Engineering Commons^™

Trajectory Analysis For Driving Safety Quantification, Michael I. Chang

UNLV Theses, Dissertations, Professional Papers, and Capstones

In order to evaluate the efficacy of the skid recovery exercise in the Driver’s Edge teenage driving program, a process is established to determine the trajectories of vehicles from recorded videos, compare them in terms of similarity through dynamic time warping (DTW), and then analyze the similarity measurements to assess whether the program has a significant effect on driving ability by repeated measures analysis of variance (rANOVA). The video is analyzed by Harris corner detection and Lucas-Kanade optical flow method to ascertain the vehicle trajectories. A homography is then estimated to translate coordinates from video into real-world. The instructor and …

A Control Integrated Circuit For A Hysteretic Flyback Power Converter, Armani Alvarez

UNLV Theses, Dissertations, Professional Papers, and Capstones

This Thesis reports the design, simulation, and layout of a switching power supply controller chip for a fly-back switching power supply. The controller chip was designed using ON’s Semiconductor C5 process. The resulting switching power supply using the controller chip consists of using various off-chip components such as a power MOSFET, transformer, and capacitors. The main components used in the integrated circuit include a bandgap reference voltage circuit which outputs a reference voltage of 1.25V, a ring oscillator having a nominal frequency of 5MHz, a high-gain comparator, and a buffer that can drive a 5pF capacitor. Under normal simulation conditions …

Efficient Networks-On-Chip Communication Support Solutions For Deep Neural Network Acceleration, Binayak Tiwari

UNLV Theses, Dissertations, Professional Papers, and Capstones

The increasing popularity of deep neural network (DNN) applications demands high computing power and efficient hardware accelerator architectures. DNN accelerators use a large number of processing elements (PEs) and on-chip memory for storing weights and other parameters. A significant challenge is faced when designing a many-core DNN accelerator to handle the data movement between the processing elements. As the communication backbone of a DNN accelerator, networks-on-chip (NoC) plays an important role in supporting various dataflow patterns and enabling processing with communication parallelism in a DNN accelerator. However, the widely used mesh-based NoC architectures inherently cannot efficiently support many-to-one (gather) and …

Radiomic Features To Predict Overall Survival Time For Patients With Glioblastoma Brain Tumors Based On Machine Learning And Deep Learning Methods, Lina Chato

UNLV Theses, Dissertations, Professional Papers, and Capstones

Machine Learning (ML) methods including Deep Learning (DL) Methods have been employed in the medical field to improve diagnosis process and patient’s prognosis outcomes. Glioblastoma multiforme is an extremely aggressive Glioma brain tumor that has a poor survival rate. Understanding the behavior of the Glioblastoma brain tumor is still uncertain and some factors are still unrecognized. In fact, the tumor behavior is important to decide a proper treatment plan and to improve a patient’s health. The aim of this dissertation is to develop a Computer-Aided-Diagnosis system (CADiag) based on ML/DL methods to automatically estimate the Overall Survival Time (OST) for …

Modeling And Analysis Of Subcellular Protein Localization In Hyper-Dimensional Fluorescent Microscopy Images Using Deep Learning Methods, Yang Jiao

UNLV Theses, Dissertations, Professional Papers, and Capstones

Hyper-dimensional images are informative and become increasingly common in biomedical research. However, the machine learning methods of studying and processing the hyper-dimensional images are underdeveloped. Most of the methods only model the mapping functions between input and output by focusing on the spatial relationship, whereas neglect the temporal and causal relationships. In many cases, the spatial, temporal, and causal relationships are correlated and become a relationship complex. Therefore, only modeling the spatial relationship may result in inaccurate mapping function modeling and lead to undesired output. Despite the importance, there are multiple challenges on modeling the relationship complex, including the model …

Effects Of Distributed Energy Resources On The Bulk Electric System, Stryder Loveday

UNLV Theses, Dissertations, Professional Papers, and Capstones

The traditional power grid was designed with a centralized resource distribution in mind, with a relatively small number of large power generation facilities supplying the vast majority of load. With recent efforts to further diversify the power grid as well as an increased interest in renewable energy sources, there has been an unprecedented amount of new DistributedEnergy Resources (DERs) added to the grid within the last decade. This often heavily influences the local demand where these resources are installed, often causing power to flow in ways not anticipated by the original design of the grid. This thesis reviews the potential …

Development Of Uav-Based Rail Track Geometry Irregularity Monitoring And Measuring Platform Empowered By Artificial Intelligence, Lihao Qiu

UNLV Theses, Dissertations, Professional Papers, and Capstones

Rail tracks need to be consistently monitored and inspected for problems associated with rust, deformation, and cracks that, at their worst, can cause catastrophic train derailments. Many non-destructive testing approaches have been explored and extensively utilized to help inspect rails’ health, but most of them require intensive human power and/or heavy sensor systems (e.g. total stations, manual/car-mounted trolly, etc.) that are not efficient or convenient to cover a long range of rails and may interfere with the normal operation of trains.In light of the rapid development of unmanned aerial systems/vehicles (UAS’s/UAVs) and high definition photographic and optical distance measuring sensors, …

Predicting The Progression Of Diabetes Mellitus Using Dynamic Plantar Pressure Parameters, Mathew Sunil Varre

UNLV Theses, Dissertations, Professional Papers, and Capstones

Introduction: Diabetic peripheral neuropathy is one of the common complications of type-2 diabetes mellitus (DM). Changes in the intrinsic plantar tissue coupled with repetitive mechanical loads and loss of sensation may lead to foot related complications (skin break down, ulcerations, and amputations) in persons with neuropathy if left untreated. The purpose of this dissertation was to stratify individuals with pre-diabetes, diabetes with and without neuropathy using dynamic plantar pressure parameters during walking, using machine learning algorithms.Methods: Plantar pressure data was collected from one hundred participants during walking with pressure measuring insoles fixed between the feet and thin socks. Simultaneously high-definition …

Tinyml For Gait Stride Classification, Priyanka Rajendra

UNLV Theses, Dissertations, Professional Papers, and Capstones

Human gait classification and analysis become very important when a person has been diagnosed with a neurological disorder or has suffered an injury which has affected their ability to walk correctly. Gait strides are an important parameter to be studied as it helps the doctor to diagnose any underlying gait condition and evaluate what type of treatment suits the best for the patient’s recovery. Studying gait strides also helps athletes to improve their performance.In today’s world, machine learning has emerged as one of the most widely used technology for classification and analysis of gait characteristics. TinyML is a field of …

Designs And Outcomes Of Transcranial Magnetic Stimulation (Tms) And Repetitive Transcranial Magnetic Stimulation (Rtms) Circuits, Daniel Senda

UNLV Theses, Dissertations, Professional Papers, and Capstones

This thesis reports the design and outcomes of several circuits intended for transcranial magnetic stimulation (TMS) and repetitive transcranial magnetic stimulation (rTMS) research. In simple terms, TMS circuits are composed of four main blocks: high voltage power source, energy storage bank, control switch, and coil. Each one of these blocks has characteristics that influence how well the circuit will perform for TMS procedures. A successful TMS research circuit must have the ability to emit controlled electromagnetic pulses through a coil connected to it. For the first block, voltages ranging from 50 V to 2 kV were used. In the second …

Analysis Of Microscopic Objects Using Computer Vision Methods, Yuan Dao

UNLV Theses, Dissertations, Professional Papers, and Capstones

As an essential and powerful tool to observe living organisms, three-dimensional fluorescence microscopy is widely used in biological research and diagnosis. The 4D fluorescence microscopy data can be obtained using time-lapsed videos of 3D images. To analyze and extract useful information from the increasingly large and complex biological image dataset, efficient and effective computational tools are in need but still lagging behind. In analyzing biological data, two major challenges are faced. First, time-lapsed fluorescence microscopic images typically have a low SNR. Second, biological objects often change their morphology and internal structure frequently. As such, conventional image processing methods may not …

Forecasting Pedestrian Trajectory Using Deep Learning, Arsal Syed

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this dissertation we develop different methods for forecasting pedestrian trajectories. Complete understanding of pedestrian motion is essential for autonomous agents and social robots to make realistic and safe decisions. Current trajectory prediction methods rely on incorporating historic motion, scene features and social interaction to model pedestrian behaviors. Our focus is to accurately understand scene semantics to better forecast trajectories. In order to do so, we leverage semantic segmentation to encode static scene features such as walkable paths, entry/exits, static obstacles etc. We further evaluate the effectiveness of using semantic maps on different datasets and compare its performance with already …

Dynamic Analysis Of A Microgrid Powered With An Inverter And Machine-Based Distributed Resources, Brandon William Blackstone

UNLV Theses, Dissertations, Professional Papers, and Capstones

The proliferation of renewable distributed energy resources, particularly photovoltaic (PV) power systems, and the increasing need for a reliable power supply has led to the concept of microgrids, a mini-grid that consists of locally connected power generation units and needs, able to operate connected or disconnected from the utility grid, using controlled and coordinated methods to provide for the users of the microgrid the best possible conditions for their needs. The main technical issues facing microgrids include some of the following, seamless transition from stand-alone to utility grid connected operation, how to preserve frequency and voltage stability, and provide the …

Healthy Aging In Place: An Integrated Smart Home Using Alexa-Compatible Technology, Tyler Kryst

UNLV Theses, Dissertations, Professional Papers, and Capstones

The purpose of this research was to address current healthcare concerns among the aging population. Around the globe, the average life expectancy is projected to continue increasing resulting in a higher prevalence of age-related disease. With the ever-increasing capabilities of modern technology, it is necessary to apply this technology to increase the health and safety of older adults. This work had three objectives. The first objective was the establishment of an integrated smart home aimed at creating a healthy aging-in-place environment through the use of Alexa. The second objective was the development and integration of a completely custom health device …

Infants' Facial Emotion Recognition, Umme Fatema

UNLV Theses, Dissertations, Professional Papers, and Capstones

Facial Emotion Recognition (FER) has become a popular computer vision topic and attracted a lot of researchers. However, these researches mostly focus on adult facial emotion. Infants’ facial structure and expression of emotions differ from adults’. That is why our target is to create a model that particularly focuses on infants’ facial emotions. In this work, we classify three types of emotions: positive, negative, and neutral. Two datasets are used in this work. One is an image-based dataset named ‘The City’ that consists of 154 images of infants aged 0-12 months. Another is the Rebel Dataset from the Department of …

Patient-Specific Finite Element Analysis For Mandibular Fracture Fixation, Ethan Snyder

UNLV Theses, Dissertations, Professional Papers, and Capstones

This thesis proposes an approach for Finite Element Analysis (FEA) of mandibular fracture fixation. Using a Computerized Tomography (CT) scan of mandible obtained from a specific person, the material characteristics, density and modulus of elasticity, were determined from a set of discrete points within the mandible that are 1mm spaced based on the Hounsfield Units of these points. The mandible geometry was sectioned to simulate a fracture. Muscle and mastication forces were added to replicate post-surgery loading. Using a standard linear miniplate, this material model was compared with two commonly used mandibular cortical shell bone models: isotropic and orthotropic. A …

A Circle Hough Transform Implementation Using High-Level Synthesis, Carlos Lemus

UNLV Theses, Dissertations, Professional Papers, and Capstones

Circle Hough Transform (CHT) has found applications in biometrics, robotics, and imageanalysis. In this work, the focus is the development of a Field Programmable Gate Array (FPGA) based accelerator that performs a series of procedures and results in circle detection. The design is performed using Vivado High-Level Synthesis (HLS) tools and targeted for a Zynq UltraScale+ ZCU106. The implementation includes the following procedures: Gaussian filter, Sobel edge operator, thresholding, and finally the CHT algorithm. The performance is evaluated based on the execution time as compared to the software (Python code) execution and the analysis tools provided by Vivado HLS tool. …

Fighter Aircraft Guidance And Control, Kaushik Raj

UNLV Theses, Dissertations, Professional Papers, and Capstones

Aircraft technology has undergone momentous improvements over the decades due to the growing demand and competition among industries. The performance requirement and stability of fighter aircraft is a major factor due to the ever-increasing flight envelope. Advanced fighter aircraft are expected to be highly maneuverable. To achieve extreme maneuverability a very good aircraft design with multiple redundant control actuators having static instabilities in certain modes, near accurate nonlinear aircraft mathematical model, and suitable nonlinear control design technique is required. Conventionally, training aircrafts such as Cessna and Piper aircraft are built for static and dynamic stability, making them easy to trim …

Current-Mode Photon-Counting Circuit With Sige Bicmos Input Stage, Gonzalo Guadalupe Arteaga

UNLV Theses, Dissertations, Professional Papers, and Capstones

Continued advances in photodetector technology have helped improve the imaging capabilities of LiDAR systems or other time-of-flight imaging efforts. As a major part of the receiver within a LiDAR system, the photodetector can offer features such as faster rise times, narrower pulse widths, or larger gains to help achieve greater timing resolution or sensitivity to faint light levels. These benefits, however, are contingent on the ability of the accompanying readout circuit to reliably process the fast, electrical signals generated by the photodetector. While common front-end circuits involve the use of feedback amplifiers, their stability can become a major concern at …

A Multi-Channel Mcp-Pmt Based Readout Integrated Circuit For Lidar Applications, Sachin Purushothaman Namboodiri

UNLV Theses, Dissertations, Professional Papers, and Capstones

Photon counting techniques are becoming more critical in fields such as LiDAR, high energy physics (HEP), and positron emission tomography (PET). For space-based aerosol-cloud-ocean (ACO) LiDAR, the total aggregate photon flux signal has a very high dynamic range, from a single-photon up to giga-photons per second for a single channel. This dissertation focuses on the design of a multichannel, photon counting readout circuit that can interface with MCP-PMTs for high dynamic range, space-based LiDAR applications. Chapter 2 presents the conventional current mode approach that has been employed to realize a photon counting circuit. A transimpedance amplifier, a 6-bit delay line …

Design And Fabrication Of A 6-Bit Current-Mode Adc For Lidar And High-Speed Applications, Jason Silic

UNLV Theses, Dissertations, Professional Papers, and Capstones

The design, simulation, fabrication, and testing of a 6-bit current-mode ADC is presented with related calculations. Capable of producing samples at greater than 100MHz, the proposed design should excel at measuring fast current transients such as those produced by arrays of avalanche photodiodes (APDs). An array of 63 current comparators feed digital logic gates, via “thermometer code,” that produce a 6-bit output word that is driven off-chip by appropriate buffers.The circuit was designed using Cadence Virtuoso design software. Simulated Differential Non-Linearity (DNL), with an 8µA reference current, is 1.656 LSB at the worst transition involving a scale change. Excluding scale …

Water Quality Prediction Based On Machine Learning Techniques, Zhao Fu

UNLV Theses, Dissertations, Professional Papers, and Capstones

Water is one of the most important natural resources for all living organisms on earth. The monitoring of treated wastewater discharge quality is vitally important for the stability and protection of the ecosystem. Collecting and analyzing water samples in the laboratory consumes much time and resources. In the last decade, many machine learning techniques, like multivariate linear regression (MLR) and artificial neural network (ANN) model, have been proposed to address the problem. However, simple linear regression analysis cannot accurately forecast water quality because of complicated linear and nonlinear relationships in the water quality dataset. The ANN model also has shortcomings …

3d Parametric Hand To Improve Prosthetic Hand Functionality, Maria C. Gerardi

UNLV Theses, Dissertations, Professional Papers, and Capstones

The use of prosthetics can significantly enhance an individual’s standard of living, not only offering functional advantages but psychological advantages as well. Unfortunately for children with upper limb reduction, options are limited and rejection rates are high due to a multitude of reasons including discomfort and poor functionality. This study proposes a new parametric 3D design model (Parametric Hand) with an adjustable thumb that can be easily manipulated to the uniqueness of an individual. The Parametric Hand was evaluated, both qualitatively and quantitatively, against the Flexy-Hand 2, a commonly used prosthetic hand. The results showed insignificant differences in all testing …

The Dynamic Control Platform: Reinventing The Wheel, One Leg At A Time, Angel Javier Solis

UNLV Theses, Dissertations, Professional Papers, and Capstones

Upright bipedal walking is a complex balance of forces and actions that is almost taken for granted. How this system is modeled, how it affects a prosthesis, and how it can be implemented in the real world are topics that the proposed Dynamic Control Platform aims to address.

The Dynamic Control Platform (DCP) is a bipedal robot designed to test bio-inspired control algorithms with the aim to smooth out the walking experience for prosthetic legs. The main control paradigm that the DCP centers on the principle of orthogonal constraint, which aims to enforce a perpendicular relationship between the center of …

3-D Silk Fibroin Porous Particles Created By The Ouzo Effect For Biomedical Applications, Ashley Nicole Lamb

UNLV Theses, Dissertations, Professional Papers, and Capstones

Due to its high biocompatibility and biodegradability, silk fibroin – produced from

Bombyx mori (B. mori) cocoons – has been at the forefront of research for many

biomedical application formats: hydrogels, films, microspheres, and porous

sponges/scaffolding, to name a few. For drug delivery, in particular, porous particles are

desirable for their large surface area, uniform and tunable pore structure, and high

porosity. This thesis focuses on the fabrication of porous particles from silk fibroin by the

very interesting Ouzo effect. The Ouzo effect, so named because of the Greek

beverage ouzo, describes the phenomenon of an ethanol + anethole oil …

Front-End Cmos Transimpedance Amplifiers On A Silicon Photomultiplier Resistant To Fast Neutron Fluence, Mario Valles Montenegro

UNLV Theses, Dissertations, Professional Papers, and Capstones

Radiation hard electronics are indispensable in providing reliable diagnostics in fast radiation detection, which are necessary for new fundamental science research. In the case of radiation detection, transimpedance amplifiers are needed to magnify low photodetector signals to readable levels reliably. This thesis investigates a way to design a transimpedance CMOS-front-end transimpedance amplifier (TIA) as a first stage front-end amplifier. The TIA is to be mounted on silicon photomultiplier (SiPM), as a photodetector in fast neutron scintillation experiments capable of sustaining up to 10^15 n/cm of fast neutrons in the range of 0.1 to 20 MeV. The proposed TIA was designed …

Neural Network In Hardware, Jiong Si

UNLV Theses, Dissertations, Professional Papers, and Capstones

This dissertation describes the implementation of several neural networks built on a field programmable gate array (FPGA) and used to recognize a handwritten digit dataset – the Modified National Institute of Standards and Technology (MNIST) database. A novel hardwarefriendly activation function called the dynamic ReLU (D-ReLU) function is proposed. This activation function can decrease chip area and power of neural networks when compared to traditional activation functions at no cost to prediction accuracy.

The implementations of three neural networks on FPGA are presented: 2-layer online training fully-connected neural network, 3-layer offline training fully-connected neural network, and two solutions of Super-Skinny …

High-Speed Radhard Mega-Pixel Cis Camera For High-Energy Physics, Wenlan Wu

UNLV Theses, Dissertations, Professional Papers, and Capstones

This dissertation describes the schematic design, physical layout implementation, system-level hardware with FPGA firmware design, and testing of a camera-on-a-chip with a novel high-speed CMOS image sensor (CIS) architecture developed for a mega-pixel array. The novel features of the design include an innovative quadruple column-parallel readout (QCPRO) scheme with rolling shutter that increases pixel rate, its ability to program the frame rate and to tolerate Total Ionizing Dose effects (TID). Two versions of the architecture, a small (128 x 1,024 pixels) and large (768 x 1,024 pixels) version were designed and fabricated with a custom layout that does not include …

On Action Quality Assessment, Paritosh Parmar

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this dissertation, we tackle the task of quantifying the quality of actions, i.e., how well an

action was performed using computer vision. Existing methods used human body pose-based features to express the quality contained in an action sample. Human body pose estimation in actions such as sports actions, like diving and gymnastic vault, is particularly challenging, since the athletes undergo convoluted transformations while performing their routines. Moreover, pose-based features do not take into account visual cues such as water splash in diving. Visual cues are taken into account by human judges. In our first work, we show that using …

A Highly-Sensitive Global-Shutter Cmos Image Sensor With On-Chip Memory For Hundreds Of Kilo-Frames Per Second Scientific Experiments, Konstantinos Moutafis

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this work, a highly-sensitive global-shutter CMOS image sensor with on-chip memory that can capture up to 16 frames at speeds higher than 200kfps is presented. The sensor fabricated and tested is a 100 x 100 pixel sensor, and was designed to be expandable to a 1000 x 1000 pixel sensor using the same building blocks and similar architecture.

The heart of the sensor is the pixel. The pixel consists of 11 transistors (11T) and 2 MOSFET capacitors. A 6T front-end is followed by a Correlated Double Sampling (CDS) circuitry that includes 2 capacitors and a reset switch. The 4T …