Publications

The analysis presented herein predicts that, under signal-independent noise limited conditions, an Information-efficient Spectral Imaging Sensor (ISIS) style hyperspectral imaging system design can obtain significant signal-to-noise ratio (SNR) and speed increase relative to a comparable traditional hyperspectral imaging (HSI) instrument. Factors of forty are reasonable for a single vector, and factors of eight are reasonable for a five-vector measurement. These advantages can be traded with other system parameters in an overall sensor system design to allow a variety of applications to be done that otherwise would be impossible within the constraints of the traditional HSI style design.

Minimum detectable irradiance levels for a diffraction grating based laser sensor were calculated to be governed by clutter noise resulting from reflected earth albedo. Features on the earth surface caused pseudo-imaging effects on the sensor`s detector arras that resulted in the limiting noise in the detection domain. It was theorized that a custom aperture transmission function existed that would optimize the detection of laser sources against this clutter background. Amplitude and phase aperture functions were investigated. Compared to the diffraction grating technique, a classical Young`s double-slit aperture technique was investigated as a possible optimized solution but was not shown to produce a system that had better clutter-noise limited minimum detectable irradiance. Even though the double-slit concept was not found to have a detection advantage over the slit-grating concept, one interesting concept grew out of the double-slit design that deserved mention in this report, namely the Barker-coded double-slit. This diffractive aperture design possessed properties that significantly improved the wavelength accuracy of the double-slit design. While a concept was not found to beat the slit-grating concept, the methodology used for the analysis and optimization is an example of the application of optoelectronic system-level linear analysis. The techniques outlined here can be used as a template for analysis of a wide range of optoelectronic systems where the entire system, both optical and electronic, contribute to the detection of complex spatial and temporal signals.

A unique end-to-end LIDAR sensor model has been developed supporting the concept development stage of the CALIOPE UV DIAL and UV laser-induced-fluorescence (LIF) efforts. The model focuses on preserving the temporal and spectral nature of signals as they pass through the atmosphere, are collected by the optics, detected by the sensor, and processed by the sensor electronics and algorithms. This is done by developing accurate component sub-models with realistic inputs and outputs, as well as internal noise sources and operating parameters. These sub-models are then configured using data-flow diagrams to operate together to reflect the performance of the entire DIAL system. This modeling philosophy allows the developer to have a realistic indication of the nature of signals throughout the system and to design components and processing in a realistic environment. Current component models include atmospheric absorption and scattering losses, plume absorption and scattering losses, background, telescope and optical filter models, PMT (photomultiplier tube) with realistic noise sources, amplifier operation and noise, A/D converter operation, noise and distortion, pulse averaging, and DIAL computation. Preliminary results of the model will be presented indicating the expected model operation depicting the October field test at the NTS spill test facility. Indications will be given concerning near-term upgrades to the model.