UTSA Runner Research Press (Univ. of Texas at San Antonio)
Not a member yet
6846 research outputs found
Sort by
Language Teacher Identity Impacts of English Teaching Policy and Ideology in Uzbekistan: Coteaching as Nexus of Practice
Full text linkThe U.S. Department of State funds exchange programs that send U.S. citizens to conduct English teaching activities with local Uzbek teachers of English in Uzbekistan. U.S. teaching assistants and Uzbek teachers of English orient toward language ideologies and exercise agency in accordance with their own identities as English teachers within these circulating discourses. Language ideologies evident in the policy discourse of these exchange programs as well as those of the Uzbek institutions circulate in their coteaching activities. These discourses in place and historical bodies influence and are influenced by interactions that take place in coteaching spaces. We remain under-informed about how the language ideologies evident in these discourses (across and among discourses in place, historical bodies, and interaction order) impact language teacher identity development. The purpose of this study is to explore these impacts of language ideologies evident in circulating discourses on language teacher identity development in the context of U.S. Department of State funded English coteaching in Uzbekistan. This ethnographic study applies nexus analysis as a conceptual framework to guide the focus and data collection of the study. Discourse analysis from critical discourse analysis, interactional linguistics, and linguistic anthropology are applied in analysis. Findings indicate impacts on professional identity development of participants in reinforcement or resistance of language ideologies. Professional identity development for teaching assistants involves development of agency and authority in their coteaching roles.Bicultural-Bilingual Studie
First-Year Experience Service-Learning Modality Differences Among First-Generation College Students
Full text linkThe purpose of this study was to investigate whether differential outcomes and barriers exist between in-person and virtual service-learning experiences for first-generation college (FGC) students enrolled in a first-year experience (FYE) course at a Hispanic Serving Institution (HSI) in south Texas. While previous literature has explored service-learning, this study fills a gap through comparison of modality differences within this student population exploring service-learning match experiences through flexibility. Guided by cultural mismatch theory and experiential learning theory, this convergent mixed methods parallel case study collected quantitative and qualitative data, analyzed them independently, and merged findings through triangulation to gain a deeper understanding of student perspectives. For quantitative data collection pre- and post- surveys were implemented and measured perceptions of barriers, community building, and educational outcomes. Quantitative results revealed that there were significant differences between the pre- and post- surveys on the barriers of transportation and time, however no significant differences emerged between the two modalities. Qualitative data created narratives from interviews and reflections, where five themes developed that were challenges, community connection, real-world connection, motivation, and positive experience. Triangulation showed that both in-person and online service participants demonstrated strong perception of community building and community connection; and educational outcomes and real-world connections. It also showed that barriers and challenges including time and transportation did exist for some participants, and providing online option created access mitigating those barriers. The findings suggest that both in-person and virtual service-learning experiences can provide meaningful educational and developmental benefits when implemented with flexibility and cultural awareness.Interdisciplinary Learning and Teachin
Securing the IoT Edge Cloud Model
Full text linkThe Internet of Things has evolved from isolated devices into cloud-based ecosystems that enable automation but widen attack surfaces, centralize data, and embed rigid trust and identity models. This dissertation advances security in edge networks while preserving privacy and function. Bulwark returns data stewardship to users through permissioned control of data generation, access, and storage, provisioning user-owned cloud resources and employing a containerized pipeline that processes sensor and video data near the source, discards non-actionable content, and forwards results to the cloud. Experiments in video and smart-home security reduce traffic and storage and validate policy-driven control. Building on this foundation, Snap IoT applies decentralized identity and zero-trust principles to enable self-assembling edge networks. A residential HVAC deployment increases sensing density, improves control accuracy, and limits disclosure through distributed edge computation. Complementing these designs, an adversarial analysis shows cloud services, combined with Domain Name System (DNS) can leak metadata when payloads are encrypted; by correlating endpoints, timing, and throughput, it groups communication streams, infers device roles, and links traffic to manufacturers. In response, Roadblock builds on these findings to enforce subnet isolation by sensor type, one-way egress, local decision making at the middle tier, and deletion of raw data, achieving substantial egress reductions, strong containment, and automated remediation in evaluations. Finally, Beacon evaluates interpretable models for edge land-use classification, quantifies the accuracy–transparency trade-off, and motivates paired deployments that preserve auditability without sacrificing performance. Together, these contributions identify critical leakage paths, return data ownership and control to users, validate decentralized and isolated edge designs, and deliver practical mechanisms that operate on low-cost hardware across heterogeneous deployments, thereby strengthening user security while preserving privacy.Computer Scienc
C 语言复合内存安全修改的性能分析
Full text linkMemory errors have been around for the entirety of computing, starting as innocent mistakes on shared systems and eventually being leveraged for malicious reasons. These issues are more pronounced in C and C++ and thus there has been a widespread effort in the cybersecurity community to introduce protections to these languages (and associated compilers) to prevent memory errors. Many solutions proposed in the research literature are fairly effective in mitigating these problems. However, often the penalties of using these mechanisms do not provide the necessary insight for practitioners to make informed decisions on where to put their efforts. This thesis increases the scope of comparison to include modern and natively memory-safe languages when comparing performance penalties incurred by protection mechanisms for C/C++ programs. Standardized abstract tasks that can be implemented across languages give a more complete picture as to the true penalties of attempting to backport memory safety protections to a language that is unsafe by design, rather than using a language that had memory safety built in from the start. This thesis reviews the plausibility of using combinations of state-of-the-art memory safety mechanisms to achieve greater breadth of protection comparable to native memory safe languages while reviewing the resulting performance penalties of those combinations. Most protections did not lose their provided protections when combined with each other but performance degradation of these combinations does not scale linearly and is highly dependent on the pattern of memory usage.内存错误贯穿了整个计算机发展史，最初是在共享系统上的无心失误，最终被用于恶意目的。这些问题在 C 和 C++ 中尤为突出，因此网络安全领域广泛致力于为这些语言（及其相关编译器）引入防护，以防止内存错误。研究文献中提出的许多方案在缓解这些问题方面相当有效。然而，这些机制的使用代价往往缺乏足够的参考信息，难以为实践者提供关于应将精力投入何处的明智指引。
本文在比较 C/C++ 程序的内存安全防护机制所带来的性能开销时，将比较范围扩展到纳入现代的、原生内存安全的语言。可在多种语言中实现的标准化抽象任务，能够更完整地展现这样做的真实代价：与其在一种从设计上就不安全的语言中回填内存安全防护，不如从一开始就使用内置内存安全的语言。本文还评估了将多种最先进的内存安全机制进行组合的可行性，旨在获得可与原生内存安全语言相当的更广泛防护，并考察这些组合由此带来的性能开销。大多数防护在相互组合后并未削弱其保护效果，但由此产生的性能下降并非线性叠加，而且高度依赖于内存使用模式。Computer Scienc
STUDY OF ENERGY STORAGE AND SUPPLY STRATEGIES FOR ADVANCING RESILIENCE AND LONG-TERM COMFORT IN LOW-INCOME COMMUNITIES DURING TEXAS COLD SNAPS
No full textThe full text of this item is not available at this time because the author has placed this item under an embargo until December 11, 2027.Texas, characterized by a predominantly hot and humid climate, has historically focused on cooling-oriented building design. However, recent cold events, most notably Winter Storm Uri (2021) and subsequent winter storms in 2023 and 2025, have exposed severe vulnerabilities in the state’s residential infrastructure, particularly within low-income communities in Southwest San Antonio. Many of these households lack adequate insulation and rely on inefficient heating systems, leaving occupants exposed to unsafe indoor temperatures during prolonged power outages.
The purpose of this study was to evaluate strategies that enhance energy resilience and thermal comfort during extreme winter conditions. Using EnergyPlus, DesignBuilder, TESPy, and OEMOF, four primary scenarios were simulated under 2050 projected weather data: (1) passive survivability of existing homes, (2) decentralized photovoltaic and battery systems, (3) district-scale Ground Source Heat Pump (GSHP) with microgrid integration, and (4) code-compliant insulation upgrades based on IECC 2021 and ASHRAE 90.1-2021 standards.
Results revealed that baseline homes experienced rapid indoor temperature drops below 10 °C within three days, violating the WHO’s recommended minimum indoor threshold of 18 °C. Adding rooftop PV and batteries moderately improved performance, while the district-scale GSHP–microgrid system maintained indoor comfort above 18 °C throughout a five-day outage. Incorporating code-compliant insulation further extended survivability by nearly two days and reduced total system costs when combined with renewable generation.
The study concludes that community-based hybrid energy systems integrating renewable power, thermal storage, and improved insulation offer the most robust, equitable, and cost-effective pathway toward sustainable and climate-resilient housing for vulnerable Texas communities.Architectur
SOFTONIC: A Photonic Design Approach to Softmax Activation for High-Speed Fully Analog AI Acceleration
Full text linkRecent advancements in Silicon Photonics (SiPh)-based AI accelerators offer promising solutions to address the energy bottlenecks of executing large models, such as Transformers. However, existing SiPh solutions primarily focus on accelerating matrix multiplication (MM) in the photonic domain, while the Softmax Activation (SMA) function—responsible for 30-40% of the total computation in Transformers—is still handled by conventional digital platforms. This reliance on digital platforms introduces significant energy and latency overheads due to the frequent conversions between photonic MM and digital SMA.
While nonlinear functions like ReLU, Sigmoid, Tanh, and Softplus have been successfully implemented in the photonic domain using components such as optical amplifiers, Mach-Zehnder modulators (MZMs), and microring resonators (MRRs), similar approaches are not suitable for SMA. The excessive area and energy overhead introduced by optical amplifiers, along with the inability of MRRs and MZMs to handle SMA’s exponential and division functions, make it difficult to realize a SiPh-based SMA.
To address these challenges, we propose SOFTONIC, a first-of-its-kind photonic SMA architecture designed for ultra-high energy efficiency and speedup in Transformer acceleration. SOFTONIC integrates a microdisk (MD)-based vector decomposition unit, a novel photonic polynomial unit with a range reduction technique, and a unique SiPh lookup table. Simulations of SOFTONIC using industry-standard CAD tools and AI workloads demonstrate a (i) 109× improvement in latency; (ii) an 80% reduction in power consumption; (iii) upto 199 × higher compute-density compared to leading digital and analog Softmax hardware solutions. These improvements come with an area overhead of 0.0157 mm2, highlighting the potential of SOFTONIC for efficient, large-scale photonic AI acceleration.Electrical and Computer Engineerin
(Re)Humanizing Pedagogy and Zone of Proximal Development: A Mixed Methods Study Investigating Mathematics CoRequisite Students
Full text linkThe purpose of this convergent parallel, one-phase, mixed methods study was to investigate non-TSI ready mathematics student outcomes and actual lived experiences for those who took a mathematics CoRequisite course at one south-central Texas, Hispanic serving community college from Fall 2018-Spring 2024. After quantitative and qualitative data collection and analysis were conducted in-parallel, a side-by-side comparison approach was carried out to converge the results to synthesize overall findings. The multiple regression models developed suggested no practical nor statistical significance in predicting a student’s final course grade. Meaning, the student demographic factors of TSIA/TSIA2 mathematics score, race/ethnicity, and gender did not influence a student’s final course grade outcome. Also, the phenomenological, instrumental case study themes surfaced it was the use of Vygotsky’s (1978) sociocultural concept of Zone of Proximal Development (ZPD) and theorized to be one, or a combination, of the eight rehumanizing practices of Gutiérrez (2018) that allowed each student to self-perceive what success/failure meant to them, and/or how they were able to be successful in their mathematics CoRequisite (CoReq) course and life beyond their mathematics CoReq classroom. Mathematics CoReq classes provided non-TSI ready mathematics students with an experience of connection, joy, and belonging within their tertiary educational journeys. These conclusions would not have been supported if stand-a-lone quantitative or qualitative studies were conducted, which is why mixed methods are warranted for this topic of investigation. Overall, this dissertation research study was used to create a space to continue to (re)surface the efforts of how one could provide non-TSI ready mathematics students with access to a quality tertiary educational classroom to learn and continue to thrive in.Interdisciplinary Learning and Teachin
Structural Transformation and Physical Properties of Oxide Materials by Using In-Situ Transmission Electron Microscopy
Full text linkOxide ceramics and their compounds have been widely used across diverse industrial applications while also being recognized as promising candidates for cosmic dust formation in circumstellar environments. Morphological, compositional and structural characterizations of these oxides are crucial for understanding their fundamental structure-property relationships at the nanoscale. Their crystallization and phase transformation from metastable or amorphous states can occur under different energetic conditions—thermal processing, ion-irradiation, energetic particles, UV photons etc. To study these phenomena, in-situ TEM observations at nanoscale resolution coupled with electron pair distribution function (ePDF) analysis can provide comprehensive structural insights. This technique can capture the dynamics of structural evolution and provide time-resolved crystallization data from a nearly identical region of the analyzing materials. Therefore, in this dissertation, the combination of in-situ and ex-situ TEM techniques with ePDF, EELS, and off-axis electron holography introduces a comprehensive characterization approach to study oxide materials crystallization process, structural evolution, phase transformation, and physical properties. By applying this methodology to cosmic dust analogs ceramics oxide nanograins, several key aspects have been investigated: local structure of amorphous phase, quantitative structural evolution, electron dose-dependent structural changes, metastable phase, threshold electron dose for crystallization and the underlying mechanism, extrapolating crystallization timescale in energetic astrophysical environments, and their dielectric and optical properties. In addition, in-situ ferroelectric properties of perovskite oxides resulting from thermal-induced (low temperature) phase transformation have been investigated by using electron holography. Ferroelectric properties were measured from nanograins by measuring electrostatic potential extracted from the reconstructed phase obtained through holography, despite dynamic diffraction effects in TEM still being a challenge. Simultaneously registered electron diffraction patterns have been analyzed and compared with simulated patterns to validate the crystalline structural transformation at low temperatures.Physics and Astronom
Performance Optimization and AI Integration in Mobile Health Applications: A Case Study of the QuitTxt Smoking Cessation Platform
Full text linkMobile health applications face significant challenges in performance optimization and the integration of artificial intelligence. This thesis investigates these challenges through the development and evaluation of QuitTxt, a smoking-cessation application built with Flutter and Firebase.
The research addresses three primary objectives: (1) systematic performance optimization, (2) integration of large language models (LLMs) using Retrieval-Augmented Generation (RAG), and (3) validation of improvements in both performance and AI capability. The performance engineering process produced substantial results—message sorting improved by 94.8% (reducing latency from 420 ms to 22 ms), application startup time decreased by 54.7%, and memory usage stabilized at under 150 MB.
The AI component introduced a RAG system tailored for therapeutic contexts. By extracting clinical content from established smoking-cessation protocols and applying a relevance-scoring algorithm, the system demonstrated high precision in retrieving clinically appropriate material. Expert evaluators assigned an average score of 4.6/5.0 across five dimensions: protocol alignment, therapeutic appropriateness, empathy, actionability, and conversational naturalness.
Multiple RAG optimization strategies were explored, revealing that professionally curated Q&A pairs (“Dataset RAG”) produced the strongest results. This approach achieved a BLEU score of 0.2915, representing a 271.6% improvement over raw web content and a 683.6% improvement over baseline retrieval. These findings highlight the value of structured preprocessing for enhancing AI system performance.
Overall, this work contributes practical methods for performance engineering in cross-platform mobile health applications and demonstrates that RAG architectures can preserve clinical protocol fidelity while leveraging the conversational strengths of modern language models.Electrical and Computer Engineerin
Destiny and Disorder: Prophecy and Politics in Shakespeare's Richard II
Full text linkThis thesis explores into the extent to which Shakespeare’s Richard II reinforced or challenged Tudor political ideology through its use of prophecy. It examines how these prophetic utterances influenced debates about monarchy, succession, and legitimacy during Elizabeth I’s reign. While some argue that Shakespeare’s history plays perpetuate Tudor propaganda, others argue that his portrayal of providence and kingship is more nuanced, allowing for ambiguity and ideological tension rather than simple endorsement.
Richard II delves into the precarious nature of kingship and the clash between divine right and pragmatic power, all set against the backdrop of competing prophecies that frame the deposition crisis. I examine how the prophetic discourse in the play intersects with Elizabethan political pageantry and the Tudor Myth, a providential narrative aimed at reconciling Richard II’s disposition with Tudor stability. My objective was to assess the validity and limitations of the Tudor Myth as applied to Richard II, through a meticulous analysis of its prophetic speeches and ideological implications. John of Gaunt, Richard II, Bolingbroke, and Bishop of Carlisle not only dramatize the Tudor Myth but also complicate it simultaneously. The binary between providential order and political disorder can be both reinforced and dismantled by grounding Tudor historiography and challenging it ideologically, thereby confronting Elizabethan anxieties about monarchy and succession.Englis