From the collection of Dean's Award recipients, one overall winner is chosen to receive the Chancellor's Award. The Chancellor's Award is presented to the graduating student who produces the most exceptional master's thesis. Dr. Todd Stewart, AFIT Chancellor and Director, and Maj Gen Everett Odgers, USAF, Retired, Treasurer of the AFIT Foundation, present the Chancellor's Award and Russ Prize.
Recent Chancellor's Award winners and the title of their theses are listed below. AFIT theses are available from the Defense Technical Information Center (DTIC), the Air University Research Web, and in the AFIT library.
2018 - Capt Steven M. Beyer
"Pattern-of-Life Modeling using Data Leakage in Smart Homes"
2017 - Maj Francis D. Hallada
"The Fresnel Zone Light Field Spectral Imager"
2016 - Capt Thompson Graves, U.S. Marine Corps
Capt Graves was the recipient of the 2016 Chancellor's Award and Russ Prize.
2015 - 2nd Lieutenant Megan E. Lewis
" Recent Advances in Compressed Sensing: Discrete Uncertainty Principles and Fast Hyperspectral Imaging"
Compressed sensing is an important field with continuing advances in theory and applications. This thesis provides contributions to both theory and application. Much of the theory behind compressed sensing is based on uncertainty principles, which state that a signal cannot be concentrated in both time and frequency. We develop a new discrete uncertainty principle and use it to demonstrate a fundamental limitation of the demixing problem, and to provide a fast method of detecting sparse signals. The second half of this thesis focuses on a specific application of compressed sensing: hyperspectral imaging. Conventional hyperspectral platforms require long exposure times, which can limit their utility, and so we propose a compressed sensing platform to quickly sample hyperspectral data. We leverage certain combinatorial designs to build good coded apertures, and then we apply block orthogonal matching pursuit to quickly reconstruct the desired imagery.
2014 - Capt David L. Levene
" An Embedded, Programmable GPS Injection Jammer for Aircraft Testing and Aircrew Training"
2013 - Capt Timothy Cleaver
" Aerodynamic Characterization of an Air-to-Air Missile with Computational Fluid Dynamics"
The US Air Force continually seeks to improve the survivability and lethality of its air superiority systems. In recent years, several converging trends have generated interest in the development of small, extremely agile multi-role missiles. The extreme maneuverability of these agile missile systems exposes them to unconventional aerodynamic environments that are not well understood.
This research reduced the technical risks of developing new agile missile systems by undertaking a thorough and reliable characterization of the aerodynamic environment associated with their extreme flight profiles. Capt Cleaver utilized a DOD-developed Computational Fluid Dynamic (CFD) tool, known as Kestrel, to predict the aerodynamic forces and moments generated by a concept missile geometry through a complete range of representative flight conditions. He provided additional analytical rigor and breadth of applicability to his analyses through judicious use of Design of Experiments (DoE) principles.
The CFD simulations of the missile geometry were able to predict induced side forces and yaw moments on symmetric configurations at intermediate angles of attack, such as those observed in AFIT subsonic wind tunnel results. At transonic Mach numbers, the CFD estimates of in-plane forces were comparable to benchmark, semi-empirical prediction tools. Finally, computational estimates of pitch moment indicate that the concept missile may be statically unstable at intermediate angles of attack. The above observations, together with the extensive aerodynamic data collected, provide invaluable feedback to the system developer who must take these performance characteristics into account when designing the integrated aerodynamic and attitude control systems featured in these next generation missiles.
2012 - 2nd Lieutenant Jack G. FitzGerald
" A Programmable Liquid Collimator for Both Coded Aperture Adaptive Imaging and Multiplexed Compton Scatter Tomography"
The first fully reconfigurable liquid collimator was designed, built, and tested. Several radioactive sources were viewed by the system. Images of the sources improved in all tests when the liquid collimator was used. There are a number of promising potential applications for this new device. The device could be mated with a UAV or ground-based robot to survey large areas for radiation after a nuclear explosion or after a reactor accident. This is the mission interest of the sponsor USSTRATCOM. The system could find radioactive effluents for treaty monitoring. Used for Multiplexed Compton Scatter Tomography (MCST), the device could perform single-sided airfoil defect inspections, bone density studies, and historic artifact scanning. If mated with a single radiation detector, large decreases in cost for a position-sensitive detector could be realized. The research could be used toward satellite surveillance of the ground for locating nuclear detonations. Similarly, space-based astronomy could benefit from this device with improvements in solar flare or Gamma-Ray Burst (GRB) observations. Finally, the device could improve medical imaging, to include Computed Tomography (CT), Single Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET).
2011 - Captain Brian T. Bohan
" Analysis of Flow Migration in an Ultra-Compact Combustor"
The Air Force is continually interested in reducing the weight of its systems and in particular the weight of the engines & ndash without losing thrust! One means of accomplishing this objective is an Ultra Compact Combustor (UCC). This state of the art combustor uses a circumferential cavity to accomplish the combustion thereby using a circumferential annulus to acquire the length needed for complete combustion while shortening the overall length of the combustor by as much as 60%. Several studies have been accomplished at small, missile sized, components with good success. At AFIT we would like to build and study a fighter-sized component that would be consistent with the next developmental engine cycle which we obtained from AFRL. The current research uncovered the design criteria necessary to achieve a high g-loading in the circumferential cavity (which is crucial to maintain high flame speeds and thus fast reaction times) as well as an understanding of the methods to reintroduce the hot circumferential flow back into the axial core flow. This mixing process is crucial to producing an exit temperature profile that turbine rotors can withstand. If the gas on the outside diameter is too hot, the turbine tips will melt and overall engine performance will be seriously degraded as an understatement. The second area of investigation within the current research stemmed around the biggest obstacle we foresee in creating a practical UCC combustor. This results from the interaction of any unburned fuel species with the oxygen rich film cooling flow in the vane hardware. Results have been shown that highlight this secondary burning phenomena in the turbine. This thesis has taken the next step in trying to mitigate the effects of this phenomena and how to reduce the overall heat load to the turbine.
2010 - Captain Casey J. Pellizzari
" Phase Unwrapping in the Presence of Strong Turbulence"
This research increases the ability of AO systems to operate under strong atmospheric turbulence conditions. Turbulence becomes stronger as range increases or the atmosphere becomes thicker. Optical sensors, laser communication, and directed energy weapons utilizing AO will be able to operate at greater distances and at lower angles to the horizon.
2009 - Major Daniel T. Schmitt
" Automated Knowledge Generation with Persistent Video Surveillance"
Some of the benefits offered by automated knowledge generation are improved intelligence and the increased speed to process persistent video into intelligence.
2008 - Captain Robert Johnson
" Improved Feature Extraction, Feature Selection, and Identification Techniques that Create a Fast Unsupervised Hyperspectral Target Detection Algorithm"
This research provided new techniques to identify targets in HSI data in real-time. These techniques (demonstrated on eight HIS test images) were demonstrated to be superior to the current methods described in literature. Further, the new techniques provided the capability to fully automate the target detection process.
2007 - Captain Robert Koo
" Feature Extraction Using Principal and Independent Component Analysis
for Hyperspectral Imagery"
With an increasing availability of new technology comes a heightened need for integration of data into real world application. This research leverages analysis of cutting edge technology for application within the Total Force frame work. Further, it will provide a baseline for future exploitation of the underlying data for use in several arenas within a variety of Air Force mission areas. This research resolves challenges associated with the implementation of independent component analysis as a target detection technique. The proposed Two-Phase Filtering Independent Component Analysis Target Detection Method has shown to be robust, and has achieved promising results when tested against a wide range of benchmark images.
2006 - Captain Steven M. Ross
" Formation Flight Control For Aerial Refueling"
2005 - Major Kurt W. Meidel
" GPS Signal Jamming Mitigation Through Multiple Model Adaptive Estimation Applied to Ultra-Tightly Coupled GPS/INS Architecture"
Compressed sensing is an important field with continuing advances in theory and applications. This thesis provides contributions to both theory and application. Much of the theory behind compressed sensing is based on uncertainty principles, which state that a signal cannot be concentrated in both time and frequency. We develop a new discrete uncertainty principle and use it to demonstrate a fundamental limitation of the demixing problem, and to provide a fast method of detecting sparse signals.
hyperspectral imaging. Conventional hyperspectral platforms require long exposure times, which can limit their utility, and so we propose a compressed sensing platform to quickly sample hyperspectral data. We leverage certain combinatorial designs to build good coded apertures, and then we apply block orthogonal matching pursuit to quickly reconstruct the desired imagery.
The second half of this thesis focuses on a specific application of compressed sensing: