Engineering Commons^™

Multi-Domain Secure Dds Networks For Aerial And Ground Vehicle Communications, Daniel Pendleton

All Theses

none

Analysis Of Cnn Performance Utilizing Jpeg Compressed Images Created On An Fpga, Timothy Shaughnessy

All Theses

JPEG (Joint Photographic Experts Group) was formed in 1986 to create a method to reduce image size primarily for ease of transfer on the Internet. Released to the public in 1992, JPEG compression is a form of lossless compression that has been a staple for compressing images. JPEG is the go-to image compressor because it provides high compression ratios while maintaining visual integrity for the human eye. Growing image sizes have made JPEG compression increasingly relevant. It is vital to keep up with growing data sizes for improved image handling performance on an edge device like a Field-Programmable Gate Array …

Analyzing An In-Line Compression Management System For Improved Performance In A High-Performance Computing Environment, Steven Platt

All Theses

High-performance computing (HPC) has enabled advancements in computation speed and resource cost by utilizing all available server resources and using parallelization for speedup. This computation scheme encourages simulation model development, massive data collection, and AI computation models, all of which store and compute on massive amounts of data. Data compression has enhanced the performance of storing and transferring this HPC application data to enable acceleration, but the benefits of data compression can also be transferred to the active allocated memory used by the application. In-line compression is a compression method that keeps the application memory compressed in allocated memory, decompressing …

Deep Reinforcement Learning Of Variable Impedance Control For Object-Picking Tasks, Akshit Lunia

All Theses

The increasing deployment of robots in industries with varying tasks has accelerated the development of various control frameworks, enabling robots to replace humans in repetitive, exhaustive, and hazardous jobs. One critical aspect is the robots' interaction with their environment, particularly in unknown object-picking tasks, which involve intricate object weight estimations and calculations when lifting objects. In this study, a unique control framework is proposed to modulate the force exerted by a manipulator for lifting an unknown object, eliminating the need for feedback from a force/torque sensor. The framework utilizes a variable impedance controller to generate the required force, and an …

Defining And Labeling Traversable Space In A Forested Environment, James Nguyen

All Theses

This thesis investigates the problem of identifying traversable terrain in outdoor conditions. We are motivated by research in recent years toward identifying drivable space for the purpose of developing autonomous vehicles. Our motivating application is similar but also different. We envision a “Hiker Helper” that assists humans with dismounted navigation in forested terrain. A common challenge in this type of environment is identifying a viable path for moving through terrain that is congested with trees, bushes, other flora, and natural obstacles that would make navigation difficult. We envision training an artificial intelligence (AI) model to automatically analyze images of this …

Detection Of Myofascial Trigger Points With Ultrasound Imaging And Machine Learning, Benjamin Formby

All Theses

Myofascial Pain Syndrome (MPS) is a common chronic muscle pain disorder that affects a large portion of the global population, seen in 85-93% of patients in specialty pain clinics [10]. MPS is characterized by hard, palpable nodules caused by a stiffened taut band of muscle fibers. These nodules are referred to as Myofascial Trigger Points (MTrPs) and can be classified by two states: active MTrPs (A-MTrPs) and latent MtrPs (L-MTrPs). Treatment for MPS involves massage therapy, acupuncture, and injections or painkillers. Given the subjectivity of patient pain quantification, MPS can often lead to mistreatment or drug misuse. A deterministic way …

A Study Of 5g Cellular Connectivity To Unmanned Aerial Vehicles, Jackson Murrin

All Theses

The market of unmanned aerial vehicles (UAVs) has seen significant growth in the past ten years on both the commercial and military sides. The applications for UAVs are endless and options by manufacturers allow users to modify their drones for their specific goals. This industry has opened up the excitement of piloting vehicles in the air, photography, videography, exploration of nature from a different point of view and many other hobbies assisted by the emergence of UAVs. The growth of this industry coincides with the roll out of new 5G cellular network technology. This upgrade in cellular network infrastructure allows …

Virtual Prototyping Of Pebb Based Power Electronics System For Ground Vehicles, Yi Li

All Theses

Power electronics are heavily involved in power and energy systems in plenty of applications nowadays. The increase of demand brings more challenges into simulations for development. Considering the complexity of the systems and high frequency operational conditions, this paper presents comprehensive research on modeling, simulating, and validation on ground vehicle propulsion system applications.

To reduce the computational burden, the Power Electronics Building Blocks concept is utilized to simplify the structure of modeling under different conversion scenarios in ground vehicle systems. In addition, the Average and Switching versions models are included. To speedup the simulation, the engagement of advanced computing technique …

Generalizable Deep-Learning-Based Wireless Indoor Localization, Ali Owfi

All Theses

The growing interest in indoor localization has been driven by its wide range of applications in areas such as smart homes, industrial automation, and healthcare. With the increasing reliance on wireless devices for location-based services, accurate estimation of device positions within indoor environments has become crucial. Deep learning approaches have shown promise in leveraging wireless parameters like Channel State Information (CSI) and Received Signal Strength Indicator (RSSI) to achieve precise localization. However, despite their success in achieving high accuracy, these deep learning models suffer from limited generalizability, making them unsuitable for deployment in new or dynamic environments without retraining. To …

Performance Modeling Of Inline Compression With Software Caching For Reducing The Memory Footprint In Pysdc, Sansriti Ranjan

All Theses

Modern HPC applications compute and analyze massive amounts of data. The data volume is growing faster than memory capabilities and storage improvements leading to performance bottlenecks. An example of this is pySDC, a framework for solving collocation problems iteratively using parallel-in-time methods. These methods require storing and exchanging 3D volume data for each parallel point in time. If a simulation consists of M parallel-in-time stages, where the full spatial problem has to be stored for the next iteration, the memory demand for a single state variable is M ×Nx ×Ny ×Nz per time-step. For an application simulation with many state …

Improving Inference Speed Of Perception Systems In Autonomous Unmanned Ground Vehicles, Bradley Selee

All Theses

Autonomous vehicle (AV) development has become one of the largest research challenges in businesses and research institutions. While much research has been done, autonomous driving still requires extensive amounts of research due to its immense, multi-factorial difficulty. Autonomous vehicles rely on many complex systems to function, make accurate decisions, and, above all, provide maximum safety. One of the most crucial components of autonomous driving is the perception system.

The perception system allows the vehicle to identify its surroundings and make accurate, but safe, decisions through the use of computer vision techniques like object detection, image segmentation, and path planning. Due …

Effects Of Surface Noise On Printing Artifacts: An Artistic Approach To Hiding Print Artifacts, Samuel New

All Theses

This research focuses on improving the quality of Fused Filament Fabrication (FFF) 3D printing by using fractal noise to mask certain print artifacts (e.g. layer lines and stair-stepping). The use of textures is quite common in digital sculpting for aesthetic reasons. This study focuses on finding specific textures that minimize visible 3D print artifacts.

Transforming Character Faces Based On Perceived Personality Traits, Kara Porter

All Theses

The ability to read other human's faces is a crucial part of everyday life. Subconsciously, the human brain analyzes someone's face within the first few seconds of seeing it, making a variety of conclusions ~\cite{FacePerp} such as gathering information about emotional state and assuming character traits this person might possess. The purpose of this thesis is to create a tool that allows a user to alter features of a character's three dimensional (3D) face mesh to look increasingly or decreasingly like the character possesses certain personality traits. Using a sample set of randomly generated faces, a survey is conducted to …

A Digital Healthcare Application For Patient Monitoring And Assessment, Brandon Shumin

All Theses

The COVID-19 pandemic strained our healthcare resources and exacerbated the existing issues of primary care shortages and burnout rates for healthcare professionals. Due in part to these factors, telehealth has seen more wide-spread use during this time. However, current asynchronous telehealth applications require stable Internet to function fully. Since many medically underserved populations in the United States lack Internet access in their homes, an application that offers patient monitoring and assessment could extend their access to medical resources. This work proposes such a digital healthcare application for iOS devices and evaluates it based on the system requirements of availability, data …

A Quantitative Comparison Of Algorithmic And Machine Learning Network Flow Throughput Prediction, Cayden Wagner

All Theses

Applications ranging from video meetings, live streaming, video games, autonomous vehicle operations, and algorithmic trading heavily rely on low latency communication to operate optimally. A solution to fully support this growing demand for low latency is called dual-queue active queue management (AQM). Dual-queue AQM's functionality is reduced without network traffic throughput prediction.

Perhaps due to the current popularity of machine learning, there is a trend to adopt machine learning models over traditional algorithmic throughput prediction approaches without empirical support. This study tested the effectiveness of machine learning as compared to time series forecasting algorithms in predicting per-flow network traffic throughput …

Improving Intelligent Transportation Safety And Reliability Through Lowering Costs, Integrating Machine Learning, And Studying Model Sensitivity, Cavender Holt

All Theses

As intelligent transportation becomes increasingly prevalent in the domain of transportation, it is essential to understand the safety, reliability, and performance of these systems. We investigate two primary areas in the problem domain. The first area concerns increasing the feasibility and reducing the cost of deploying pedestrian detection systems to intersections in order to increase safety. By allowing pedestrian detection to be placed in intersections, the data can be better utilized to create systems to prevent accidents from occurring. By employing a dynamic compression scheme for pedestrian detection, we show the reduction of network bandwidth improved by 2.12× over the …

Application Of Image Processing And Convolutional Neural Networks For Flood Image Classification And Semantic Segmentation, Jaku Rabinder Rakshit Pally

All Theses

Floods are among the most destructive natural hazards that affect millions of people across the world leading to severe loss of life and damage to property, critical infrastructure, and the environment. Deep learning algorithms are exceptionally valuable tools for collecting and analyzing the catastrophic readiness and countless actionable flood data. Convolutional neural networks (CNNs) are one form of deep learning algorithms widely used in computer vision which can be used to study flood images and assign learnable weights and biases to various objects in the image. Here, we leveraged and discussed how connected vision systems can be used to embed …

Lossy Compression And Its Application On Large Scale Scientific Datasets, Tasmia Reza

All Theses

High Performance Computing (HPC) applications are always expanding in data size and computational complexity. It is becoming necessary to consider fault tolerance and system recovery to reduce computation and resource cost in HPC systems. The computation of modern large scale HPC applications are facing bottleneck due to computation complexities, increased runtime and large data storage requirements. These issues can not be ignored in current supercomputing era. Data compression is one of the effective ways to address data storage issue. Among data compression, the lossy compression is much more feasible and efficient than the traditional lossless compression due to low I/O …

Parking Lot Monitoring System Using An Autonomous Quadrotor Uav, Venkataraman Ganesh

All Theses

The main goal of this thesis is to develop a drone-based parking lot monitoring system using low-cost hardware and open-source software. Similar to wall-mounted surveillance cameras, a drone-based system can monitor parking lots without affecting the flow of traffic while also offering the mobility of patrol vehicles. The Parrot AR Drone 2.0 is the quadrotor drone used in this work due to its modularity and cost efficiency. Video and navigation data (including GPS) are communicated to a host computer using a Wi-Fi connection. The host computer analyzes navigation data using a custom flight control loop to determine control commands to …

Detecting Occlusions Of Automobile Parts Being Inspected By A Camera System During Manufacturing Assembly, Jayadevan Puthumanappilly

All Theses

This thesis considers the problem of detecting occlusions in automobile parts on a moving assembly line in an automotive manufacturing plant. This work builds on the existing ``Visual Inspector'' (VI) system developed as a joint research project between Clemson University and the BMW Spartanburg manufacturing plant. The goal is to develop a method that can successfully detect occlusions in real-time. VI is a detector and classifier system that uses video cameras to determine the correct installation of a part in the assembly line. In the current version of VI, an occluded part is flagged simply as `not OK' - as …

Scientific Application Acceleration Utilizing Heterogeneous Architectures, Edwin Weill

All Theses

Within the past decade, there have been substantial leaps in computer architectures to exploit the parallelism that is inherently present in many applications. The scientific community has benefited from the emergence of not only multi-core processors, but also other, less traditional architectures including general purpose graphical processing units (GPGPUs), field programmable gate arrays (FPGAs), and Intel's many integrated cores (MICs) architecture (i.e. Xeon Phi). The popularity of the GPGPU has increased rapidly because of their ability to perform massive amounts of parallel computation quickly and at low cost with an ease of programmability. Also, with the addition of high-level programming …

A Device To Record Naturally Daily Wrist Motion, Surya Sharma

All Theses

We introduce a new device to record and store wrist motion activity data. The motivation to create this device was the fact that this data can be used to detect periods of eating or the number of bites consumed. There is no similar device available in the market. This device uses new components that have been recently introduced to the market, and newer techniques that can be used for low quantity production. The production cost for this device was $52, similar to other fitness trackers on the market. The device was capable of recording wrist motion activity for 24 hours …

A Load-Based Approach To Forming A Connected Dominating Set For An Ad Hoc Network, Raihan Hazarika

All Theses

Efficient routing in mobile ad hoc networks (MANETs) is highly desired and connected dominating sets (CDS) have been gaining significant popularity in this regard. The CDS based approach reduces the search for a minimum cost path between a pair of source and destination terminals to the set of terminals forming the backbone network. Researchers over the years have developed numerous distributed and localized algorithms for constructing CDSs which minimize the number of terminals forming the backbone or which provide multiple node-disjoint paths between each pair of terminals. However none of this research focuses on minimizing the load at the bottleneck …

Verifying A Systematic Application To Accelerator Roadmap Using Shallow Water Wave Equations, Ranajeet Anand

All Theses

With the advent of parallel computing, a number of hardware architectures have become available for data parallel applications. Every architecture is unique with respect to characteristics such as floating point operations per second, memory bandwidth and synchronization costs. Data parallel applications possess inherent parallelism that needs to be studied and the hardware that can best exploit this parallelism can be identified and selected for large-scale implementation. The application that I have considered for my thesis is - numerical solution of shallow water wave equations using finite difference method. These equations are a set of partial differential equations that model the …

Electric Power Synchrophasor Network Cyber Security Vulnerabilities, Christopher Beasley

All Theses

Smart grid technologies such as synchrophasor devices (Phasor Measurement Units (PMUs)), make real-time monitoring, control, and analysis of the electric power grid possible. PMUs measure voltage and current phasors across the electrical power grid, add a GPS time stamps to measurements, and sends reports to the Phasor Data Concentrators (PDCs) in the control centers. Reports are used to make decisions about the condition and state of the power grid. Since this approach relies on Internet Protocol (IP) network infrastructure, possible cybersecurity vulnerabilities have to be addressed to ensure that it is stable, secure, and reliable. In literature, attacks that are …

A Study Of Time-Based Features And Regularity Of Manipulation To Improve The Detection Of Eating Activity Periods During Free Living, Jose Reyes

All Theses

This thesis considers the problem of detecting when people eat by tracking their wrist motion. The goal of this work is to automatically detect the start and stop times of these eating activities. It builds upon previous work done by our research group, which developed an algorithm for automatically detecting peaks in activities associated with food preparation and clean-up. This peak detector is then used for segmenting data. These segments are then classified as eating or non-eating activities using a naive Bayes classifier based on probabilities obtained from computing different features in each segment. The original work introduced 4 features, …

Privacy Preserving Statistics, Oluwakemi Hambolu

All Theses

Over the past few years, there have been an increase in the development and improvement of circumvention tools like Tor and Psiphon. These tools provide an environment for citizens of oppressive regimes to access websites freely without fear of identification, these tools aid democracy activists and journalists in West Africa in using the Internet securely. A similar circumvention tool was developed by us. This tool circumvents DNS and IP address blocking/filtering, by leveraging technologies developed by criminal botnet enterprises. To improve and maintain the circumvention tool we developed, it is important to quantify the number and country of origin of …

Ear Contour Detection And Modeling Using Statistical Shape Models, Satish Ravindran

All Theses

Ear detection is an actively growing area of research because of its applications in human head tracking and biometric recognition. In head tracking, it is used to augment face detectors and to perform pose estimation. In biometric systems, it is used both as an independent modality and in multi-modal biometric recognition. The ear shape is the preferred feature used to perform detection because of its unique structure in both 2D color images and 3D range images. Ear shape models have also been used in literature to perform ear detection, but at a cost of a loss in information about the …

Sociallink: A Social Network Based Trust System For P2p File Sharing Systems, Fang Qi

All Theses

In peer-to-peer (P2P) file sharing systems, many autonomous peers without preexisting trust relationships share files with each other. Due to their open environment and distributed structure, these systems are vulnerable to the significant impact from selfish and misbehaving nodes. Free-riding, whitewash, collusion and Sybil attacks are common and serious threats, which severely harm non-malicious users and degrade the system performance. Many trust systems were proposed for P2P file sharing systems to encourage cooperative behaviors and punish non-cooperative behaviors. However, querying reputation values usually generates latency and overhead for every user. To address this problem, a social network based trust system …

Statistical Regression Methods For Gpgpu Design Space Exploration, Nimisha Raut

All Theses

General Purpose Graphics Processing Units (GPGPUs) have leveraged the performance and power efficiency of today's heterogeneous systems to usher in a new era of innovation in high-performance scientific computing. These systems can offer significantly high performance for massively parallel applications; however, their resources may be wasted due to inefficient tuning strategies. Previous application tuning studies pre-dominantly employ low-level, architecture specific tuning which can make the performance modeling task difficult and less generic. In this research, we explore the GPGPU design space featuring the memory hierarchy for application tuning using regression-based performance prediction framework and rank the design space based on …