Tascon Industries (United States)

The performance of acoustical ground sensors depends heavily on the local atmospheric and terrain conditions. This paper describes a prototype physics-based decision aid, called the Acoustic Battlefield Aid (ABFA), for predicting these environ-mental effects. ABFA integrates advanced models for acoustic propagation, atmospheric structure, and array signal process-ing into a convenient graphical user interface. The propagation calculations are performed in the frequency domain on user-definable target spectra. The solution method involves a parabolic approximation to the wave equation combined with a ter-rain diffraction model. Sensor performance is characterized with Cramer-Rao lower bounds (CRLBs). The CRLB calcula-tions include randomization of signal energy and wavefront orientation resulting from atmospheric turbulence. Available performance characterizations include signal-to-noise ratio, probability of detection, direction-finding accuracy for isolated receiving arrays, and location-finding accuracy for networked receiving arrays. A suite of integrated tools allows users to create new target descriptions from standard digitized audio files and to design new sensor array layouts. These tools option-ally interface with the ARL Database/Automatic Target Recognition (ATR) Laboratory, providing access to an extensive library of target signatures. ABFA also includes a Java-based capability for network access of near real-time data from sur-face weather stations or forecasts from the Army's Integrated Meteorological System. As an example, the detection footprint of an acoustical sensor, as it evolves over a 13-hour period, is calculated.

In predicting and measuring laser effect on retinal tissue for most of the visible to near infrared spectrum, one is concerned with the melanosome as the major absorber of incident energy. Differences in the location and density of melanosomes in the retinal pigment epithelium may have an impact on the effect of laser energy delivered to those tissues. Current models use estimates of numbers of melanosomes usually in an even distribution across a 5 - 8 micrometer deep volume. The goal of our study is to identify the three-dimensional distribution of melanosomes within the retinal pigment epithelium (RPE) for the use of those modeling laser tissue effects. We examined normal retinal pigment epithelium using three-dimensional (3-D) reconstruction from images obtained by transmission electron microscopy (TEM), light microscopy (LM) and confocal microscopy. Images were captured on a digital camera system attached to the microscope for both the transmission electron and light microscopy. Three-dimensional reconstruction was performed after digital deconvolution of microscopic images (Vaytek, Inc.). Three- dimensional images were then utilized for analysis of distribution of melanosomes and organelles within the pigment epithelial block. The distribution of melanosomes will be useful for accurate mathematical modeling of laser impact on the retina.