Signal Processing Commons^™

Gaussian Mixture Models And Neural Networks For Automatic Speaker Identification, Usha Gayatri Chalkapally

Electrical & Computer Engineering Theses & Dissertations

Automatic Speaker Recognition is the process of automatically recognizing who is speaking on the basis of individual information contained in speech signals. This technique of Automatic Speaker Recognition makes it possible to use the speaker's voice to verify their identity and control access to services such as voice dialing, banking by telephone, telephone shopping, database access services, information services, voice mail, security control for confidential information areas, and remote access to computers.

In this thesis, the techniques of Gaussian Mixture Models and Neural Networks for Automatic Speaker Identification are presented. Algorithms for Speaker Identification using Gaussian Mixture Models were developed, …

Apparatus And Method For Using Adaptive Algorithms To Exploit Sparsity In Target Weight Vectors In An Adaptive Channel Equalizer, Richard K. Martin, Robert C. Williamson, William A. Sethares

AFIT Patents

An apparatus and method is disclosed for using adaptive algorithms to exploit sparsity in target weight vectors in an adaptive channel equalizer. An adaptive algorithm comprises a selected value of a prior and a selected value of a cost function. The present invention comprises algorithms adapted for calculating adaptive equalizer coefficients for sparse transmission channels. The present invention provides sparse algorithms in the form of a Sparse Least Mean Squares (LMS) algorithm and a Sparse Constant Modulus Algorithm (CMA) and a Sparse Decision Directed (DD) algorithm.

Multiple Masks-Based Pixel Comparison Steganalysis Method For Mobile Imaging, Sos S. Agaian, Gilbert L. Peterson, Benjamin M. Rodriguez

Faculty Publications

No abstract provided.

Performance-Metric Driven Atmospheric Compensation For Robust Free-Space Laser Communication, Peter N. Crabtree

Theses and Dissertations

The effect of turbulence on laser propagation is a significant challenge to current electro-optical systems. While atmospheric compensation techniques in space object imaging and high-energy laser weapons have been thoroughly investigated, optimizing these techniques for Laser Communication (LaserCom) has not been examined to the same degree. Average Strehl ratio is the typical design metric for current atmospheric compensation systems. However, fade probability is the relevant metric for LaserCom. This difference motivated the investigation into metric-driven atmospheric compensation. Metric-based tracking techniques for fade mitigation is the first major focus of this research. In a moderate range air-to-air scenario, focal plane spot …

Radiometric Analysis Of Daytime Satellite Detection, Katherine B. Lilevjen

Theses and Dissertations

A radiometric model for daylight satellite detection is developed and used to evaluate the effects of various parameters on signal-to-noise ratio (SNR). Detection of reflected sunlight from a low-earth orbit, diffuse, planar satellite by a single-pixel infrared photovoltaic detector is considered. Noise considered includes photon noise from the background and signal, as well as thermal noise. Parameters considered include atmospheric conditions, optical parameters, and detector parameters. The Phillips Laboratory Expert-assisted User System, an atmospheric modeling tool that employs the MODTRAN and FASCODE transmission codes, is used to model wavelength-dependent atmospheric transmission and background radiance. The SNR is found to increase …

Passive Ranging Using Atmospheric Oxygen Absorption Spectra, Michael R. Hawks

Theses and Dissertations

The depth of absorption bands in observed spectra of distant, bright sources can be used to estimate range to the source. A novel approach is presented and demonstrated using observations of the oxygen absorption band near 762 nm. Range is estimated by comparing observed values of band-average absorption against curves derived from either historical data or model predictions. Curves are based on fitting a random band model to the data, which reduces average range error by 67% compared to the Beer's Law model used in previous work. A new modification to existing band models for long, inhomogeneous paths is presented …

Multiframe Shift Estimation, Stephen A. Bruckart

Theses and Dissertations

The purpose of this research was to develop a fundamental framework for a new approach to multiframe translational shift estimation in image processing. This thesis sought to create a new multiframe shift estimator, to theoretically prove and experimentally test key properties of it, and to quantify its performance according to several metrics. The new estimator was modeled successfully and was proven to be an unbiased estimator under certain common image noise conditions. Furthermore its performance was shown to be superior to the cross correlation shift estimator, a robust estimator widely used in similar image processing cases, according to several criteria. …

Verification Of A Decision Level Fusion Algorithm Using A Proven Atr System And Measured Sar Data, James Douglas Thompson

Theses and Dissertations

Decision level fusion (DLF) algorithms combine outputs of multiple single sensors to make one confident declaration of a target. This research compares performance results of a DLF algorithm using measured data and a proven ATR system with results from simulated data and a modeled ATR system. This comparison indicates that DLF offers significant performance improvements over single sensor looks. However, results based on simulated data and a modeled ATR are slightly optimistic and overestimate results from measured data and a proven ATR system by nearly 10% over all targets tested.

Doppler Aliasing Reduction In Wide-Angle Synthetic Aperture Radar Using Phase Modulated Random Stepped-Frequency Waveforms, Andrew W. Hyatt

Theses and Dissertations

This research effort examines the theory, application and results of side-looking airborne radar operation in hot clutter. Hot clutter is an electronic counter-measure used to degrade the performance of airborne radar. Hot clutter occurs by illuminating the ground with an airborne jammer at some velocity, azimuth, elevation, and range from the airborne radar. When the received RCS scattered hot clutter waveform is perfectly coherent with the radar waveform, the radar believes the returns created by the hot clutter jammer resulted from the transmitting radar. Hot clutter degrades radar performance at locations in azimuth and Doppler. The effect of hot clutter …

A Climatological Study Of Equatorial Gps Data And The Effects On Ionospheric Scintillation, Katharine A. Wicker

Theses and Dissertations

Ionospheric scintillation is detrimental to radio signals, especially those from the global positioning system. Such scintillation is caused when a signal permeates the ionosphere through plasma bubbles. The signal’s phase and amplitude can be altered, and a receiver on the ground can lose lock on the GPS signal. Measured using a zero to one index known as S4, scintillation severity is based upon season, solar cycle, time of day, location and frequency. The most severe scintillation occurs at the equatorial anomaly, or fifteen degrees north and south of the equator. Seven years of data from fifteen different locations around the …

Radio Frequency Enabled Soil Redox Potential Sensor Networks, Chris Holme

Theses : Honours

There is a need for' cost effective tools and data collection methods for field measurements; to increase both productivity and volumes of collected data in the quest for enhanced understanding and management of environmental systems. To such end, various technologies that may be combined into a cost effective soil redox sensor network were explored. Suitability of each technology, as a component of said network was evaluated. A prototype soil redox sensor network was constructed and basic laboratory and field testing was completed. Results indicate that the prototype sensor network functions correctly within bounds. Both laboratory and field testing show that …

Partition-Based Interpolation For Color Filter Array Demosaicking And Super-Resolution Reconstruction, Min Shao, Kenneth E. Barner, Russell C. Hardie

Electrical and Computer Engineering Faculty Publications

A class of partition-based interpolators that addresses a variety of image interpolation applications are proposed. The proposed interpolators first partition an image into a finite set of partitions that capture local image structures. Missing high resolution pixels are then obtained through linear operations on neighboring pixels that exploit the captured image structure. By exploiting the local image structure, the proposed algorithm produces excellent performance on both edge and uniform regions. The presented results demonstrate that partition-based interpolation yields results superior to traditional and advanced algorithms in the applications of color filter array (CFA) demosaicking and super-resolution reconstruction.

Scene-Based Nonuniformity Correction Technique For Focal-Plane Arrays Using Readout Architecture, Balaji Narayanan, Russell C. Hardie, Robert A. Muse

Electrical and Computer Engineering Faculty Publications

Spatial fixed pattern noise is a common and major problem in modern infrared imagers due to the nonuniform response of the photodiodes in the focal plane array of the imaging system. In addition, the nonuniform response of the readout and the digitization electronics, involved in multiplexing the signals from the photodiodes, causes further nonuniformity. In this paper, we describe a novel scene based nonuniformity correction algorithm that treats the aggregate nonuniformity in separate stages. Firstly, the nonuniformity from the readout amplifiers is corrected using knowledge of the readout architecture of the imaging system. In the second stage, the nonuniformity resulting …

The Navigation Potential Of Signals Of Opportunity-Based Time Difference Of Arrival Measurements, Kenneth A. Fisher

Theses and Dissertations

This research introduces the concept of navigation potential, NP, to quantify the intrinsic ability to navigate using a given signal. NP theory is a new, information theory-like concept that provides a theoretical performance limit on estimating navigation parameters from a received signal that is modeled through a stochastic mapping of the transmitted signal and measurement noise. NP theory is applied to SOP-based TDOA systems in general as well as for the Gaussian case. Furthermore, the NP is found for a received signal consisting of the transmitted signal, multiple delayed and attenuated replicas of the transmitted signal, and measurement noise. Multipath-based …

Deconvolution Analysis Of Laser Pulse Profiles From 3-D Ladar Temporal Returns, Michael D. Walter

Theses and Dissertations

Three-dimensional laser imaging systems offer important advantages for battlefield applications, such as night-time targeting and tactical reconnaissance. Recently developed technologies used by coherent detection systems that collect temporally resolved images include arrays of Avalanche Photo-Diodes (APD), Geiger mode APDs, and photo-diodes. Frequently, LADAR systems produce waveforms from each detector that characterize the convolution of the transmitted laser pulse with the target surface. The pulse convolution generates uncertainty as to the precise location of a target surface, which can severely impact various weapon systems' targeting capability. This work analyzes two deconvolution techniques: Wiener filtering and an iterative process derived from the …

Using Liquid Crystal Spatial Light Modulators For Closed Loop Tracking And Beam Steering With Phase Holography, Michael J. Perry

Theses and Dissertations

Optical Phased Array (OPA) technology offers advantages in the reduction of size, weight, and power of optical steering devices. Nematic liquid crystal (LC) spatial light modulators (SLMs) have been studied as a potential candidate for building non-mechanical OPAs. They can steer a laser beam and split the beam into multiple beams. This thesis builds upon the prior research showing each split beam can be individually controlled, including variation in intensity. A closed loop tracking scenario shows the flexibility of the SLM by tracking and stabilizing an incoming beam. Results show that applying a phase grating to the SLM has limitations …

Determination Of Structure From Motion Using Aerial Imagery, Paul R. Graham

Theses and Dissertations

The structure from motion process creates three-dimensional models from a sequence of images. Until recently, most research in this field has been restricted to land-based imagery. This research examines the current methods of land-based structure from motion and evaluates their performance for aerial imagery. Current structure from motion algorithms search the initial image for features to track though the subsequent images. These features are used to create point correspondences between the two images. The correspondences are used to estimate the motion of the camera and then the three-dimensional structure of the scene. This research tests current algorithms using synthetic data …

Daytime Detection Of Space Objects, Alistair D. Funge

Theses and Dissertations

Space Situational Awareness (SSA) requires repeated object updates for orbit accuracy. Detection of unknown objects is critical. A daytime model was developed that evaluated sun flares and assessed thermal emissions from space objects. Iridium satellites generate predictable sun glints. These were used as a model baseline for daytime detections. Flares and space object thermal emissions were examined for daytime detection. A variety of geometric, material and atmospheric characteristics affected this daytime detection capability. In a photon noise limited mode, simulated Iridium flares were detected. The peak Signal-to- Noise Ratios (SNR) were 6.05e18, 9.63e5, and 1.65e7 for the nighttime, daytime and …

The Use Of X-Ray Pulsars For Aiding Gps Satellite Orbit Determination, Dennis W. Woodfork Ii

Theses and Dissertations

This research proposes the use of an existing "signal of opportunity" - namely x-ray pulsars - to improve the accuracy and robustness of the GPS satellite and clock estimation algorithm. Improvement in satellite and clock accuracy results in a direct benefit to the user. A simulation has been developed to determine the effects of using x-ray pulsar measurements on the GPS Operational Control Segment. The epoch-specific position, velocity, and clock errors of all GPS satellites in the constellation were estimated using both pseudoranges and time-difference-of-arrival (TDOA) measurements from pulsars. The primary measure of accuracy is a constellation Signal-In-Space Range Error …

Noise Estimation In The Presence Of Bpsk Digital Burst Transmissions, Susan E. Bettison

Theses and Dissertations

This research explores noise estimation techniques in an attempt to improve upon a previously developed digital burst transmission Binary Phase Shift Keyed (BPSK) demodulator. The demodulator success is dependent on the accuracy of the estimate of Power Spectral Density (PSD) of the unknown noise. Given a discrete time signal transformed into the frequency domain, the research seeks to determine if it is possible to effectively estimate the PSD of the unknown noise. The demodulator was developed using a new signal model for digital burst transmissions based on linear spectral subspace theory. Using this model and the redundancy properties of BPSK …

Consistency Results For The Roc Curves Of Fused Classifiers, Kristopher S. Bjerkaas

Theses and Dissertations

The U.S. Air Force is researching the fusion of multiple classifiers. Given a finite collection of classifiers to be fused, one seeks a new classifier with improved performance. An established performance quantifier is the Receiver Operating Characteristic (ROC) curve, which allows one to view the probability of detection versus the probability of false alarm in one graph. Previous research shows that one does not have to perform tests to determine the ROC curve of this new fused classifier. If the ROC curve for each individual classifier has been determined, then formulas for the ROC curve of the fused classifier exist …

A Fast And Simple Algorithm For Computing M Shortest Paths In Stage Graph, M. Sherwood, Laxmi P. Gewali, Henry Selvaraj, Venkatesan Muthukumar

Electrical & Computer Engineering Faculty Research

We consider the problem of computing m shortest paths between a source node s and a target node t in a stage graph. Polynomial time algorithms known to solve this problem use complicated data structures. This paper proposes a very simple algorithm for computing all m shortest paths in a stage graph efficiently. The proposed algorithm does not use any complicated data structure and can be implemented in a straightforward way by using only array data structure. This problem appears as a sub-problem for planning risk reduced multiple k-legged trajectories for aerial vehicles.

A Fast And Simple Algorithm For Computing M-Shortest Paths In State Graph, M. Sherwood, Laxmi P. Gewali, Henry Selvaraj, Venkatesan Muthukumar

Electrical & Computer Engineering Faculty Research

We consider the problem of computing m shortest paths between a source node s and a target node t in a stage graph. Polynomial time algorithms known to solve this problem use complicated data structures. This paper proposes a very simple algorithm for computing all m shortest paths in a stage graph efficiently. The proposed algorithm does not use any complicated data structure and can be implemented in a straightforward way by using only array data structure. This problem appears as a sub-problem for planning risk reduced multiple k-legged trajectories for aerial vehicles.

A Computer-Based Articulation Training Aid For Short Words (Cata), Mukund Devarajan

Electrical & Computer Engineering Theses & Dissertations

Several improvements in the vowel articulation training aid (VATA) are described, as well as the efforts to extend the visual feedback system to operate with short words in the form of consonant, vowel and consonant (CVC). The extended version of the visual feedback system is referred to as CATA (Computer-based Articulation Training Aid); the vowel version of the aid (VATA) only operates with ten American English monopthong vowels. Improvements in VATA include the use of a neural network (NN) recognizer method to prune a large database of vowel recordings to eliminate noisy and/or mispronounced tokens. The spectral jitter problem, previously …

Redundant Discrete Wavelet Transform Based Super-Resolution Using Sub-Pixel Image Registration, Daniel L. Ward

Theses and Dissertations

The limited resolution of video imagery taken by aircraft, over geographical areas of interest, hinders the accurate extraction of useful information. The frame resolution of the video is determined by the camera that created it. Information exists about the camera which can be used to increase frame resolution beyond the resolution capability of the camera. This is achieved by a process called super-resolution, which uses multiple low-resolution video frames to create one high-resolution image.

Evaluating The Performance Of Multiple Classifier Systems: A Matrix Algebra Representation Of Boolean Fusion Rules, Justin M. Hill

Theses and Dissertations

Given a finite collection of classifiers one might wish to combine, or fuse, the classifiers in hopes that the multiple classifier system (MCS) will perform better than the individuals. One method of fusing classifiers is to combine their final decision using Boolean rules (e.g., a logical OR, AND, or a majority vote of the classifiers in the system). An established method for evaluating a classifier is measuring some aspect of its Receiver Operating Characteristic (ROC) curve, which graphs the trade-off between the conditional probabilities of detection and false alarm. This work presents a unique method of estimating the performance of …

Inquisitive Pattern Recognition, Amy L. Magnus

Theses and Dissertations

The Department of Defense and the Department of the Air Force have funded automatic target recognition for several decades with varied success. The foundation of automatic target recognition is based upon pattern recognition. In this work, we present new pattern recognition concepts specifically in the area of classification and propose new techniques that will allow one to determine when a classifier is being arrogant. Clearly arrogance in classification is an undesirable attribute. A human is being arrogant when their expressed conviction in a decision overstates their actual experience in making similar decisions. Likewise given an input feature vector, we say …

Feature Guided Image Registration Applied To Phase And Wavelet-Base Optic Flow, Kate R. Duffy

Theses and Dissertations

Optic Flow algorithms are useful in problems such as computers vision, navigational systems, and robotics. However, current algorithms are computationally expensive or lack the accuracy to be effective compared with traditionally navigation systems. Recently, lower accuracy inertial navigation systems (INS) based on Microelectromechanical systems (MEMS) technology have been proposed to replace more accurate traditional navigation systems.

Digital Image Processing, Russell C. Hardie, Majeed M. Hayat

Electrical and Computer Engineering Faculty Publications

In recent years, digital images and digital image processing have become part of everyday life. This growth has been primarily fueled by advances in digital computers and the advent and growth of the Internet. Furthermore, commercially available digital cameras, scanners, and other equipment for acquiring, storing, and displaying digital imagery have become very inexpensive and increasingly powerful. An excellent treatment of digital images and digital image processing can be found in Ref. [1]. A digital image is simply a two-dimensional array of finite-precision numerical values called picture elements (or pixels). Thus a digital image is a spatially discrete (or discrete-space) …

Yet Another Algorithm For Pitch Tracking (Yaapt), Kavita Kasi

Electrical & Computer Engineering Theses & Dissertations

This thesis presents a pitch detection algorithm that is extremely robust for both high quality and telephone speech. The kernel method for this algorithm is the Normalized Cross Correlation (NCCF) reported by David Talkin [16]. Major innovations include: processing of the original acoustic signal and a nonlinearly processed version of the signal to partially restore very weak F0 components; intelligent peak picking to select multiple F0 candidates and assign merit factors; and, incorporation of highly robust pitch contours obtained from smoothed versions of low frequency portions of spectrograms. Dynamic programming is used to find the ''best" pitch track among all …