for efficient implementation on parallel computers. Numerical results confirm the high efficiency of the proposed iterative algorithms.

The past two decades has witnessed a renaissance in passive radar research. One of the areas of research in passive radar that has received recent attention is the use of reflected GNSS signals as the signal-of-opportunity for bistatic synthetic aperture radar (BSAR), known as space-surface BSAR (SS-BSAR) [1-7]. SS-BSAR is unique because it uses GNSS signals, which, in the case of the US owned and operated Global Positioning System (GPS), provide almost constant coverage to almost the entire earth [8-10]. Furthermore, the GPS satellites transmit left and right-hand circularly polarized signals combined during transmission to form a right-hand circularly polarized (RHCP) signal; the benefits being, when compared to horizontal or vertical polarized waveforms, is the signal reflection re-radiates in the opposite, or left-hand circularly polarized (LHCP), polarization with signal loss ranging from 15 to 20 dB [5, 10]. One major drawback to using GPS as the signal-of-opportunity is that the received signal level is extremely low, and lower when reflected (see Table 1).

The traditional radar RF transceivers, similar to communication transceivers, have the basic elements such as baseband waveform processing, IF/RF up-down conversion, transmitter power circuits, receiver front-ends, and antennas, which are shown in the upper half of Figure 1. For modern radars with diversified and sophisticated waveforms, we can frequently observe that the transceiver behaviors, especially nonlinear behaviors, are depending on the waveform amplitudes, frequency contents and instantaneous phases. Usually, it is a troublesome process to tune an RF transceiver to optimum when different waveforms are used. Another issue arises from the interference caused by the waveforms - for example, the range side-lobe (RSL) caused by the waveforms, once the signals pass through the entire transceiver chain, may be further increased due to distortions. This study is inspired by the two existing solutions from commercial communication industry, digital pre-distortion (DPD) and adaptive channel estimation and Interference Mitigation (AIM), while combining these technologies into a single chip or board that can be inserted into the existing transceiver system. This device is then named RF Transceiver Optimizer (RTO). The lower half of Figure 1 shows the basic element of RTO. With RTO, the digital baseband processing does not need to take into account the transceiver performance with diversified waveforms, such as the transmitter efficiency and chain distortion (and the intermodulation products caused by distortions). Neither does it need to concern the pulse compression (or correlation receiver) process and the related mitigation. The focus is simply the information about the ground truth carried by the main peak of correlation receiver outputs. RTO can be considered as an extension of the existing calibration process, while it has the benefits of automatic, adaptive and universal. Currently, the main techniques to implement the RTO are the digital pre- or –post distortions (DPD), and the main technique to implement the AIM is the Adaptive Pulse Compression (APC). The basic algorithms and experiments with DPD will be introduced which is also the focus of this paper. The discussion of AIM algorithms will be presented in other papers, while the initial implementation of AIM and correlation receiver in FPGA devices will also be introduced in this paper.

In this paper, we show that a single transmission of a random noise waveform may not sufficient to obtain a successful tomographic image of an object. In order to overcome this shortcoming, multiple independent and identically distributed (iid) random noise waveforms over a frequency range from 8 to 10 GHz are transmitted to reconstruct the final image of various objects. Diffraction tomography theorem is applied for each noise waveform transmission so that the image of the multiple objects is reconstructed based on the backward scattered field at the end of each noise waveform transmission realization. After all iid noise waveforms are transmitted, the final tomographic image of the target is reconstructed by averaging all obtained images from multiple transmissions. Several numerical simulations in the spatial frequency domain are performed, and the successful tomographic image of the multiple cylindrical PEC objects is achieved after transmission of multiple iid ultra-wideband (UWB) random noise waveforms.

This work demonstrates the feasibility of using the advanced pulse compression noise (APCN) radar waveform for synthetic aperture radar (SAR). Using a simple image formation process (IFP), we not only show that we can successfully form images using the APCN waveform, but we grow our understanding of how different combinations of APCN waveforms and side lobe weighting functions impact SAR image quality. In this paper, an analysis is presented that compares the target range point spread function (PSF) for several simulated SAR images.

Super-resolution (SR) is a radar processing technique closely related to the pulse compression (or correlation receiver). There are many super-resolution algorithms developed for the improved range resolution and reduced sidelobe contaminations. Traditionally, the waveforms used for the SR have been either phase-coding (such as LKP3 code, Barker code) or the frequency modulation (chirp, or nonlinear frequency modulation). There are, however, an important class of waveforms which are either random in nature (such as random noise waveform), or randomly modulated for multiple function operations (such as the ADS-B radar signals in [1]). These waveforms have the advantages of low-probability-of-intercept (LPI). If the existing SR techniques can be applied to these waveforms, there will be much more flexibility for using these waveforms in actual sensing missions. Also, SR usually has great advantage that the final output (as estimation of ground truth) is largely independent of the waveform. Such benefits are attractive to many important primary radar applications. In this paper the general introduction of the SR algorithms are provided first, and some implementation considerations are discussed. The selected algorithms are applied to the typical LPI waveforms, and the results are discussed. It is observed that SR algorithms can be reliably used for LPI waveforms, on the other hand, practical considerations should be kept in mind in order to obtain the optimal estimation results.

Displaced Phase Center Antenna (DPCA) and Space-Time Adaptive Processing (STAP) are two general methods to cancel clutter in order to detect small, slowly moving targets that may be obscured by clutter. To detect these targets, the radar detection threshold needs to be as low as possible to ensure some minimum probability of detection (Pd). Unfortunately, lowering the radar threshold naturally results in a higher false alarm rate. Although there are standard methods such as M of N to reduce the false alarms, new techniques can potentially drive the false alarm rate down even further. Many “theoretical” papers have shown that STAP can be designed to outperform DPCA because of its potential additional “degrees-of-freedom”. However, in “practice,” this isn’t always the case. For example, it is well known that STAP can have training issues in heterogeneous clutter. Typically, a radar signal processor will implement one method or the other to detect these small endoclutter targets. The technique being explored here is a two-fold approach in which the existing STAP code first processes the data in order to find a list of candidate targets. Next, a DPCA technique is also used to find a separate list of candidate detections from the same data. Although the algorithms are working on the same data, the processing is “independent” between them so the target lists are different. After both techniques have finished processing, the modified radar signal processing code “intelligently” combines the two detection lists into a single detection list. It will be shown that the combined list of detections from the two methods results in better detection performance than either method used separately.

Multiple-input multiple-output (MIMO) radar utilizes the flexible configuration of transmitting and receiving antennas to construct images of target scenes. Because of the target scenes' sparsity, the compressive sensing (CS) technique can be used to realize a feasible reconstruction of the target scenes from undersampling data. This paper presents the signal model of MIMO radar and derive the corresponding CS measurement matrix, which shows success of the CS technique. Also the basis pursuit method and total-variation minimization method are adopted for different scenes' recovery. Numerical simulations are provided to illustrate the validity of reconstruction for one dimensional and two dimensional scenes.

The Army Research Laboratory (ARL) is developing an indoor experimental facility to evaluate and assess airborne synthetic-aperture-radar-(SAR)-based detection capabilities. The rail-SAR is located in a multi-use facility that also provides a base for research and development in the area of autonomous robotic navigation. Radar explosive hazard detection is one key sensordevelopment area to be investigated at this indoor facility. In particular, the mostly wooden, multi-story building houses a two (2) story housing structure and an open area built over a large sandbox. The housing structure includes reconfigurable indoor walls which enable the realization of multiple See-Through-The-Wall (STTW) scenarios. The open sandbox, on the other hand, allows for surface and buried explosive hazard scenarios. The indoor facility is not rated for true explosive hazard materials so all targets will need to be inert and contain surrogate explosive fills. In this paper we discuss the current system status and describe data collection exercises conducted using canonical targets and frequencies that may be of interest to designers of ultra-wideband (UWB) airborne, ground penetrating SAR systems. A bi-static antenna configuration will be used to investigate the effects of varying airborne SAR parameters such as depression angle, bandwidth, and integration angle, for various target types and deployment scenarios. Canonical targets data were used to evaluate overall facility capabilities and limitations. These data is analyzed and summarized for future evaluations. Finally, processing techniques for dealing with RF multi-path and RFI due to operating inside the indoor facility are described in detail. Discussion of this facility and its capabilities and limitations will provide the explosive hazard community with a great airborne platform asset for sensor to target assessment.

Sandia National Laboratories produces copious amounts of high-resolution, single-polarization Synthetic Aperture Radar (SAR) imagery, much more than available researchers and analysts can examine. Automating the recognition of terrains and structures in SAR imagery is highly desired. The optical image processing community has shown that superpixel segmentation (SPS) algorithms divide an image into small compact regions of similar intensity. Applying these SPS algorithms to optical images can reduce image complexity, enhance statistical characterization and improve segmentation and categorization of scene objects. SPS algorithms typically require high SNR (signal-to-noise-ratio) images to define segment boundaries accurately. Unfortunately, SAR imagery contains speckle, a product of coherent image formation, which complicates the extraction of superpixel segments and could preclude their use. Some researchers have developed modified SPS algorithms that discount speckle for application to SAR imagery. We apply two widely-used SPS algorithms to speckle-reduced SAR image products, both single SAR products and combinations of multiple SAR products, which include both single polarization and multi-polarization SAR images. To evaluate the quality of resulting superpixels, we compute research-standard segmentation quality measures on the match between superpixels and hand-labeled ground-truth, as well as statistical characterization of the radar-cross-section within each superpixel. Results of this quality analysis determine the best input/algorithm/parameter set for SAR imagery. Simple Linear Iterative Clustering provides faster computation time, superpixels that conform to scene-relevant structures, direct control of average superpixel size and more uniform superpixel sizes for improved statistical estimation which will facilitate subsequent terrain/structure categorization and segmentation into scene-relevant regions.

Screening crowds for threats requires a stand-off sensor with wide area coverage, high spatial resolution and a high temporal update rate. We have assessed the capability of the NIRAD high speed 94 GHz FMCW surveillance radar against this requirement. NIRAD’s sub-degree beamwidth, 25 cm range bins and 10 Hz azimuthal frame rate yield high resolution radar videos of scenes over ranges from tens to hundreds of meters, capable of tracking people walking or running around the scene. We present how people are detected and tracked in the scene to enable analysis of their radar cross section images to reveal signatures which may indicate the presence of a carried threat item.

Quantum radar serves to drastically improve the resolution of current radar technology using quantum phenomena. This paper will first review some of the proposed ideas and engineering designs behind both entanglement radar and coherent state radar design schemes. Entanglement radar is based on first entangling two photons, then sending one of the entangled photons out towards the target, and keeping the other one at home. A correlation between the two photons is analyzed to obtain information. Coherent state quantum radar relies on using coherent state photons and a quantum detection scheme in order to beat the diffraction limit. Based on the above, a proposed design concept to implement of a coherent state quantum radar is presented for simultaneously determining target range and azimuth/elevation angles.

RF electronic targets cannot be detected by traditional linear radar because their radar cross sections are much smaller than that of nearby clutter. One technology that is capable of separating RF electronic targets from clutter, however, is nonlinear radar. Presented in this paper is a combination of stepped-frequency ultra-wideband radar with nonlinear detection. By stepping the transmit frequency across an ultra-wide bandwidth and recording the amplitude and phase of the harmonic return signal, a nonlinear frequency response of the radar environment is constructed. An inverse Fourier transform of this response reveals the range to a nonlinear target.

Monitoring seasonal snow accumulation is important for evaluation of snow models, for short- and long-term snow cover monitoring, and for both military and civilian activities in cold climates. Improved spatial analysis of snow depth and volume can help decision makers plan for future events and mitigate risk. Current snow depth measurement methods fall short of operational requirements. This research explored a new approach for determining snow depth using Ku-band multi-pass (monostatic) airborne interferometric synthetic aperture radar (InSAR). A perturbation method that isolated and compared high frequency terrain phase to elevation was used to generate Snow-Off and Snow-On DEMs from the InSAR phase data. Differencing the InSAR DEMs determined elevation change caused by accumulated snow. Comparison of InSAR snow depths to manual snow depth measurements indicated average InSAR snow depth errors of -8cm, 95cm, -49cm, 176cm, 87cm, and 42cm for six SAR pairs. The source of these errors appears to be mostly related to uncorrected slope and tilt in fitted low frequency planes. Results show that this technique has promise but accuracy could be substantially improved by the use of bistatic SAR systems, which would allow for more stable and measurable interferometric baselines.

Crohn's disease is a condition that causes inflammation and associated complications along any section of the digestive tract. Over the years, numerous radiological and endoscopic methods as well as the use of ultrasound have been developed to examine and diagnose inflammatory bowel disorders such as Crohn's disease. While such techniques have much merit, an alternative medical solution that is safe, non-invasive, and inexpensive is proposed in this paper. Reflections from electromagnetic signals transmitted by an ultra-wide band (UWB) radar allow for not only range (or extent) information but also spectral analysis of a given target of interest. Moreover, the radar cross-section (RCS) of an object measures how detectable the electromagnetic return energy of such an object is to the radar. In the preliminary phase of research, we investigate how disparities in the dielectric properties of diseased versus non-diseased portions of the intestines can aid in the detection of Crohn's disease. RCS analysis from finite-difference time-domain (FDTD) method simulations using a simple 3D model of the intestines are presented. The ultimate goal of our research is to design a UWB radar system using a suitable waveform to detect and monitor Crohn's disease.

Urinary incontinence is defined as the inability to stop the flow of urine from the bladder. In the US alone, the annual societal cost of incontinence-related care is estimated at 12.6 billion dollars. Clinicians agree that those suffering from urinary incontinence would greatly benefit from a wearable system that could continually monitor the bladder, providing continuous feedback to the patient. While existing ultrasound-based solutions are highly accurate, they are severely limited by form-factor, battery size, cost and ease of use. In this study the authors propose an alternative bladder-state sensing system, based on Ultra Wideband (UWB) Radar. As part of an initial proof-of-concept, the authors developed one of the first dielectrically and anatomically-representative Finite Difference Time Domain models of the pelvis. These models (one male and one female) are derived from Magnetic Resonance images provided by the IT'IS Foundation. These IT'IS models provide the foundation upon which an anatomically-plausible bladder growth model was constructed. The authors employed accurate multi-pole Debye models to simulate the dielectric properties of each of the pelvic tissues. Two-dimensional Finite Difference Time Domain (FDTD) simulations were completed for a range of bladder volumes. Relevant features were extracted from the FDTD-derived signals using Principle Component Analysis (PCA) and then classified using a k-Nearest-Neighbour and Support Vector Machine algorithms (incorporating the Leave-one-out cross-validation approach). Additionally the authors investigated the effects of signal fidelity, noise and antenna movement relative to the target as potential sources of error. The results of this initial study provide strong motivation for further research into this timely application, particularly in the context of an ageing population.

Breast cancer is one of the most common cancers in women. In the United States alone, it accounts for 31% of new cancer cases, and is second only to lung cancer as the leading cause of deaths in American women. More than 184,000 new cases of breast cancer are diagnosed each year resulting in approximately 41,000 deaths. Early detection and intervention is one of the most significant factors in improving the survival rates and quality of life experienced by breast cancer sufferers, since this is the time when treatment is most effective. One of the most promising breast imaging modalities is microwave imaging. The physical basis of active microwave imaging is the dielectric contrast between normal and malignant breast tissue that exists at microwave frequencies. The dielectric contrast is mainly due to the increased water content present in the cancerous tissue. Microwave imaging is non-ionizing, does not require breast compression, is less invasive than X-ray mammography, and is potentially low cost. While several prototype microwave breast imaging systems are currently in various stages of development, the design and fabrication of anatomically and dielectrically representative breast phantoms to evaluate these systems is often problematic. While some existing phantoms are composed of dielectrically representative materials, they rarely accurately represent the shape and size of a typical breast. Conversely, several phantoms have been developed to accurately model the shape of the human breast, but have inappropriate dielectric properties. This study will brie y review existing phantoms before describing the development of a more accurate and practical breast phantom for the evaluation of microwave breast imaging systems.

An assessment of bio-radiolocation monitoring of respiratory rhythm during sleep is given. Full-night respiratory inductance plethysmography (RIP) and bio-radiolocation (BRL) records were collected simultaneously in a sleep laboratory. Polysomnography data from 5 subjects without sleep breathing disorders were used. A multi-frequency bioradar with step frequency modulation was applied. It has 8 operating frequencies ranging from 3.6 to 4.0 GHz. BRL data are recorded in two quadratures. Respiratory cycles were detected in time domain. Obtained data was used for the evaluation of correlation between BRL and RIP respiration rate estimates. Strong correlation between corresponding time series was revealed. BRL method is reliably implemented for estimation of respiratory rhythm and respiratory rate variability during full night sleep.

This paper presents the results of experiments and mathematical simulation carried out to confirm the possibility of using holographic radar for the detection of breast tumors. In the work the software designed for the numerical solution of electromagnetic problems using the Finite-Difference Time-Domain Method. The simulation was performed with the three probe frequencies 4, 7 and 15 GHz. The model is a parallelepiped with dimensions 200×200×100 mm - mimicking the normal tissue of the breast, with the inclusion of a sphere - malignant neoplasm of breast tissue, the radius and depth of which have been varied. Frequency dispersion of normal and malignant tissues dielectric properties (conductivity and permittivity) was taken into account. It was shown both by theoretical and experimental results that it is preferable to use lower-frequency probing signal, namely, 4GHz, which can detect the inclusion of 5 mm diameter up to a depth of 10 mm. While using of probing signals of 7 and 15 GHz the depth limit of detection inclusion is not more than 5 mm, which is caused by the high attenuation in a medium. However, their usage is preferred because of higher resolution.

In this paper the ability of four radar sensors in detecting breath activity has been tested. In particular, range gating UWB, CMOS UWB, CW phase detecting, and FMCW radars have taken into account. Considering a realistic scenario, the radar antenna has been pointed towards the thorax of a breathing subject and the recorded signals have been compared with those of a piezoelectric belt placed around the thorax. Then the ability of the radars in detecting small movements has been tested by means of an oscillating copper plate placed at various distances from the radar antenna. All the considered radars were able to detect the plate movements with a distance-dependent resolution.

Unattended catastrophic falls result in risk to the lives of elderly. There are growing efforts and rising interest in detecting falls of the aging population, especially those living alone. Radar serves as an effective non-intrusive sensor for detecting human activities. For radar to be effective, it is important to achieve low false alarms, i.e., the system can reliably differentiate between a fall and other human activities. In this paper, we discuss the time-scale based signal analysis of the radar returns from a human target. Reliable features are extracted from the scalogram and are used for fall classifications. The classification results and the advantages of using a wavelet transform are discussed.

We analyze the performance of a wideband orthogonal frequency division multiplexing (OFDM) signal in estimating the micro-Doppler frequency of a target having multiple rotating scatterers (e.g., rotor blades of a helicopter, propellers of a submarine). The presence of rotating scatterers introduces Doppler frequency modulation in the received signal by generating sidebands about the transmitted frequencies. This is called the micro-Doppler effects. The use of a frequency-diverse OFDM signal in this context enables us to independently analyze the micro-Doppler characteristics with respect to a set of orthogonal subcarrier frequencies. Therefore, to characterize the accuracy of micro-Doppler frequency estimation, we compute the Cram´er-Rao Bound (CRB) on the angular-velocity estimate of the target while considering the scatterer responses as deterministic but unknown nuisance parameters. Additionally, to improve the accuracy of the estimation procedure, we formulate and solve an optimization problem by minimizing the CRB on the angular-velocity estimate with respect to the transmitting OFDM spectral coefficients. We present several numerical examples to demonstrate the CRB variations at different values of the signal-to-noise ratio (SNR) and the number of OFDM subcarriers. The CRB values not only decrease with the increase in the SNR values, but also reduce as we increase the number of subcarriers implying the significance of frequency-diverse OFDM waveforms. The improvement in estimation accuracy due to the adaptive waveform design is also numerically analyzed. Interestingly, we find that the relative decrease in the CRBs on the angular-velocity estimate is more pronounced for larger number of OFDM subcarriers.

The large utility-scale wind turbines are reported to have negative impact on nearby radars due to complex scattering mechanisms, which is usually referred to as the radar Wind Turbine Clutter (WTC). Extremely complicated time-varying Doppler spectrum have been observed. Conventional ground clutter filter techniques thus have failed in mitigating the non-stationary components in the frequency domain. Rotation of the blades is a micro-motion as the wind turbine always stays at the same location. The time-evolving spectrum associated with the blade rotation is therefore a Micro-Doppler signature, which is important in characterizing radar WTC. This paper will disclose some latest findings from our recent studies in characterizing the Micro-Doppler radar signatures of wind turbine through electromagnetic modeling.

Micro-range Micro-Doppler can be used to isolate particular parts of the radar signature, and in this case we demonstrate the differences in the signature between a walking horse versus a walking horse with a rider. Using micro-range micro-Doppler, we can distinguish the radar returns from the rider as separate from the radar returns of the horse.

In this paper, we present the results of a modeling study to examine the interference effect of microDopplers caused by offshore wind farms on airborne sensors operating in the synthetic aperture radar (SAR) and ground moving target indicator (GMTI) modes. The modeling is carried out by generating CAD instantiations of the dynamic wind turbine and using the high-frequency electromagnetic code Xpatch to perform the scattering calculations. Artifacts in the resulting SAR and GMTI signatures are evaluated for interference with tracking of boats in coastal waters. Results of signal filtering algorithms to reduce the dynamic turbine clutter in both SAR images and GMTI displays are presented.

The detection of small unmanned aerial vehicles (UAVs) using radar can be challenging due to the small radar cross section and the presence of false targets such as birds. We present the initial results of micro-Doppler radar data collected on a small helicopter at G-band and compare the results to previously measured birds. The resulting signature differences can be used to help discriminate small UAVs from naturally occurring moving clutter such as birds.

In this paper, we investigate the use of software defined radar (SDR) to analyze the micro-Doppler signatures. The first SDR we use is based on the Universal Software Radio Peripheral (USRP) and GNU Radio, and another SDR which has several operation modes is based on field-programmable gate arrays (FPGA). Typically, the USRP-based SDR is not optimized for radar applications due to its narrow bandwidth and time-varying additional delay caused by USRP components and operating system. The FPGA-based SDR is more suitable for applications where high-resolution range information is required. Our studies indicate that both of the SDR systems are capable of producing the micro-Doppler signatures. System design challenges and measurement results will be discussed in detail.

Micro-Doppler and vibrometry measurements require coherent radars with low phase noise. We report the development of a novel, very low phase noise 94 GHz radar, called T-220, which offers superior performance for micro-Doppler and vibrometry studies compared with our previous work. The radar uses a combination of direct digital synthesis (DDS) chirp generation, frequency upconversion and frequency multiplication to yield very low phase noise and rapid, contiguous chirps, necessary for Doppler studies and other coherent processing applications. Dual fan beam antennas are used to achieve negligible transmit-receive leakage, with fine azimuth resolution and modest elevation coverage. The resulting PPI imagery is very high fidelity with little or no evidence of phase noise effects.

In recent years, the automotive industry has experienced an evolution toward more powerful driver assistance systems that provide enhanced vehicle safety. These systems typically operate in the optical and microwave regions of the electromagnetic spectrum and have demonstrated high efficiency in collision and risk avoidance. Microwave radar systems are particularly relevant due to their operational robustness under adverse weather or illumination conditions. Our objective is to study different signal processing techniques suitable for extraction of accurate micro-Doppler signatures of slow moving objects in dense urban environments. Selection of the appropriate signal processing technique is crucial for the extraction of accurate micro-Doppler signatures that will lead to better results in a radar classifier system. For this purpose, we perform simulations of typical radar detection responses in common driving situations and conduct the analysis with several signal processing algorithms, including short time Fourier Transform, continuous wavelet or Kernel based analysis methods. We take into account factors such as the relative movement between the host vehicle and the target, and the non-stationary nature of the target’s movement. A comparison of results reveals that short time Fourier Transform would be the best approach for detection and tracking purposes, while the continuous wavelet would be the best suited for classification purposes.

An occluded or dark region in synthetic aperture radar (SAR) imagery, known as a shadow, is created when incident radar energy is obstructed by a target with height from illuminating resolution cells immediately behind the target in the ground plane. Shadows depend on the physical dimensions and mobility of a target, platform and radar imaging parameters, and scene clutter. Target shadow dimensions and intensity can be important radar observables in SAR imagery for target detection, location, and tracking or even identification. Stationary target shadows can provide insight as to the physical dimensions of a target, while moving target shadows may show more accurately the location and motion of the target over time versus Doppler energy which may be shifted or smeared outside the scene. However, SAR shadows prove difficult to capture as a target or platform moves, since the quality of the no-return area may quickly be washed-out in a scene over many clutter resolution cells during an aperture. Prior work in the literature has been limited to describing partial shadow degradation effects from platform or target motion of vehicles such as static target shadow tip or interior degradation during an aperture, or shadow degradation due to target motion solely in cross-range. In this paper, we provide a more general formulation of SAR shadow dimensions and intensity for non-specific targets with an arbitrary motion.

The development of sensors that are capable of penetrating smoke, dust, fog, clouds, and rain is critical for maintaining situational awareness in degraded visual environments and for providing support to the Warfighter. Atmospheric penetration properties, the ability to form high-resolution imagery with modest apertures, and available source power make the extremely high-frequency (EHF) portion of the spectrum promising for the development of radio frequency (RF) sensors capable of penetrating visual obscurants. Comprehensive phenomenology studies including polarization and backscatter properties of relevant targets are lacking at these frequencies. The Army Research Laboratory (ARL) is developing a fully-polarimetric frequency-modulated continuous-wave (FMCW) instrumentation radar to explore polarization and backscatter properties of in-situ rain, scattering from natural and man-made surfaces, and the radar cross section and micro-Doppler signatures of humans at EHF frequencies, specifically, around the 220 GHz atmospheric window. This work presents an overview of the design and construction of the radar system, hardware performance, data acquisition software, and initial results including an analysis of human micro-Doppler signatures.

We propose a method to image a complex scene with spotlight synthetic aperture radar (SAR) despite the presence of multiple moving targets. Many recent methods use sparsity-based reconstruction coupled with phase error corrections of moving targets to reconstruct stationary scenes. However, these methods rely on the assumption that the scene itself is sparse and thus unfortunately cannot handle realistic SAR scenarios with complex backgrounds consisting of more than just a few point targets. Our method makes use of sparse and low-rank (SLR) matrix decomposition, an efficient method for decomposing a low-rank matrix and sparse matrix from their sum. For detecting the moving targets and reconstructing the stationary background, SLR uses a convex optimization model that penalizes the nuclear norm of the low rank background structure and the L1 norm of the sparse moving targets. We propose an L1-norm regularization reconstruction method to form the input data matrix, which is grossly corrupted by the moving targets. Each column of the input matrix is a reconstructed SAR image with measurements from a small number of azimuth angles. The use of the L1-norm regularization and a sparse transform permits us to reconstruct the scene with significantly fewer measurements so that moving targets are approximately stationary. We demonstrate our SLR-based approach using simulations adapted from the GOTCHA Volumetric SAR data set. These simulations show that SLR can accurately image multiple moving targets with different individual motions in complex scenes where methods that assume a sparse scene would fail.

Full waveform Lidar systems have the ability of recording the complete signal reflected from the illuminated target. Therefore, more detail information can be obtained compared to conventional Lidar systems. The problem that is faced in using full waveform Lidar is the acquisition of high volume data, a solution proposed to solve this problem is compressive sensing. By using a compressive sensing approach we can reduce the sampling rate and still be able to recover the signal. The reduction is incorporated in the acquisition hardware, where we perform sensing of the signal with compression. In this paper we propose to use a deterministic compressive sensing approach by using a chaotic signal as the sensing matrix. The proposed approach gives the range profile information without the requirement of further processing techniques. For comparison we used two different types of transmitted signals: chaotic and Linear Frequency Modulated (LFM) signals. Simulations demonstrate that chaotic signals give better results than the LFM signals. By using a chaotic signal we can obtain the impulse response of the target by using less than 20 percent of the samples.

Compressive sensing techniques have been widely used to decrease the data acquisition time while generating high-resolution images due to the sparsity of the target space in through-the-wall radar imaging application. The CS-based imaging techniques mainly discretize the continuous target space into grid points and generate a dictionary of model data to form an optimization problem. The choice of the grid for generating the sparsity inducing basis or dictionary is a central point of CS and sparse approximation. However, good sparse recovery performance is based on the assumption that the targets are positioned at the pre-discretized grid locations; otherwise, the performance would significantly degrade. In this paper, the first-order approximation to estimate the targets' off-grid shifts and the joint sparse recovery method are used for reducing the effect of the grid to locate the off-grid target. Numerical examples demonstrate the robust results with lower localization errors using the joint sparse recovery method are obtained for off-grid targets compared to standard sparse reconstruction techniques.

Wideband receive-mode beamforming applications in wireless location, electronically-scanned antennas for radar, RF sensing, microwave imaging and wireless communications require digital aperture arrays that offer a relatively constant far-field beam over several octaves of bandwidth. Several beamforming schemes including the well-known true time-delay and the phased array beamformers have been realized using either finite impulse response (FIR) or fast Fourier transform (FFT) digital filter-sum based techniques. These beamforming algorithms offer the desired selectivity at the cost of a high computational complexity and frequency-dependant far-field array patterns. A novel approach to receiver beamforming is the use of massively parallel 2-D infinite impulse response (IIR) fan filterbanks for the synthesis of relatively frequency independent RF beams at an order of magnitude lower multiplier complexity compared to FFT or FIR filter based conventional algorithms. The 2-D IIR filterbanks demand fast digital processing that can support several octaves of RF bandwidth, fast analog-to-digital converters (ADCs) for RF-to-bits type direct conversion of wideband antenna element signals. Fast digital implementation platforms that can realize high-precision recursive filter structures necessary for real-time beamforming, at RF radio bandwidths, are also desired. We propose a novel technique that combines a passive RF channelizer, multichannel ADC technology, and single-phase massively parallel 2-D IIR digital fan filterbanks, realized at low complexity using FPGA and/or ASIC technology. There exists native support for a larger bandwidth than the maximum clock frequency of the digital implementation technology. We also strive to achieve More-than-Moore throughput by processing a wideband RF signal having content with N-fold (B = N F_clk/2) bandwidth compared to the maximum clock frequency F_clk Hz of the digital VLSI platform under consideration. Such increase in bandwidth is achieved without use of polyphase signal processing or time-interleaved ADC methods. That is, all digital processors operate at the same F_clk clock frequency without phasing, while wideband operation is achieved by sub-sampling of narrower sub-bands at the the RF channelizer outputs.

The U.S. Army Research Laboratory (ARL) has developed the impulse-based, ground vehicle-based, forward-looking ultra-wideband (UWB), synthetic aperture radar (SAR) to detect concealed targets. Although the impulse-based architecture offers its own advantages, one of the important challenges is that when using this architecture it is very difficult to transmit a radar signal with an arbitrary bandwidth and shape. This feature is crucial for the radar to be compliant with the local frequency authority. In addition, being able to transmit signals with an arbitrary spectral shape is an important step in creating the next generation of smart (cognitive) radars. Therefore, we have designed a next-generation prototype radar to take advantage of the stepped frequency architecture. The design and building of the radar hardware is underway. In this paper, we study the radar transmit and acquisition scheme; the trade-offs between SAR image performance and various key radar parameters; and data reconstruction techniques for radar signals with an arbitrary spectrum. This study demonstrates performance, provides some guidelines for the radar design, and serves as a foundation for the signal and image processing stage.

In recent years, a new class of Moving Target Indicator (MTI) radars has emerged, namely those whose mission included detecting moving people, or “dismounts.” This new mode is frequently termed Dismount-MTI, or DMTI. Obviously, detecting people is a harder problem than detecting moving vehicles, necessitating different specifications for performance and hardware quality. Herein we discuss some performance requirements typical of successful DMTI radar modes and systems.

Backprojection has long been applied to SAR image formation. It has equal utility in forming the range-velocity maps for Ground Moving Target Indicator (GMTI) radar processing. In particular, it overcomes the problem of targets migrating through range resolution cells.

A fundamental relationship that is the foundation for all radar is that a target’s range is proportional to an echo delay time. The actual relationship requires knowledge of the velocity of propagation of the signal whose echo delay time is measured. A typical assumption for radar ranging is to use free-space velocity of propagation. However, atmospheric dielectric properties yield a measurably slower velocity of propagation that is a function of temperature, atmospheric pressure, and especially humidity. This results in range measurement errors. A simplified model is developed to estimate the error in range measurements for airborne ground-surveillance radars.

The advent of high-speed large gate-count Field Programmable Gate Array (FPGA) components facilitates the implementation of high-performance parametric digital waveform generators for radar applications. One such waveform is the popular Linear-FM chirp. The state-of-the-art allows us to generate high-fidelity precision wideband Linear-FM chirp waveforms with relative ease, and furthermore enhance these waveforms with a number of features including spectral notches, phase equalization, compound pulses, and more. Design equations are presented as well as a number of feature enhancements.

The paper contains feasibility study of a method for bioradar monitoring of small laboratory animals loco-motor activity improved by using a corner reflector. It presents results of mathematical simulation of bioradar signal reflection from the animal with the help of finite-difference time-domain method. It was proved both by theoretical and experimental results that a corner reflector usage during monitoring of small laboratory animals loco-motor activity improved the effectiveness of the method by reducing the dependency of the power flux density level from the distance between antennas block and the object.

The paper summarizes results of step-frequency radars application in medicine. Remote and non-contact control of physiological parameters with modern bioradars provides a wide range of possibilities for non-contact remote monitoring of a human psycho-emotional state and physiological condition. The paper provides information about technical characteristics of bioradars designed at Bauman Moscow State Technical University and experiments using them. Results of verification experiment showed that bioradars of BioRASCAN type may be used for simultaneous remote measurements of breathing and heart rate parameters. In addition, bioradar assisted experiments for detecting of different sleep disorders are described. Their results proved that method of bioradiolocation allows correct estimation of obstructive sleep apnea severity compared to the polysomnography method, which satisfies standard medical recommendations.

Cardiovascular diseases (CVD) are a major cause of deaths all over the world. Microwave radar can be an alternative sensor for heart diagnostics and monitoring in modern healthcare that aids early detection of CVD symptoms. In this paper measurements from a switch array radar system are presented. This UWB system operates below 3 GHz and does time-lapse imaging of the beating heart inside the human body. The array consists of eight fat dipole elements. With a switch system, every possible sequence of transmit/receive element pairs can be selected to build a radar image from the recordings. To make the radar waves penetrate the human tissue, the antenna array is placed in contact with the body. Removal of the direct signal leakage through the antennas and body surface are done by high-pass (HP) filtering of the data prior to image processing. To analyze the results, measurements of moving spheres in air and simulations are carried out. We see that removal of the direct signal introduces amplitude distortion in the images. In addition, the effect of small target motion between the collection times of data from the individual elements is analyzed. With low pulse repetition frequency (PRF) this motion will distort the image. By using data from real measurements of heart motion in simulations, we analyze how the PRF and the antenna geometry influence this distortions.

The possibility of detecting moving people around the corner using multipath propagation of radar waves has previous been demonstrated in experimental set-ups. Here, we present measurements from a realistic scene using a radar system operating at X-band in a stepped-frequency mode. The moving objects include person(s), bicycle, and car(s). A semi-monostatic single receiver-transmitter radar system was used as a data collector. All detections were made by using Doppler filter processing. We can identify target returns after one and two wall reflections. Two persons moving in the scene can be separated, when allowed by the range resolution.

Enhancing the operational safety of small, maneuverable rotorcraft has been a critical consideration in the development of next generation situational awareness sensor suites. From landing assistance to target detection and obstacle avoidance, millimeter wave radars have become the leading candidate for such solutions due to their ability to operate in degraded visual environments, whether it is weather, induced debris, or night conditions that must be dealt with. Power lines pose arguably the largest safety risk for helicopter operation due to their difficulty in detection and proper identification to support avoidance maneuvering, where even under perfect conditions they can be nearly invisible to the naked eye. The backscatter phenomenology from braided power lines has been well-studied and formulated in previous literature, albeit mainly in controlled laboratory settings or limited field trials. Subsequently, the ability to simply detect power lines at operational distances up to around 2 km has been demonstrated. In this work, an analysis is performed on the measureable characteristics of power lines captured in a representative operational environment for helicopters. The test location included a diverse set of power line configurations with surrounding ground and tower clutter, representing a realistic scenario. A radiometrically calibrated w-band real-beam FMCW sensor allows the study and estimation of target RCS, as well as evaluation against the developed theory. All analysis is performed on dynamically captured data from a helicopter, where platform dynamics and system stability also play a significant role in a processed result. Results from this work will aid the effective development of next generation situational awareness systems.

Diabetic retinopathy is the leading cause of blindness in adults in the United States. The presence of exudates in fundus imagery is the early sign of diabetic retinopathy so detection of these lesions is essential in preventing further ocular damage. In this paper we present a novel technique to automatically detect exudates in fundus imagery that is robust against spatial and temporal variations of background noise. The detection threshold is adjusted dynamically, based on the local noise statics around the pixel under test in order to maintain a pre-determined, constant false alarm rate (CFAR). The CFAR detector is often used to detect bright targets in radar imagery where the background clutter can vary considerably from scene to scene and with angle to the scene. Similarly, the CFAR detector addresses the challenge of detecting exudate lesions in RGB and multispectral fundus imagery where the background clutter often exhibits variations in brightness and texture. These variations present a challenge to common, global thresholding detection algorithms and other methods. Performance of the CFAR algorithm is tested against a publicly available, annotated, diabetic retinopathy database and preliminary testing suggests that performance of the CFAR detector proves to be superior to techniques such as Otsu thresholding.