MSU Libraries Digital Repository (Michigan State University)
Not a member yet
106711 research outputs found
Sort by
TOWARD RELIABLE AND SCALABLE LONG RANGE NETWORKING FOR RURAL IOT
Full text linkThesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2025The Internet of Things (IoT) holds great promise for transforming rural applications such as precision agriculture, infrastructure monitoring, and environmental sensing. However, enabling reliable and scalable wireless connectivity in rural areas remains a fundamental challenge due to sparse infrastructure, energy constraints, and wide-area coverage requirements. This dissertation presents a system-level exploration into building practical, low-power, and cost-effective LoRa-based networks tailored for rural IoT.Specifically, I address four key challenges: (1) unreliable backhaul due to the absence of cellular or wired infrastructure, (2) weak LoRa signal coverage in complex and obstructed rural terrains, (3) limited scalability of existing backscatter systems for battery-free communication, and (4) inflexible physical layer encoding that fails to meet the diverse demands of rural applications. I propose and validate a series of novel techniques, including opportunistic satellite backhaul using lightweight link estimation and routing, polarization-aligned underground communication for cross-soil sensing, concurrent non-linear chirp backscatter for scalable battery-free transmission, and reconfigurable chirp encoding for adaptive coverage, throughput, and energy balancing. These techniques are evaluated through real-world deployments, hardware prototypes, and empirical experiments across rural-scale testbeds.Together, these contributions advance the design of robust and scalable LoRa networks for rural IoT. Looking forward, this work motivates future research on integrating space\u2013air\u2013ground architectures, embedding joint sensing and communication capabilities, and co-designing cross-layer protocols that adapt to the dynamic and heterogeneous nature of rural environments. This dissertation lays the foundation for next-generation rural IoT systems that are not only technically efficient but also practically deployable across underserved regions.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
EXPLICIT TRAINING PRACTICES FOR DEVELOPMENTALLY BASED BEHAVIORAL INTERVENTIONS
Full text linkThesis (Ph.D.)--Michigan State University. Special Education - Doctor of Philosophy, 2025Effective implementation of behavior analytic practices for children with autism spectrum disorders (ASD) often requires explicit training. Although there is a need for more training in many human services fields, barriers such as costs and regulations may hinder the development of effective training practices (Larson et al., 2005). The absence of explicit training practices, particularly in naturalistic behavioral interventions (NDBIs) for children with ASD may lead to increased negative interactions with clients and their families (Jimenez-Gomez et al., 2019, Rose, 2020; Rohrer & Weiss, 2023; Rohrer et al., 2021; Taylor et al., 2019). Therefore, providers of NDBIs need to assess cheaper, yet effective ways of training providers to implement their practices. The purpose of this dissertation was to assess the current training practices and trends reported in evidence-based naturalistic interventions. Three interrelated, yet separate studies were conducted.Chapter two is a scoping review of the reported training practices and strategies in NDBIs literature from 2000-2024. This review aimed to assess how providers are trained to implement evidence-based naturalistic interventions such as NDBIs. This study extends previous NDBI reviews as it adds pertinent information regarding training that may aid in the dissemination of NDBI practices. Chapter three assessed a training package aimed at training providers to implement an NDBI known as reciprocal imitation training (RIT). Behavior technicians (BT) underwent a behavior skills training (BST) training protocol to increase implementation fidelity percentages. The results of this study indicated that BST was a cost effective and simple strategy that could be used to train BTs to implement RIT. Chapter four assessed changes in children\u2019s spontaneous behaviors and responses to adult behavior following a change in adult PRIDE behaviors during a naturalistic intervention session. A reversal design was implemented to assess any change in child behaviors when an adult doubled the number of PRIDE behaviors expressed during a naturalistic teaching session. This study emulates a training study by expressing simulated \u201cpre-training\u201d and \u201cpost-training\u201d frequencies of PRIDE behaviors, thus informing of the possible benefits of training providers to engage in higher levels of PRIDE behaviors. This study extends parent-child interaction therapy (PCIT) literature. Chapter five summarizes and integrates themes witnessed across studies including the use of BST components and the focus on parent training procedures observed in NDBI literature.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
The Influencer Dissonance Model
Full text linkThesis (Ph.D.)--Michigan State University. Communication - Doctor of Philosophy, 2025This research introduces the Influencer Dissonance Model (IDM) to explain how influencer attributes, group conformity, psychological discomfort, and task type compliance interact to drive attitudinal shifts that ultimately influence purchase intention in followers exposed to counter-normative recommendations from social media influencers (SMIs). Through a series of experimental studies, this work explains why and how unconventional recommendations are an effective way for influencers to change follower attitudes and incite behaviors that ultimately have the power to generate revenue for brands. This research builds on the Social Identity Model of Deindividuation Effects (SIDE model) and Vicarious Dissonance Theory (VDT) to develop a new model that explains the underlying mechanisms driving attitudinal shifts, specifically in online influencer contexts. Results suggest that influencer type (micro- vs. celebrity-influencer) may no longer play a significant role in today\u2019s online influence landscape. Additionally, this study validates two new constructs\u2014influencer attributes and group conformity\u2014for use specifically in influencer contexts. Notably, counter to classic VDT findings, which suggest that individuals who perceive free will in task completion exhibit stronger attitudinal shifts, this research found that pressured sharing in online environments triggers stronger attitudinal shifts, which in turn predict purchase intention.These findings offer novel insights into the psychological dynamics of online influence and provide recommendations for influencers, marketers, and brands navigating the ethical and strategic implications of counter-normative recommendations in online spaces.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
Transnational Immigrant Youth Navigating Languages, Literacies, and Learning in a U.S. High School
No full textThesis (Ph.D.)--Michigan State University. Curriculum, Instruction, and Teacher Education - Doctor of Philosophy, 2025Although immigrant youth are the fastest growing population within the U.S. Schools in recent years, schools and educators have increasingly faced challenges in addressing their unique educational needs. This study explores the language, literacy and educational experiences of transnational immigrant youth studying in a U.S. high school. Informed by sociocultural perspective on literacy and transnationalism as theoretical frameworks and answers two research questions: (1) What are the discourses circulating within a high school about immigrant youth, and how do these discourses shape perceptions of their language, literacy, and learning?, and (2) In what ways do immigrant youth engage in language and literacy practices, and how do they mobilize these practices to navigate school and learning? Data were collected through participant observation and fieldnotes in the school, interviews with focal students and their teachers, and analysis of student and teacher-generated documents and artifacts.Findings showed that while teachers at the focal high school expressed appreciation for the perseverance and dedication of immigrant youth, their narratives were dominated by deficit discourses that positioned these students as academically behind due to limited English proficiency. Despite the school\u2019s linguistic diversity, the school did not have any systematic professional development or institutional mechanism to support teachers to serve these students, resulting in the overburdening of the ESL teacher and paraprofessional, who were expected to support both language development and content instruction. In contrast to this institutional neglect, immigrant youth actively engaged in meaning-making and community-building beyond school by utilizing digital platforms such as social media and online gaming. These digital spaces enabled them to sustain transnational ties, share cultural and linguistic resources, and cultivate a sense of belonging, demonstrating their agency and resilience in navigating their educational and social worlds.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
Face Modeling Under Diverse Illuminations
No full textThesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 20253D face modeling is a longstanding problem within the computer vision and computer graphicscommunities with widespread applications in AR/VR, movie production, and entertainment that involves a vast array of problems including modeling facial appearance, shape, expressions, lighting, and pose. One prominent issue in this domain is the inability to properly model non-Lambertian lighting effects, including hard shadows caused by directional lights. To this end, we propose two methods to better handle facial hard shadows. The first leverages physically-inspired supervision along the hard shadow boundaries to encourage the model to focus on properly synthesizing these regions. The second method pushes one step further and directly leverages the 3D face to physically constrain the generated hard shadows and guarantees shadow geometric consistency with respect to the face. Beyond facial shadows, which follow the geometric structure of the face, foreign shadows caused by external objects (e.g. hats, sunglasses, and leaves) are another consideration when modeling in-the-wild faces. We thus propose a self-supervised foreign shadow removal method that formulates the problem as an image-inpainting problem and properly removes foreign shadows while preserving the subject\u2019s facial details. However, all of these proposed methods do not have full control over facial illumination. For example, our shadow removal method, while able to completely remove shadows, is not able to control the degree of shadow softening and cannot perform the converse shadow intensifying operation. Moreover, our directional relighting methods represent light as a direction and are not able to control light size, rendering them unable to model the associated lighting effects. We thus propose COMPOSE, the first single-image portrait shadow editing method with full control over all shadow attributes, including shadow intensity, shape, and position that also preserves other attributes of the portrait\u2019s original environmental lighting. This is achieved by decomposing the problem into an environment-preserving diffuse image estimation and subsequently estimating a shadowed image where the shadow parameters are varied. By performing image compositing between the diffuse and shadowed images, COMPOSE precisely controls all shadow attributes during illumination editing. Moving beyond illumination editing, we investigate the remainder of the face modeling problem in editing appearance, shape, expressions, and pose. We first propose INFAMOUS-NeRF, an implicit face modeling method that introduces hypernetworks to estimate subject-specific model weights, thus alleviating the burden of the NeRF MLP to encode all subject information in a single shared set of weights. While improving representation power with subject-specific weights, INFAMOUS-NeRF also maintains editability by encouraging subjects with similar attributes (e.g. same expression) to share the same latent code for those attributes, which maintains semantic alignment of latent spaces. Finally, we investigate the tradeoff between representation power and efficiency. Gaussian splatting methods attempt to resolve this tradeoff by offering both high representation power and efficiency with real-time rendering. However, their efficiency nonetheless depends on the number of rasterized gaussians. Moreover, the time needed to optimize/prepare the avatar (enrollment time) is often high. We therefore propose EGGHead: a novel face modeling method that, given a fixed gaussian budget, efficiently assigns gaussians to different semantic regions of the avatar. Our algorithm reduces the total number of gaussians needed while minimizing the impact on representation power. Our pipeline operates in a single forward pass, and achieves SoTA reconstruction quality, novel view synthesis quality, and enrollment/rendering speeds compared to other single-image face modeling methods.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
IMPROVED COOLING OF HIGH-SPEED AXIAL FLUX PERMANENT MAGNET MACHINES USING SOFT MAGNETIC COMPOSITES
No full textThesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2025High-speed machines are desirable due to their light weight and small size. However,high-speed machines present several challenges. First, stresses increase with rotor speed. Second, losses increase with rotor speed. Finally, the reduced size provides less surface area for cooling, despite their increased losses. Much work has been done to solve these three problems. However, while direct cooling generally proves superior to indirect liquid cooling, the coolant in a direct cooling system is impractical to exclude from the air gap, and fluid in the air gap of a high-speed machine results in added drag, which can be significant at high speeds. This thesis presents a novel cooling system for high-speed, axial-flux machines, using SMC cores to improve upon existing cooling methods by combinig the benefits of direct and indirect liquid cooling. Because SMC cores are homogeneous, they are liquid-tight. SMC cores can also be formed into shapes impossible with laminated steels. By integrating cooling channels inside of an SMC stator core, coolant can be circulated in direct contact with the core while still containing the coolant from the air gap, combining the advantages of direct and indirect cooling. A machine is designed and teseted experimentally, and the data used to adjust the finite-element analysis (FEA) model parameters. The FEA model was then used to compare the prototype machine with the novel cooling system to a machine designed with laminated steel with an attached cooling plate. The results of the comparison demonstrate the greater efficacy of the novel cooling system compared to the traditional design with an attached cooling plate. In order for the traditional design to meet the same maximum coil temperature as the SMC design, the attached cooling plate would require at least 4.4MPa contact pressure, equal to 2.8kN at the machine diameter. Even with this contact pressure, the SMC machine was still smaller while producing the same torque (1.65Nm L vs. 1.57Nm L ), even though the SMC design had more total loss. The SMC design rejected roughly 20% more heat per air gap surface area (49.6 mW mm2 vs. 40.9 mW mm2 ) than the laminated steel design. The numerical results demonstrate the potential of SMCs to provide improved cooling in high-speed machines. Combined with their reduced losses and manufacturing advantages, SMCs are a viable option for motor cores for high-speed electric machines.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
BUILDING TRUSTING CONNECTIONS : THREE PAPERS EXPLORING TRUST AND CULTURALLY RESPONSIVE PRACTICE
No full textThesis (Ph.D.)--Michigan State University. School Psychology - Doctor of Philosophy, 2025While student populations in the United States have become increasingly culturally and linguistically diverse, disparities in academic achievement and psychological well-being persist between culturally and linguistically diverse (CLD) youth and their White peers. As a result, school practitioners and educators are called to engage in practices that will promote the equitable success and well-being of all students. Two methods for promoting student outcomes are culturally responsive practices and fostering trusting school relationships. This dissertation consists of three papers which collectively explore how culturally responsive practices and trust may be leveraged to support students. The first paper utilized interpretative phenomenological analysis to understand how refugee youth conceptualize the systems they access for services. A primary finding of the first paper is that trust is a key determinant of if and when refugee youth seek support. The second paper is a conceptual manuscript that synthesizes the literature on culturally responsive practices in school website design. The primary product of the second paper is a tool for schools to use to evaluate and improve their websites to make them more culturally responsive. The third paper examines how students\u2019 trust in their teachers and peers relates to their psychological well-being (i.e., symptoms of depression and anxiety) and academic-oriented psychosocial perceptions (i.e., behavioral and emotional engagement with learning, hope, GPA, grit, and school belonging). Results indicated that trust in peers and teachers did not meaningfully account for a significant amount of variance in students\u2019 psychological well-being. However, students\u2019 trust in their teachers did meaningfully explain their behavioral and emotional engagement with learning and school belonging. The limitations, future directions, and suggestions for practice of these individual papers are also provided.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
FAST DATA-DRIVEN FRAMEWORKS FOR SOLVING THREE DIMENSIONAL SCATTERING PROBLEMS
No full textThesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2025Scattering problems sit at the center of many practical applications involving some form of wave-based physics. These problems can broadly be categorized as forward and inverse scattering problems. Given their wide applicability, numerous numerical methods have been proposed over the last few decades for both classes of problems. A common limitation in these methods is the computational cost which renders them unpractical for time-sensitive applications. To overcome this bottleneck, several machine learning frameworks have been proposed recently. However, most of these work are focused on the two dimensional (2D) scattering problems. Efficient data-driven methods for the three dimensional (3D) version of the problem are yet to be explored. In this thesis, we develop data-driven deep learning models to solve 3D scattering problems in acoustics and electromagnetics. First, we seek to explore the advantages and disadvantages of using a common spherical harmonic representation for shapes and scattered data to learn solutions to the forward scattering problem. To this end, we first compute the acoustic scattering data at different frequencies for a large collection of 3D particles. Then, both the shape and scattering data of these particles are embedded into a series of spherical harmonic coefficients. A residual neural network is finally trained to learn the mapping between these two representations. We discuss the results and compare the proposed framework to methods based on spatial point-cloud representations.Next, we present a data-driven framework to perform fast 3D shape reconstruction from acoustic scattering data. The framework is implemented by (a) using a compact probabilistic shape latent space learned by a 3D variational auto-encoder, and (b) a convolutional neural network trained to extract far-field features due to multiple incident waves, and map the acoustic scattering information to this shape representation. We demonstrate the proposed framework's 3D shape reconstruction capabilities on random rock-like particles and airplane objects from the popular ShapeNet data set. We further evaluate the framework's performance, specifically testing its robustness when trained with lower-resolution scattering data, and when both the scattered data and receiver locations are affected by noise.Lastly, we present the extension of the proposed framework to electromagnetic inverse scattering problems. We extensively evaluate the proposed method against the state-of-the-art machine learning model. Our experiments demonstrate that the proposed method successfully reconstructs 3D shapes of complex scatterer geometries from ShapeNet. It is also robust to noise and achieves similar or better reconstruction quality than the state-of-the-art 3D inverse scattering method, while being orders of magnitude faster at inference.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
SALMONELLA TYPHIMURIUM-SPECIFIC SIGNATURES AS DIAGNOSTIC TARGET
No full textThesis (M.S.)--Michigan State University. Comparative Medicine and Integrative Biology - Master of Science, 2025Salmonella Typhimurium is most frequently associated with foodborne outbreaks. Standard microbial cultivation methods along with PCR are a gold standard for the detection of this pathogen, but these are labor intensive and take about 72 hours for a confirmatory diagnosis. Thus, rapid, specific, and cost-effective assays to detect and differentiate Salmonella infection from other foodborne pathogens are direly needed for early recognition to help curtail outbreaks and improve timely clinical interventions. The objective of this study is to select aptamers specifically binding to OPS S. Typhimurium using one round selection of SELEX (Systematic Evolution of Ligands by Exponential Enrichment) and also to validate its sensitivity (limit of detection) and specificity (failure to bind to non-target cells). Briefly, oligopolysaccaride (OPS) of Salmonella Typhimurium were extracted and used for a one step aptamer selection. The selected candidate aptamers were validated for binding to OPS of S. Typhimurium using a dot blot assay. Aptamer bound to OPS dots were eluted, reamplified and cloned for DNA sequencing of individual candidates. Of 19 OPS-specific aptamer clones, two redundant aptamer sequences were identified (aptamer N23). This candidate aptamer (N23) was characterized and a dissociation constant (Kd) of 58.70nM was obtained. Using dot blot assays, the sensitivity (limit of detection) of OPS S. Typhimurium was determined to be 0.8pg that equivalents to 200 cells of S. Typhimurium. To establish analytical specificity, aptamer N23 binding was tested against OPS and LPS of S. Typhimurium, as well as the LPS from different foodborne pathogens, including; E. coli O157, E. coli O111: B4, S. Minnesota, S. Enteritidis and Campylobacter jejuni. The results showed aptamer N23 had specificity to OPS and LPS of S. Typhimurium as well as LPS S. Enteritidis. This study demonstrated that the ssDNA aptamer selected by one round selection method which was designed to bind specific target, possessed a greater level of accuracy than previously selected aptamers.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
Investigations Toward Power Ramp-up of FRIB and Increasing Scientific Reach
No full textThesis (Ph.D.)--Michigan State University. Physics - Doctor of Philosophy, 2025The Facility for Rare Isotope Beams (FRIB) is a state-of-the-art nuclear physics research facility. FRIB contains two superconducting radiofrequency (SRF) heavy-ion linear accelerators: the FRIB driver linac, and the FRIB Re-Accelerator (ReA). The ultimate purpose of the driver linac is to safely and reliably deliver beams with an unprecedented power of 400~kW to the target. This thesis presents studies that contribute to the goal of achieving this beam power as well as increasing the scientific capabilities of both SRF accelerators at FRIB. These studies include research and development of room-temperature RF cavities, code development, creating new simulation models, updating existing models, and validating with beam measurements.A critical problem of the power ramp-up process is mitigating beam losses. In the past, the main criterion for a low-loss accelerator was low radioactivation of the equipment and the possibility of hands-on maintenance, and the beam loss rate of 1 W/m has been a rule of thumb applicable for room-temperature accelerators up to 1 GeV proton energy. In a superconducting linac, a lost fraction of the beam can significantly degrade crucial components of the accelerator, such as SRF cavities, and more strict requirements for the beam loss rate should be adapted. This thesis explores ways to mitigate losses caused by changes in the calibration of SRF cavities and longitudinal beam halos from the liquid lithium charge stripper. It discusses dual-charge-state acceleration of heavy ion beams in the FRIB radiofrequency quadrupole (RFQ).Studies presented in the thesis are ways to increase the scientific reach of both the FRIB driver linac and ReA. The design of a chopper system to allow for clean time-of-flight measurements in ReA is discussed. Also explored is the improvement of simulation models for simultaneous multi-charge state beam transport in the FRIB driver linac bending sections and the beam measurements that validate these models.The research discussed in this thesis has led to the complete multi-physics design of three room-temperature RF cavities. An application to quickly calculate synchronous phases of SRF cavities has been developed and implemented. Studies of the simulation models of the FRIB driver linac made them more accurate.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references