Signal Processing Commons^™

Acoustic Source Localization Using Straight Line Approximations, Swarnadeep Bagchi, Ruairí De Fréin

Conference papers

The short paper extends an acoustic signal delay estimation method to general anechoic scenario using image processing techniques. The technique proposed in this paper localizes acoustic speech sources by creating a matrix of phase versus frequency histograms, where the same phases are stacked in appropriate bins. With larger delays and multiple sources coexisting in the same matrix, it becomes cluttered with activated bins. This results in high intensity spots on the spectrogram, making source discrimination difficult. In this paper, we have employed morphological filtering, chain-coding and straight line approximations to ignore noise and enhance the target signal features. Lastly, Hough …

Development Of The Subwave Rov And Neural-Inertial Positioning System, Jason Farmer

Theses and Dissertations

This report documents the development of the Subwave, a remotely-operated underwater vehicle (ROV), and an artificial neural network based inertial positioning system. The Subwave uses the open-source ArduSub software framework, commercial-off-the-shelf hardware components, and several custom systems. It is designed as a platform for researching autonomous underwater vehicles (AUVs). The first step for an AUV is navigating waypoints, which requires the AUV to know its global position. Since global navigation satellite systems (GNSSs) are denied underwater, the available underwater positioning systems were surveyed and determined that all the available systems were too large and expensive for the Subwave. It was …

Exploratory Analysis Of Machine Learning For Images: Methods And Applications, Meenu Ajith

Electrical and Computer Engineering ETDs

This research focuses on implementing four different applications of machine learning on images. The various categories of digital images considered for these applications are grayscale, RGB, and infra-red images. The first framework uses an unsupervised learning strategy for detecting fire and smoke from an infra-red image dataset. This problem was solved using a classical machine learning algorithm since the dataset was small and unlabeled. Next, a semi-supervised deep learning model was used for facial expression recognition. Here we detect emotions from a moderately large dataset containing labeled and unlabeled grayscale images. The third application focused on single image superresolution, which …

Hybrid Smart Transformer For Enhanced Power System Protection Against Dc With Advanced Grid Support, Moazzam Nazir

All Dissertations

The traditional grid is rapidly transforming into smart substations and grid assets incorporating advanced control equipment with enhanced functionalities and rapid self-healing features. The most important and strategic equipment in the substation is the transformer and is expected to perform a variety of functions beyond mere voltage conversion and isolation. While the concept of smart solid-state transformers (SSTs) is being widely recognized, their respective lifetime and reliability raise concerns, thus hampering the complete replacement of traditional transformers with SSTs. Under this scenario, introducing smart features in conventional transformers utilizing simple, cost-effective, and easy to install modules is a highly desired …

Combinatorial Cnn Meta-Structures In Deep Learning Applications, Aswathy Rajendra Kurup

Electrical and Computer Engineering ETDs

The study aims in applying combinatorial structures motivated from the basic CNN in various applications. Convolutional neural networks (CNNs) have grown to be very popular in the field of deep learning. The ability of such networks to learn both Spatial and temporal characteristics in the data have helped in deploying them in various fields. Through this research we explore different kinds of CNN-derived architectures and how these structures can be trained and setup in combinatorial environment to solve problems in deep learning applications. First, we introduce the basic idea of a deep learning meta-structures which is used in the application …

Emotion Detection Using An Ensemble Model Trained With Physiological Signals And Inferred Arousal-Valence States, Matthew Nathanael Gray

Electrical & Computer Engineering Theses & Dissertations

Affective computing is an exciting and transformative field that is gaining in popularity among psychologists, statisticians, and computer scientists. The ability of a machine to infer human emotion and mood, i.e. affective states, has the potential to greatly improve human-machine interaction in our increasingly digital world. In this work, an ensemble model methodology for detecting human emotions across multiple subjects is outlined. The Continuously Annotated Signals of Emotion (CASE) dataset, which is a dataset of physiological signals labeled with discrete emotions from video stimuli as well as subject-reported continuous emotions, arousal and valence, from the circumplex model, is used for …

Deep Learning Based Localization Of Zigbee Interference Sources Using Channel State Information, Dylan Kensler

All Theses

As the field of Internet of Things (IoT) continues to grow, a variety of wireless signals fill the ambient wireless environment. These signals are used for communication, however, recently wireless sensing has been studied, in which these signals can be used to gather information about the surrounding space. With the development of 802.11n, a newer standard of WiFi, more complex information is available about the environment a signal propagates through. This information called Channel State Information (CSI) can be used in wireless sensing. With the help of Deep Learning, this work attempts to generate a fingerprinting technique for localizing a …

Computational Models To Detect Radiation In Urban Environments: An Application Of Signal Processing Techniques And Neural Networks To Radiation Data Analysis, Jose Nicolas Gachancipa

Beyond: Undergraduate Research Journal

Radioactive sources, such as uranium-235, are nuclides that emit ionizing radiation, and which can be used to build nuclear weapons. In public areas, the presence of a radioactive nuclide can present a risk to the population, and therefore, it is imperative that threats are identified by radiological search and response teams in a timely and effective manner. In urban environments, such as densely populated cities, radioactive sources may be more difficult to detect, since background radiation produced by surrounding objects and structures (e.g., buildings, cars) can hinder the effective detection of unnatural radioactive material. This article presents a computational model …

Credit Card Fraud Detection Using Machine Learning Techniques, Nermin Samy Elhusseny, Shimaa Mohamed Ouf, Amira M. Idrees Ami

Future Computing and Informatics Journal

This is a systematic literature review to reflect the previous studies that dealt with credit card fraud detection and highlight the different machine learning techniques to deal with this problem. Credit cards are now widely utilized daily. The globe has just begun to shift toward financial inclusion, with marginalized people being introduced to the financial sector. As a result of the high volume of e-commerce, there has been a significant increase in credit card fraud. One of the most important parts of today's banking sector is fraud detection. Fraud is one of the most serious concerns in terms of monetary …

A Comparison Of Correlation-Agnostic Techniques For Magnetic Navigation, Clark N. Taylor, Josh Hiatt

Faculty Publications

Navigation using a Global Navigation Satellite System (GNSS) is common for autonomous vehicles (ground or air). Unfortunately, GNSS-based navigation solutions are often susceptible to jamming, interference, and a limited number of satellites. A proposed technique to aid in navigation when a GNSS-based system fails is magnetic navigation - navigation using the Earth's magnetic anomaly field. This solution comes with its own set of problems including the need for quality magnetic maps in every area in which magnetic navigation will be used. Many of the currently available magnetic maps are generated from a combination of dated magnetic surveys, resulting in maps …

Automatic Contact Tracing Using Bluetooth Low Energy Signals And Imu Sensor Readings, Suriyadeepan Ramamoorthy, Joyce Mahon, Michael O'Mahony, Jean Francois Itangayenda, Tendai Mukande, Tlamelo Makati

Other resources

In this report, we present our solution to the challenge provided by the SFI Centre for Machine Learning (ML-Labs) in which the distance between two phones needs to be estimated. It is a modified version of the NIST Too Close For Too Long (TC4TL) Challenge, as the time aspect is excluded. We propose a feature-based approach based on Bluetooth RSSI and IMU sensory data, that outperforms the previous state of the art by a significant margin, reducing the error down to 0.071. We perform an ablation study of our model that reveals interesting insights about the relationship between the distance …

Soft-Mask De-Mixing For Anechoic Mixtures, Swarnadeep Bagchi, Ruairí De Fréin

Articles

This paper extends a computationally efficient, soft-mask based source separation (SS) technique called Redress, to anechoic mixing scenarios. SS methods are an integral part of hearing aid research. We call the resulting method D-Redress. In its original form, Redress was intended for instantaneous mixing scenarios. Numerical evaluations demonstrate that soft-mask based techniques reduce the level of artifacts in the separated speech. Monte Carlo trials on 1000 real speech mixtures demonstrate that the D-Redress successfully extends Redress in terms of Overall-Perceptual (OPS), Target-Perceptual (TPS) scores and Human-Ear Intelligibility (HEI).

Photon Attenuation Layer Silicon-Based X-Ray Detector: Quantum Efficiency Investigation And Algorithmic Processing Of Individual Detection Events, Kevin Larkin

ENGS 88 Honors Thesis (AB Students)

State-of-the-art X-ray detectors either use silicon direct detection—which is inefficient, as silicon is mostly transparent to high-energy X-rays—or rely on expensive scintillators to first down-convert X-ray photons to lower energies. Dr. Eldred Lee of Dartmouth College recently completed doctoral research that suggested the promise of a novel alternative: silicon-based X-ray detectors that utilize photon attenuation layers rather than scintillators. While Dr. Lee’s work is certainly rigorous and meritorious, background noise controls were rendered less effective by circumstances. The work presented in this thesis succeeded in implementing much better background noise controls for a series of new tests to compare the …

Music Visualization Using Source Separated Stereophonic Music, Hannah Eileen Chookaszian

Master's Theses

This thesis introduces a music visualization system for stereophonic source separated music. Music visualization systems are a popular way to represent information from audio signals through computer graphics. Visualization can help people better understand music and its complex and interacting elements. This music visualization system extracts pitch, panning, and loudness features from source separated audio files to create the visual. Most state-of-the art visualization systems develop their visual representation of the music from either the fully mixed final song recording, where all of the instruments and vocals are combined into one file, or from the digital audio workstation (DAW) data …

Neural Network Based Diagnosis Of Breast Cancer Using The Breakhis Dataset, Ross E. Dalke

Master's Theses

Breast cancer is the most common type of cancer in the world, and it is the second deadliest cancer for females. In the fight against breast cancer, early detection plays a large role in saving people’s lives. In this work, an image classifier is designed to diagnose breast tumors as benign or malignant. The classifier is designed with a neural network and trained on the BreakHis dataset. After creating the initial design, a variety of methods are used to try to improve the performance of the classifier. These methods include preprocessing, increasing the number of training epochs, changing network architecture, …

Standalone Wireless Humming Music Synthesizer, Elizaveta Viktorovna Koush, Conrad Kinsey, Alberto Munoz

Electrical Engineering

This project, called a standalone wireless humming music synthesizer, aimed to provide a tool for musicians to make creating music easier. It did this by removing learning barriers and adding an assistive functionality element that current synthesizers fail to address.

Existing methods for synthesizers typically include a keyboard or some collection of buttons to play melodies on. This requires the knowledge of music theory and a learning curve for the instrument, and this device aims to tackle this barrier. A device that can synthesize your voice into high-quality audio removes this musical knowledge barrier and allows anyone to easily convert …

Signal Adc Converter Simulation On Cadence Virtuoso For Audio Applications, Maxwell Kazuki Fukada

Electrical Engineering

Audio signals are representations of sounds with a mixture of multiple analog signals between the frequency of 20Hz to 20,000Hz. To record snippets of audio data onto a mobile phone or computer, the signal needs to be converted to a digital format. For this purpose, many devices utilize a converter, specifically a sigma-delta modulator with a digital filter. By using a converter, electronics can receive binary data about the audio signal accurately and quickly without losing important signal information. This project aims to simulate a fully functional audio converter with a sigma-delta modulator and decimation filter. The system will receive …

Hum Synthesizer, Daniel Keith Mcfall, Jason Chang

Electrical Engineering

Most modern music, especially the popular and rising genres such as pop music and rap music are more reliant on music synthesizers than ever before. While other instruments are readily used to sample background noises, the main form of sound generation is becoming more digital rather than using acoustic instruments. However, many people that want to make music aren’t familiarized with traditional music theory and would have difficulty using a traditional synthesizer. The Hum Synthesizer aims to allow a wider audience to access the functions of a traditional synthesizer by using voice controls rather than keyboard controls. If the user …

Dynamic Response Of Elastic Two-Story Steel Moment Frame Scaled Structure Equipped With Viscous Dampers, Garrett L. Barker, Alexander L. Poirier

Architectural Engineering

The authors of this report are Architectural Engineering undergraduate students at California Polytechnic State University, San Luis Obispo. Damping is a complex, experimentally derived value that is affected by many structural properties and has a profound effect on the dynamic response of structures. Deducing the inherent damping of a steel moment frame and affecting the damping ratio with viscous dampers are two topics explored in this paper. Dampers are commonly implemented in resilient structures that perform better in a design basis earthquake, reducing the seismic cost and downtime. Undergraduate coursework does not delve into the factors that affect damping and …

Full Pattern Analysis And Comparison Of The Center Fed And Offset Fed Cassegrain Antennas With Large Focal Length To Diameter Ratios For High Power Microwave Transmission, Derek W. Mantzke

Theses and Dissertations

High power microwaves (HPM) have been a topic of research since the Cold War era. This paper will present a comparison between two Cassegrain-type antennas: the axially, or center fed, and the offset fed. Specifically, the 10 GHz operating frequency will be investigated with large focal length to diameter () ratios. Beam patterns which encompass the entire radiation pattern will be included for data validation and optimization. The simulations will follow a design of experiments factorial model to ensure all possible combinations of prescribed parameters are included, including an analysis of variance (ANOVA) study to find parameter influence on the …

Methods For Focal Plane Array Resolution Estimation Using Random Laser Speckle In Non-Paraxial Geometries, Phillip J. Plummer

Theses and Dissertations

The infrared (IR) imaging community has a need for direct IR detector evaluation due to the continued demand for small pixel pitch detectors, the emergence of strained-layer-super-lattice devices, and the associated lateral carrier diffusion issues. Conventional laser speckle-based modulation transfer function (MTF) estimation is dependent on Fresnel propagation and a wide-sense-stationary input random process, limiting the use of this approach for lambda (wavelength)-scale IR devices. This dissertation develops two alternative methodologies for speckle-based resolution evaluation of IR focal plane arrays (FPAs). Both techniques are formulated using Rayleigh-Sommerfield electric field propagation, making them valid in the non-paraxial geometries dictated for resolution …

Polarization-Based Image Segmentation And Height Estimation For Interferometric Sar, Augusta J. Vande Hey

Theses and Dissertations

To find scatterers in a synthetic aperture radar (SAR) image, a modification is proposed to improve peak region segmentation (PRS) with region merging. The modification considers the polarization of each pixel before it is added to a segment to ensure the segment only contains pixels of the same polarization. Prior to region merging, the polarization of the segments is compared, so that only segments with the same polarization are merged into a single region. The segmented regions are used to find the height of each scatterer through interferometric SAR (IFSAR) processing. Multiple methods of IFSAR are examined to find the …

Effects Of Motion Measurement Errors On Radar Target Detection, Darnell D. Parker

Theses and Dissertations

This thesis investigates the relationships present between signal-to-clutter ratios, motion measurement errors, image quality metrics, and the task of target detection, in order to discover what factor merit greater focus in order to attain the highest probability of target detection success. This investigation is accomplished by running a high number of Monte Carlo trials through a coherent target detector and analyzing the results. The aforementioned relationships are demonstrated via sample synthetic aperture radar imagery, histograms, receiver operating characteristics curves, and error bar plots.

Theory And Design Of A Highly Compressed Dropped-Channel Polarimetric Synthetic Aperture Radar, John T. Becker

Theses and Dissertations

Compressed sensing (CS) is a recent mathematical technique that leverages the sparsity in certain sets of data to solve an underdetermined system and recover a full set of data from a sub-Nyquist set of measurements of the data. Given the size and sparsity of the data, radar has been a natural choice to apply compressed sensing to, typically in the fast-time and slow-time domains. Polarimetric synthetic aperture radar (PolSAR) generates a particularly large amount of data for a given scene; however, the data tends to be sparse. Recently a technique was developed to recover a dropped PolSAR channel by leveraging …

Ml-Based Online Traffic Classification For Sdns, Mohammed Nsaif, Gergely Kovasznai, Mohammed Abboosh, Ali Malik, Ruairí De Fréin

Articles

Traffic classification is a crucial aspect for Software-Defined Networking functionalities. This paper is a part of an on-going project aiming at optimizing power consumption in the environment of software-defined datacenter networks. We have developed a novel routing strategy that can blindly balance between the power consumption and the quality of service for the incoming traffic flows. In this paper, we demonstrate how to classify the network traffic flows so that the quality of service of each flow-class can be guaranteed efficiently. This is achieved by creating a dataset that encompasses different types of network traffic such as video, VoIP, game …

Closed-Loop Brain-Computer Interfaces For Memory Restoration Using Deep Brain Stimulation, David Xiaoliang Wang

Electrical Engineering Theses and Dissertations

The past two decades have witnessed the rapid growth of therapeutic brain-computer interfaces (BCI) targeting a diversity of brain dysfunctions. Among many neurosurgical procedures, deep brain stimulation (DBS) with neuromodulation technique has emerged as a fruitful treatment for neurodegenerative disorders such as epilepsy, Parkinson's disease, post-traumatic amnesia, and Alzheimer's disease, as well as neuropsychiatric disorders such as depression, obsessive-compulsive disorder, and schizophrenia. In parallel to the open-loop neuromodulation strategies for neuromotor disorders, recent investigations have demonstrated the superior performance of closed-loop neuromodulation systems for memory-relevant disorders due to the more sophisticated underlying brain circuitry during cognitive processes. Our efforts are …

Analysis And Implementation Of Low Fidelity Radar-Based Remote Sensing For Unmanned Aircraft Systems, Matthew Duck

Theses and Dissertations

Radar-based remote sensing is consistently growing, and new technologies and subsequent techniques for characterization are changing the feasibility of understanding the environment. The emergence of easily accessible unmanned aircraft system (UAS) has broadened the scope of possibilities for efficiently surveying the world. The continued development of low-cost sensing systems has greatly increased the accessibility to characterize physical phenomena. In this thesis, we explore the viability and implementation of using UAS as a means of radar-based remote sensing for ground penetrating radar (GPR) and polarimetric scatterometry. Additionally, in this thesis, we investigate the capabilities and implementations of low-cost microwave technologies for …

Recognizing Traffic Signaling Gestures Through Automotive Sensors., Benjamin James Bartlett

Theses and Dissertations

As technology advances with each new day, so do the applications and uses of the different modalities of technology, including transportation, particularly in ADAS vehicles. These systems allow the vehicle to avoid collisions, change lanes, adjust the vehicle’s speed, and more without the need of driver input. However, each sensor type has a weakness, and most advanced driver- assisted system (ADAS) vehicles rely heavily on sensors, such as RGB cameras, radars, and LiDAR sensors. These visual-based sensors may collect very noisy data in cloudy, raining, foggy, or other obscuring phenomena. Radar, on the other hand, does not rely on visual …

Performance Enhancement Of Wide-Range Perception Issues For Autonomous Vehicles, Suvash Sharma

Theses and Dissertations

Due to the mission-critical nature of the autonomous driving application, underlying algorithms for scene understanding should be given special care during their development. Mostly, they should be designed with precise consideration of accuracy and run-time. Accuracy should be considered strictly which if compromised leads to faulty interpretation of the environment that may ultimately result in accidental scenarios. On the other hand, run-time holds an important position as the delayed understanding of the scene would hamper the real-time response of the vehicle which again leads to unforeseen accidental cases. These factors come as the functions of several factors such as the …

A Harmonic Radar System For Honey Bee Tracking To Better Understand Colony Collapse Disorder, William B. Woo

Theses and Dissertations

Honey bees are some of the most important pollinators for agriculture in the world and are pivotal to the health of worldwide ecosystems. Like all insects, bees struggle with exposure to parasites, diseases, and other environmental factors that can negatively affect the overall health of the colony. Recently, a new unexplainable phenomenon called Colony Collapse Disorder (CCD) has been wreaking havoc on bee populations worldwide. As a result, a system capable of tracking bees is required to understand the different contributions of chemicals, parasites, etc. to CCD. This research seeks to show data supporting the development of systems for an …