Publications

Bickel, Douglas L.; Raynal, Ann M.; Dubbert, Dale F.; Doerry, Armin; Burns, Bryan L.

Abstract not provided.

Raynal, Ann M.; Bickel, Douglas L.; Dubbert, Dale F.; Burns, Bryan L.; Doerry, Armin

Abstract not provided.

Doerry, Armin

Abstract not provided.

Doerry, Armin; Brock, Billy C.

Abstract not provided.

Doerry, Armin

Abstract not provided.

Doerry, Armin; Small, Donald M.

Abstract not provided.

Proceedings of SPIE - The International Society for Optical Engineering

Raynal, Ann M.; Burns, Bryan L.; Verge, Tobias J.; Bickel, Douglas L.; Dunkel, Ralf; Doerry, Armin

Knowing the statistical characteristics of a target's radar cross-section (RCS) is crucial to the success of radar target detection algorithms. A wide range of applications currently exist for dismount (i.e. human body) detection and monitoring using ground-moving target indication (GMTI) radar systems. Dismounts are particularly challenging to detect. Their RCS is orders of magnitude lower than traditional GMTI targets, such as vehicles. Their velocity of about 0 to 1.5 m/s is also much slower than vehicular targets. Studies regarding the statistical nature of the RCS of dismounts focus primarily on simulations or very limited empirical data at specific frequencies. This paper seeks to enhance the existing body of work on dismount RCS statistics at Ku-band, which is currently lacking, and has become an important band for such remote sensing applications. We examine the RCS probability distributions of different sized humans in various stances, across aspect and elevation angle, for horizontal (HH) and vertical (VV) transmit/receive polarizations, and at diverse resolutions, using experimental data collected at Ku-band. We further fit Swerling target models to the RCS distributions and suggest appropriate detection thresholds for dismounts in this band. © 2010 SPIE.

Proceedings of SPIE - The International Society for Optical Engineering

Doerry, Armin

A typical Ground Moving Target Indicator (GMTI) radar specification includes the parameters Probability of Detection (PD) - typically on the order of 0.85, and False Alarm Rate (FAR) - typically on the order of 0.1 Hz. The PD is normally associated with a particular target 'size', such as Radar Cross Section (RCS) with perhaps some statistical description (e.g. Swerling number). However, the concept of FAR is embodied at a fundamental level in the detection process, which traditionally employs a Constant-FAR (CFAR) detector to set thresholds for initial decisions on whether a target is present or not. While useful, such a metric for radar specification and system comparison is not without some serious shortcomings. In particular, when comparing FAR across various radar systems, some degree of normalization needs to occur to account for perhaps swath width and scan rates. This in turn suggests some useful testing strategies. © 2010 SPIE.

Proceedings of SPIE - The International Society for Optical Engineering

Doerry, Armin

The classical rule-of-thumb for Synthetic Aperture Radar (SAR) is that a uniformly illuminated antenna aperture may allow continuous stripmap imaging to a resolution of half its azimuth dimension. This is applied to classical line-by-line processing as well as mosaicked image patches, that is, a stripmap formed from mosaicked spotlight images; often the more efficient technique often used in real-time systems. However, as with all rules-of-thumb, a close inspection reveals some flaws. In particular, with mosaicked patches there is significant Signal to Noise ratio (SNR) degradation at the edges of the patches due to antenna beam roll-off. We present in this paper a calculation for the optimum antenna beamwidth as a function of resolution that maximizes SNR at patch edges. This leads to a wider desired beamwidth than the classical calculation. © 2010 SPIE.

Doerry, Armin; Burns, Bryan L.; Bickel, Douglas L.; Raynal, Ann M.

Abstract not provided.

Doerry, Armin; Bickel, Douglas L.; Raynal, Ann M.; Bow, Wallace J.; Denton, Michael M.

Abstract not provided.

Doerry, Armin; Bickel, Douglas L.; Raynal, Ann M.; Bow, Wallace J.; Denton, Michael M.

Abstract not provided.

Doerry, Armin

Abstract not provided.

Doerry, Armin

The performance of a Ground Moving Target Indicator (GMTI) radar system depends on a variety of factors, many which are interdependent in some manner. It is often difficult to 'get your arms around' the problem of ascertaining achievable performance limits, and yet those limits exist and are dictated by physics. This report identifies and explores those limits, and how they depend on hardware system parameters and environmental conditions. Ultimately, this leads to a characterization of parameters that offer optimum performance for the overall GMTI radar system. While the information herein is not new to the literature, its collection into a single report hopes to offer some value in reducing the 'seek time'.

Raynal, Ann M.; Doerry, Armin

Doppler radars can distinguish targets from clutter if the target's velocity along the radar line of sight is beyond that of the clutter. Some targets of interest may have a Doppler shift similar to that of clutter. The nature of sea clutter is different in the clutter and exo-clutter regions. This behavior requires special consideration regarding where a radar can expect to find sea-clutter returns in Doppler space and what detection algorithms are most appropriate to help mitigate false alarms and increase probability of detection of a target. This paper studies the existing state-of-the-art in the understanding of Doppler characteristics of sea clutter and scattering from the ocean to better understand the design and performance choices of a radar in differentiating targets from clutter under prevailing sea conditions.

Doerry, Armin

A Linear Frequency-Modulated (LFM) chirp is a function with unit amplitude and quadratic phase characteristic. In a focused Synthetic Aperture Radar (SAR) image, a residual chirp is undesired for targets of interest, as it coarsens the manifested resolution. However, for undesired spurious signals, a residual chirp is often advantageous because it spreads the energy and thereby diminishes its peak value. In either case, a good understanding of the effects of a residual LFM chirp on a SAR Impulse Response (IPR) is required to facilitate system analysis and design. This report presents an analysis of the effects of a residual chirp on the IPR. As reference, there is a rich body of publications on various aspects of LFM chirps. A quick search reveals a plethora of articles, going back to the early 1950s. We mention here purely as trivia one of the earlier analysis papers on this waveform by Klauder, et al.

Doerry, Armin; Brock, Billy C.

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. Consequently there is a need for better low grazing angle performance for trihedrals. This is facilitated by extending the bottom plate. A relevant analysis of RCS for an infinite ground plate is presented. Practical aspects are also discussed.

Doerry, Armin

Ground Moving Target Indicator (GMTI) radar maps echo data to range and range-rate, which is a function of a moving target's velocity and its position within the antenna beam footprint. Even stationary clutter will exhibit an apparent motion spectrum and can interfere with moving vehicle detections. Consequently it is very important for a radar to understand how stationary clutter maps into radar measurements of range and velocity. This mapping depends on a wide variety of factors, including details of the radar motion, orientation, and the 3-D topography of the clutter.

Doerry, Armin

Angular orientation errors of the real antenna for Synthetic Aperture Radar (SAR) will manifest as undesired illumination gradients in SAR images. These gradients can be measured, and the pointing error can be calculated. This can be done for single images, but done more robustly using multi-image methods. Several methods are provided in this report. The pointing error can then be fed back to the navigation Kalman filter to correct for problematic heading (yaw) error drift. This can mitigate the need for uncomfortable and undesired IMU alignment maneuvers such as S-turns.

Proceedings of SPIE - The International Society for Optical Engineering

Dickey, Fred M.; Doerry, Armin

There has been little interest in the information associated with the shadows in high resolution Synthetic Aperture Radar (SAR) images. In this paper we give an algorithm for the reconstruction of an object's shape from the shadows cast by the object in a sequence of SAR images. The algorithm is a back-projection type algorithm based on the intersection of solids. The effects of diffraction and synthetic aperture occlusion on SAR shadow resolution are also addressed.

Proceedings of SPIE - The International Society for Optical Engineering

Doerry, Armin

Coherent Change Detection (CCD) is a technique for observing very subtle changes between two Synthetic Aperture Radar (SAR) images. It is an Interferometric processing technique that measures the coherence between two images, and denotes 'change' where coherence is not observed, and 'no change' where coherence is observed. Consequently, the strategy must be to form both images with as much initial coherence as possible, and then see where in spite of our best efforts coherence cannot be achieved. Many things contribute to destroying coherence, but we want to eliminate all sources except for temporal change in the scene being imaged itself. This requires that a number of variables in the data collection be well controlled, and that the processing must be adapted to mitigate the effects of residual imperfections to achieve maximum coherence. It must be emphasized that the coherence calculation, that is, calculating the actual CCD product from two images is the easy part. The hard part is making sure that the two input images have the underlying characteristics to yield a quality result. The purpose of this paper is to discuss "What it takes to get good CCD results.".

Doerry, Armin

Demand is increasing for imaging ships at sea. Conventional SAR fails because the ships are usually in motion, both with a forward velocity, and other linear and angular motions that accompany sea travel. Because the target itself is moving, this becomes an Inverse- SAR, or ISAR problem. Developing useful ISAR techniques and algorithms is considerably aided by first understanding the nature and characteristics of ship motion. Consequently, a brief study of some principles of naval architecture sheds useful light on this problem. We attempt to do so here. Ship motions are analyzed for their impact on range-Doppler imaging using Inverse Synthetic Aperture Radar (ISAR). A framework for analysis is developed, and limitations of simple ISAR systems are discussed.

Doerry, Armin

Synthetic Aperture Radar (SAR) performance testing and estimation is facilitated by observing the system response to known target scene elements. Trihedral corner reflectors and other canonical targets play an important role because their Radar Cross Section (RCS) can be calculated analytically. However, reflector orientation and the proximity of the ground and mounting structures can significantly impact the accuracy and precision with which measurements can be made. These issues are examined in this report.

Proceedings of SPIE - The International Society for Optical Engineering

Doerry, Armin

It is well-known that Non-Linear FM (NLFM) chirp modulation can advantageously shape the transmitted signal's Power Spectral Density such that the autocorrelation function (i.e. matched filter output) exhibits substantially reduced sidelobes from its Linear FM (LFM) counterpart. Consequently, no additional filtering is required and maximum Signal-to-Noise Ratio (SNR) performance is preserved. This yields a 1-2 dB advantage in SNR over the output of a LFM waveform with equivalent sidelobe filtering. However precision NLFM chirps are more difficult to design, produce, and process. This paper presents design and implementation techniques for Nonlinear FM waveforms. A simple iterative design procedure is presented that yields a NLFM phase/frequency function with the desired inherent sidelobe response. We propose to then generate the NLFM waveform by using a cascaded integrator/accumulator structure. Several specific architectures are examined to meet target performance criteria, including bandwidth constraints and sidelobe reduction goals. We first examine a fixed parameter set to generate a fixed polynomial phase function. Polynomial coefficients are selected to be constant during the pulse. Alternatively, a NLFM waveform can be generated via integrating a stepped parameter set, whereby parameters are constant over specific intervals, with the pulse width encompassing multiple intervals. The parameter changes in steps during the course of the pulse as a function of time. Alternatively yet, the parameter steps can be made a function of the pulse's instantaneous frequency.

Doerry, Armin

A principal measure of Synthetic Aperture Radar (SAR) image quality is the manifestation in the SAR image of a spatial impulse, that is, the SAR's Impulse Response (IPR). IPR requirements direct certain design decisions in a SAR. Anomalies in the IPR can point to specific anomalous behavior in the radar's hardware and/or software.