Matching Items (726)
Filtering by
- Creators: Computer Science and Engineering Program
Description
This thesis presents the development and testing of a novel ultrasonic communication system for IoT applications, focusing on secure locks and underwater data transmission. A hybrid On-Off Keying (OOK) and Phase Shift Keying (PSK) modulation scheme is implemented on a 500 Hz carrier, combined with amplitude modulation on an ultrasonic carrier. The system proposes to address challenges such as multi-path propagation, noise sensitivity reduction, and clock synchronization, achieving a data rate of 18-24 bytes per second in low bandwidth mode under optimal conditions. Hardware-software co-design principles are applied using low-cost microcontrollers and analog circuitry. Experimental results demonstrate robustness of the proposed system when operating in controlled environments but at the same time highlight its limitations in noisy settings, especially with obstacles reflecting ultrasonic signals. Future work could focus on improving data retention and possibly integrating neural networks for waveform detection. Likewise, if time permits, construct a duplex link and test its underwater propagation range.
ContributorsNikitin, Evan (Author) / Meuth, Ryan (Thesis director) / Indela, Soumya (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
This paper presents the design and implementation of a data visualization system that connects REDCap, a secure data capture platform, with Tableau to support real-time tracking of participants in child mental health intervention research. Conducted within the ASU YFaCS Lab, the project replaces manual Excel-based workflows with automated dashboards that visualize recruitment, screening, and enrollment patterns. The system improves reporting efficiency, reduces human error, and supports Institutional Review Board (IRB) compliance. The final dashboard framework is modular, reusable, and scalable across other intervention studies and research contexts.
ContributorsBandi, Rajanandini (Author) / Osburn, Steven (Thesis director) / O'Hara, Karey (Committee member) / Torres, Marisela (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2025-05
Description
Have you ever been nervous to work out? FitPath is here for you! FitPath is a personalized fitness application that allows you to generate custom workouts that best fit your needs. You can also connect with other gym-goers to see their tracked workouts and meal plans, allowing you to learn and absorb lifestyles that fit you best! Join FitPath: Your Journey. Your Goals. Your Path.
ContributorsSingh, Yash (Author) / Raghu, Narendiran (Co-author) / Jutla, Pranav (Co-author) / Byrne, Jared (Thesis director) / Howell, Travis (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / Economics Program in CLAS (Contributor)
Created2025-05
Description
Understanding the factors that drive foraging decisions in worker honey bees (Apis mellifera L.) is a first step toward understanding bees’ collective behavior. In this study, we tested the hypothesis that a honey bee's decision to leave the hive to forage is best predicted by its age, followed in decreasing importance by a recent visit to the hive's "dance floor" where returned foragers communicate their findings, the time of day, and the time of year. Using spatial tracking data collected from an observation hive at the University of Konstanz, Germany, a data cleaning pipeline was developed and the best machine learning algorithm to predict individual foraging events based on these factors was identified. The Random Forest model achieved the highest predictive performance after addressing dataset imbalance, and provided insight into the relative importances of each factor. Our results disproved the central role of age and confirmed the significance of recent social contact in predicting foraging behavior, with time of day and seasonality contributing secondary effects. These findings deepen our understanding of honey bee decision-making and highlight the power of computational approaches in studying collective behavior.
ContributorsHamada, Diya (Author) / Daniels, Bryan (Thesis director) / Pavlic, Theodore (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
Food waste is a prominent issue today in both environmental and economic terms with households being among the top contributors. The rise of artificial intelligence (AI) in consumer technology has created new opportunities to address everyday challenges, such as food waste. Solutions are being developed to bridge the gap between sustainable living and accessible solutions to make a difference in people’s lives. This thesis explores the market viability and societal impact of FridgeScan AI, an artificial intelligence (AI) driven mobile application designed to help users manage the contents of their refrigerators. The application was designed to use image recognition and machine learning to scan food items, manage fridge contents, generate shopping lists, and provide recipe suggestions based on what users have in their fridge. This research builds on the FridgeScan AI capstone project developed at Arizona State University. It combines a literature review and a user survey to assess the current market for smart kitchen technologies, specifically user expectations and concerns. Key themes include sustainability, cost savings, and data privacy. The survey was distributed among 53 individuals aged 18 and over to collect insights on the value, features, and ethical considerations of such an application. The results from the survey then informed the development of a simplified business model that analyzes potential revenue and strategies for deployment. Ultimately, the goal is to assess user interest in tools that may help reduce food waste and improve household organization, and how their concerns around data collection and camera use could impact that. This thesis concludes by exploring both the potential and the challenges of adopting AI-based home applications and offers possible directions for further development and evaluation.
ContributorsFarias, Sabrina (Author) / Chavez Echeagaray, Maria Elena (Thesis director) / Lee, Quak Foo (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
Advancements in large language models (LLMs) have enabled the development of intelligent educational tools that support inquiry-based learning across technical domains. In cybersecurity education, where accuracy and safety are paramount, systems must go beyond surface-level relevance to provide information that is both trustworthy and domain-appropriate. To address this challenge, we introduce CyberBOT, a question-answering chatbot that leverages a retrieval-augmented generation (RAG) pipeline to incorporate contextual information from course-specific materials and validate responses using a domain-specific cybersecurity ontology. The ontology serves as a structured reasoning layer that constrains and verifies LLM-generated answers, reducing the risk of misleading or unsafe guidance. CyberBOT has been deployed in a large graduate-level course at Arizona State University (ASU), where more than one hundred students actively engage with the system through a dedicated web-based platform. Computational evaluations in lab environments highlight the potential capacity of CyberBOT, and a forthcoming field study will evaluate its pedagogical impact. By integrating structured domain reasoning with modern generative capabilities, CyberBOT illustrates a promising direction for developing reliable and curriculum-aligned AI applications in specialized educational contexts.
ContributorsDe Maria, Riccardo (Author) / Liu, Huan (Thesis director) / Agrawal, Garima (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
The efficient solution of large-scale linear systems, particularly those arising from sparse matrices, is fundamental to numerous applications in science, engineering, and machine learning. Direct methods, such as LU decomposition, offer robustness but face challenges related to computational cost and memory usage when applied naively to sparse problems, primarily due to the phenomenon of fill-in. This thesis investigates the practical performance characteristics of LU decomposition for sparse matrices, focusing on MATLAB's widely used built-in lu() function. A baseline comparison is first established by contrasting a custom LU implementation (myLU) designed for dense matrices with MATLAB's lu() applied to both dense and sparse inputs. Performance is evaluated based on execution time, numerical accuracy, and factor sparsity (fill-in).
Subsequently, the thesis explores a key enabler of sparse algorithm efficiency: data storage schemes. The impact of Coordinate (COO), Compressed Sparse Row (CSR), and Compressed Sparse Column (CSC) formats on the performance of the fundamental Sparse Matrix-Vector Multiplication (SpMV) operation is experimentally analyzed and compared against MATLAB's optimized built-in SpMV. Results demonstrate that MATLAB's sparsity-aware lu() function significantly outperforms dense approaches in both speed and memory efficiency, via controlled fill-in, when handling sparse matrices. Furthermore, the SpMV analysis confirms the superior performance of compressed storage formats (CSR/CSC) over COO, while highlighting the exceptional optimization of MATLAB's internal routines. Collectively, these findings underscore the critical importance of utilizing both sparsity-aware algorithms and efficient underlying data structures for tackling large-scale sparse linear systems effectively.
ContributorsVincentsundar, Vishal (Author) / Osburn, Steven (Thesis director) / Zhou, Ben (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2025-05
Description
This thesis addresses the need to introduce foundational computer science concepts (particularly coding and data structures) to elementary and middle school students through an AI-enhanced curriculum. Existing educational resources often neglect structured teaching of data structures which is a key element in computational thinking and problem-solving. This work bridges the educational gap by developing a curriculum that uses adaptive AI methodologies, gamification, and personalized instruction to create engaging and individual learning experiences.
The curriculum integrated interactive digital activities, adaptive learning paths, real-time feedback, and project-based scenarios. A proof of concept prototype demonstrates the practical application of AI-driven instructional techniques and highlights how adaptive challenges and personalized feedback can significantly enhance student engagement and conceptual understanding.
The thesis also discusses practical implementation strategies, assessment methodologies, and ethical considerations surrounding the use of AI in educational contexts. By highlighting both potential benefits and limitations, this research provides valuable insights and recommendations for educators, curriculum developers, and policymakers aiming to integrate advanced technological tools effectively in early computer science education.
ContributorsYoung, Danika (Author) / Zhu, Haolin (Thesis director) / Millman, Steven (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Computer Science and Engineering Program (Contributor) / School of Applied Sciences and Arts (Contributor) / Dean, W.P. Carey School of Business (Contributor)
Created2025-05
Description
This paper outlines the development of an efficient glossary navigation system that uses
artificial intelligence to provide instant results through a dynamic search bar. The project uses
several tools, including Python, Hypertext Markup Language (HTML), Cascading Style Sheets
(CSS), JavaScript, and the Google Gemini API. Market research was conducted to ensure the
glossary stands out from other products, as few competitors integrate AI. Unlike the traditional
long, continuous pages of most competitors, this project displays only the words, showing
definitions only when needed. The integration of the Google Gemini-powered dynamic search
engine was successful, achieving the goal of eliminating manual navigation and simplifying the user experience with a streamlined, quick-access layout.
ContributorsPatel, Samir (Author) / Osburn, Steven (Thesis director) / Pokidaylo, Boris (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2025-05
Description
Maps are crucial in emergency response operations, offering vital geographical data that aids teams in navigation and strategy. However, during catastrophic events like forest fires or floods, which can sever major routes, maps can quickly become outdated. In such scenarios, search and rescue teams and other first responders may find themselves scrambling to navigate and plan without dependable maps, complicating their mission.
Emergency response teams, specifically aerial imaging companies and wildfire teams, have a clear need to generate maps using aerial imaging. The conventional approach to synthesizing maps from aerial imagery is taking aerial images and accurately stitching them together, called an orthomosaic. Unfortunately, this process is both very expensive and slow. Highly accurate sensors and cameras required for real time map creation can cost hundreds of thousands of dollars. In other cases where this equipment is not available, producing maps can take overnight. This results in issues of inaccessibility and information arriving too slowly for effective emergency response.
This study investigates a cost effective, efficient solution utilizing standard cameras mounted on drones to capture aerial images and projecting them over satellite data in real time. By utilizing homography estimation image transforms, we evaluate two methods of image transformations: standard key point based matching and EXIF data camera extrinsics estimation. Both methods have strengths and weaknesses that inform future developments, including creating a hybrid model where the strengths of both methods are leveraged.
Ultimately, the work done in this experiment serves as a foundation for future research into achieving real time cost effective orthomosaic creation. Successfully implementing these methods would empower emergency responders with situational awareness enhancing the speed and accuracy of critical missions.
ContributorsKattenbraker, Luke (Author) / Nayak, Samik (Co-author) / Osburn, Steven (Thesis director) / O'Connor, Peter (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05