Daniel P. Campbell

Obtaining a 35x Speedup in 2D Phase Unwrapping Using Commodity Graphics Processors

Graphics processing units (GPUs) are a powerful tool for numerical computation. The GPU architecture and computational model are uniquely designed for high-resolution high-speed grid-based calculations. This capability can be utilized to accelerate certain classes of compute-intensive radar signal processing algorithms. Characteristics of a problem well-suited for computation on a GPU include high levels of data parallelism, low control logic, uniform boundary conditions, and well-defined input and output. We describe the implementation of two-dimensional multigrid least-squares weighted phase unwrapping on a GPU and demonstrate a large speedup over C and MATLAB implementations. Details of the GPU computation are provided. Background information on the GPU architecture and its applicability to general-purpose computation is discussed.

Optical system-on-a-chip for chemical and biochemical sensing: the chemistry

Planar waveguides have evanescent fields sensitive to index of refraction changes in the volume immediately above the waveguide surface. Optically combining one guided sensing beam with a reference beam in an interferometric configuration generates measurable signals. Applying a chemically selective film over the sensing arm of the interferometer provides the basis for a chemical sensor. Tailored chemistries can be passive (e.g.; inducing swelling or dissolution in a film) or active (e.g.; containing reactive or binding sites). Fast and reversible chemistries are the goal, in most cases for both gaseous and liquid applications. Passive mechanisms are used when the target analyte is relatively inert, i.e. aromatic and chlorinated hydrocarbons. Active chemistries developed include tailoring the acid-base strength of the sensing film for pH or ammonia response, and antibody-antigen binding. Currently the integrated optic waveguide platform consists of thirteen interferometers on a 1 X 2-cm glass substrate. A different sensing film deposited on each channel allows for multiple analyte sensing, interferant cancellation, patterned outputs for analyte identification, or extended dynamic range. Sensitivities range from the low ppm to low ppb for both vapor and aqueous applications, 0.01 pH units and ng/mL for biologicals.

Gigapixel spotlight synthetic aperture radar backprojection using clusters of GPUs and CUDA

Synthetic aperture radar (SAR) image formation via backprojection offers a robust mechanism by which to form images on general, non-planar surfaces, without often restrictive assumptions regarding the planarity of the wavefront at the locations being imaged. However, backprojection presents a substantially increased computational load relative to other image formation algorithms that typically depend upon fast Fourier transforms. In this paper, we present an image formation framework for accelerated SAR backprojection that utilizes a cluster of computing nodes, each with one or more graphics processing units (GPUs). We address the parallelization of the backprojection process among multiple nodes and the scalability thereby obtained, several optimization approaches, and performance as a function of both allocated resources and desired precision. Finally, we demonstrate the achieved performance on a simulated gigapixel-scale data set.

Real-time implementations of ordered-statistic CFAR

Ordered-statistic constant false alarm rate (OS-CFAR) detectors provide improved robustness over cell-averaging CFAR (CA-CFAR) detectors in multiple target and heterogeneous clutter environments. However, this benefit comes at the cost of generally increased processing time due to the need for a rank-ordering of the CFAR training data. Realtime implementations of OS-CFAR must consider this additional processing burden. In this paper, we present real-time FPGA and CPU/GPU implementations of OS-CFAR. A novel sorting architecture that scales linearly with window size is presented alongside traditional compare-and-swap and rank-only architectures in an FPGA. A rank-only GPU implementation is demonstrated alongside multi-threaded sorting and rank-only CPU implementations. Effects of training window size on throughput and power consumption are considered.

Integrated optic chemical sensor for environmental monitoring

An integrated optic chemical sensor has been developed to monitor benzene, toluene, ethylbenzene and xylene (BTEX) in water. The sensor uses planar waveguide interferometry, where the evanescent field associated with a guided wave probes the refractive index changes immediately above the waveguide surface. Currently, up to thirteen interferometers are fabricated on a 1 X 2 cm glass chip. One arm of each interferometer is coated with a chemically interactive film, and the other arm is buried under an inert layer of silicon dioxide (SiO2). The interference pattern formed by combining the guided waves from the two arms is read by a linear photodiode array, and onboard electronics convert the raw optical intensities into analyte concentrations. The sensor is packaged in a 1.5 inch diameter, 18 inch long stainless steel housing suitable for use in monitoring wells of with cone penetrometers. It is plug-and-play compatible with E-SMARTTM monitoring networks.

Chapter 9 – Interferometric Biosensors

Publisher Summary Interferometry is an optical method that compares differences experienced by two light beams traveling along similar paths. A bioconjugate reaction taking place within one of these beams provides the basis for a biosensor. The most common form chosen for the interferometer involves the propagation of the light within planar waveguides, which are preferred for their long interaction length. Planar waveguides have evanescent fields sensitive to index of refraction changes in the volume immediately above the waveguide surface. These fields extend up to 5000 A above the surface. Placing a chemically sensitive film within this region concentrates the reaction within this field. Chemical or physical interactions change the index of refraction causing the propagating light speed, or phase, to change in a direction opposite to that of the index change. To measure this change, a reference propagating beam, placed adjacent to a sensing beam, is optically combined with the sensing beam to create an interference pattern of alternating dark and light fringes. When chemical or physical changes occur in the sensing arm, the interference pattern will shift producing a sinusoidal output. Several interferometric sensing schemes are presented and their sensitivities are compared. Sensitivity to a biochemical event is a function of the extent of evanescent field interaction with the bioconjugate reaction as well as the path length of this interaction.

Integrated optic sensor for pH and ammonia

Planar waveguides have evanescent fields sensitive to index of refraction changes in the volume immediately above the waveguide surface. Optically combining one guided sensing beam with a reference beam in an interferometric configuration generates measurable signals. Applying a chemically selective film over the sensing arm of the interferometer provides the basis for a chemical sensor. Acid-base chemistry either on the waveguide surface or incorporated into a polymer film, deposited on the waveguide, allows for reversibly sensing pH in solution or ammonia in air. The pKas of the sensing groups can be spaced across the pH range to permit continuous change over the entire range. Alternatively, sensing and reference arms can have different sensing groups with pKas, which bracket the acidity or basicity of the target analyte. A polymer film thicker than the evanescent field shields the optical beam from environmental changes, but also permits facile transfer of either protons or ammonia. Multiple interferometers can be fabricated on a single integrated optic chip. Channels not used for acid-base sensing can be used to cancel out interferants, or to measure other analytes. Sensitivities achieved to date are in the ppbv range for ammonia and <0.01 pH in the pH sensor.

Planar-Waveguide Interferometers for Chemical Sensing

Interferometry is an optical technique that compares the differences experienced by two light beams traveling along similar paths. Planar waveguides have evanescent fields sensitive to changes in the index of refraction in the volume immediately above the waveguide surface. Placing a chemically sensitive film within this region provides the basis for chemical sensing. Film–analyte interactions change the index of refraction, causing the propagating light speed or phase to change in a direction of opposite sign to that of the index change. To measure this change, a reference propagating beam, which is adjacent to the sensing beam, is combined optically with the sensing beam, thus creating an interference pattern of alternating dark and light fringes. When chemical or physical changes occur in the sensing arm, the interference pattern shifts, producing a sinusoidal output. Waveguides and interferometers come in a variety of designs, but all rely on the evanescent field interacting with a chemically selective film to produce a measured response. The sensing mechanism can be passive (a physical change) or active (reactive sites in the film). Through a judicious choice of sensing films, interferometers can be designed to detect a wide variety of chemical and biological materials. Multi-interferometer devices with several different sensing films can be used to detect and identify a variety of different chemical or biological analytes either through specific sensing chemistry or through analysis of patterned response from an array of different films.