Computer Engineering Commons^™

Low-Resource Ecoacoustic Audio Classification, Enis Berk Coban

Dissertations, Theses, and Capstone Projects

Ecoacoustic monitoring via machine learning enables scalable analysis but is often constrained by labeled data scarcity, particularly in remote regions like the Arctic. This thesis confronts low-resource ecoacoustic audio classification by developing and evaluating complementary machine learning methodologies. We introduce EDANSA, the first publicly available, expert- labeled Arctic dataset of its kind, curated via novel active learning, alongside a baseline CNN. We systematically evaluate transfer learning, showing general audio embeddings effectively bootstrap classifiers for challenging Arctic sounds, significantly outperforming direct label mapping. Optimizing label utility, we investigate standard data augmentation and introduce novel audio data valuation via Shapley values, revealing …

Storage Location Optimization In Automated Storage And Retrieval Systems: A Deep Reinforcement Learning Approach, Lingjun Wang, Aldy Gunawan, Pieter Vansteenwegen

Research Collection School Of Computing and Information Systems

This study investigates the optimization of storage location in automated storage and retrieval systems (AS/RS). We introduce an optimization approach based on the Deep Q-Network (DQN) algorithm to enhance warehouse task efficiency and minimize stacker travel during storage and retrieval. To accelerate the algorithm training process, we integrate a prioritized experience replay mechanism. Furthermore, we decouple action selection from value estimation within the DQN framework to address the issue of value overestimation. The proposed model is evaluated against three heuristic methods. The experimental results demonstrate that our approach significantly outperforms these baselines.

Optimizing Hip And Knee Assistance For Walking And Sit-To-Stand Transitions: An Intrinsic Muscle Mechanics Based Predictive Approach, Neethan Ratnakumar

Dissertations

As the global population ages, the demand for wearable assistive technologies continues to rise, driven by their potential to enhance mobility and independence in older adults. Effectively designed controllers for lower-limb exoskeletons to assist sit-to-stand (STS) and walking are crucial for delivering efficient, safe, and comfortable assistance during daily activities. Traditionally, controller optimization involves biomechanical modeling and user-specific customization. Musculoskeletal simulations play a central role in this process by providing insights into human-exoskeleton interaction dynamics, thereby informing and refining control strategies.

This work presents a simulation-driven approach for developing exoskeleton controllers for walking and STS using two distinct methods: optimal …

Multimodal Learning In Real-World Application: Enhancing Feature Representation And Training Strategies, Nana Lin

Graduate Doctoral Dissertations

Multimodal learning has emerged as a critical paradigm for developing intelligent systems that can understand and reason across diverse inputs such as images, text, and audio data. Despite significant advances, effective deployment of multimodal models in practice remains a challenging task. This dissertation explores how multimodal learning can be effectively applied to high-stakes, real-world scenarios, with a focus on enhancing feature representation and training efficiency. Specifically, this research investigates multimodal learning strategies in two key domains: healthcare and surveillance.

In the healthcare domain, we explored the data fusion and alignment approaches for cognitive decline diagnoses. First, we propose the LOVEMA …

The Utilisation Of The Fourth Industrial Revolution (4ir) Technologies In E-Government Service Delivery: A Systematic Literature Review, Arnet Zitha, Noluntu Mpekoa, Sheethal Tom

African Conference on Information Systems and Technology

The integration of Fourth Industrial Revolution (4IR) technologies is transforming e-Government by boosting citizen engagement and enhancing service efficiency. However, gaps still exist in understanding the various applications, impacts, and barriers to adoption. This systematic review synthesises literature from 16 studies published between 2017 and 2025, illustrating how technologies like blockchain, artificial intelligence, big data, Internet of Things, and machine learning are employed and their effects on e-Government service delivery. The review reveals that 4IR technologies continue to play a vital role in e-Government services by addressing security threats, simplifying verification and authentication, building trust, and improving the quality and …

Human-Machine Communication: Complete Volume. Volume 10

Human-Machine Communication

This is the complete volume of HMC Volume 10.

The Emergence Of Ai Chatbots In Education, Trek Martin

Journal of Graduate Education Research

The recent advent of popular AI applications in educational contexts has sparked renewed interest in the question of AI and guided learning platforms as teaching tools. What are the possibilities for learning? In the attempt to answer that question, limitations of the field must be brought to full attention, as well an understanding of whether or not those limitations will continue into the immediate future; this research examines the technical evolution of artificial intelligence in education, from early symbolic reasoning systems to modern machine-learning-based chatbots. It then examines that evolution in terms of the key challenges it faced throughout, and …

Smart Mobility Technologies In Urban Areas Of Emerging Economies: A Bibliometric Analysis, Peter Mugisha, Rose Luke, Joash Mageto, Hossana Twinomurinzi

African Conference on Information Systems and Technology

Despite the adoption of smart mobility solutions in emerging economies, challenges such as traffic congestion, pollution and inadequate infrastructure still persist. This study analyses 540 scholarly articles published between 2003 and 2024 to evaluate how smart mobility technologies – such as Intelligent Transportation Systems (ITS), Internet of Things (IoT) and Artificial Intelligence (AI) – have been implemented in these regions. Data was retrieved from Scopus and Web of Science and analysed using Biblioshiny for bibliometric mapping and Atlas.ti for thematic analysis. The review identifies research trends and gaps, showing how ITS has improved transport management in cities like Nairobi, and …

Free-Running Ring Oscillators For Crystal-Free Communication Systems: Design, Simulation Challenges, And Frequency Stability Improvements, Haziq Rohail

Dissertations and Theses

This thesis presents techniques for enhancing the frequency stability of ring oscillators (ROs) for crystal-free wireless communication systems. The first major contribution is a tutorial-style study of frequency stability metrics, providing clear definitions, conversions methods, and comparative analysis of commonly used figures of merit such as phase noise, Allan deviation, and jitter. The second contribution addresses the challenges of simulating phase noise in free-running ROs. Key techniques including Periodic Steady-State Noise (PNoise), Harmonic Balance Noise (HBNoise), and transient noise analysis are evaluated in terms of accuracy, convergence behavior, and simulation runtime. Based on these results, practical guidelines are offered for …

Explaining Time Series Classifiers Through Post-Hoc Xai Methods Capturing Temporal Dependencies, Ephrem Tibebe Mekonnen

Conference papers

Time series classification is essential in domains such as healthcare and finance, where accurate predictions can have significant real-world consequences. However, in many high-stakes applications, understanding why a model makes a certain decision is just as important as the prediction itself. While deep learning models excel at capturing complex temporal patterns, their black-box nature limits transparency, making it difficult to trust and interpret their decisions. Although eXplainable AI (XAI) methods have advanced considerably for image and tabular data, applying them to time series remains challenging due to the intricate temporal dependencies and high dimensionality of the data. Post-hoc model-agnostic XAI …

Intelligent Motion Tracking: A Low-Cost Surveillance Framework Using Soc Devices And Sensor Fusion, Aung Myat Khaung, Nicholas Michael Stiffler

Research from the Berry Summer Thesis Institute, 2025

This thesis presents the design and implementation of a lightweight surveillance system capable of realtime motion detection, object tracking, and behavioral history reconstruction in controlled environments. The system uses System-on-Chip devices such as Raspberry Pi boards equipped with NOIR cameras, monocular cameras, and break-beam sensors that work together to detect and track single or multiple moving objects like colored balls. The prototype is validated in structured settings with the goal of eventual deployment in more dynamic environments, addressing the challenge of reliably tracking visually similar objects with minimal distinguishing features. The architecture integrates computer vision with sensor fusion by combining …

Learning Neural Point Processes For Long Event Sequences, Zhuoqun Li

LSU Doctoral Dissertations

This research presents a comprehensive series of studies aimed at advancing learning neural point processes for long event sequences, with applications spanning disaster resilience, crime forecasting, and healthcare. Structured around three interconnected studies, this work addresses core challenges in temporal point process (TPP) modeling, efficient handling of long event sequences, and improving accuracy over extended forecasting horizons by reinforcement learning.

The first study proposes the Sparse Transformer Hawkes Process (STHP) to model long asynchronous event sequences. Traditional neural network-based TPPs struggle with long event sequences due to computational inefficiencies. To address this, the STHP model combines two components: a temporal …

Exploring Arabic Large Language Models: A Comprehensive Review, Lamar Aljahdali, Joud Kaki

Effat Undergraduate Research Journal

This paper presents a comprehensive review of Arabic large language models (LLMs), exploring their capabilities, limitations, and potential impact on the Arabic NLP landscape. We analyze the performance of prominent LLMs, including JAIS, AraBERT, and BLOOM, highlighting their strengths and weaknesses on various NLP tasks. The review delves into critical challenges faced by Arabic LLMs, such as domain adaptation, cross-lingual capabilities, and ethical considerations. Additionally, the paper emphasizes the importance of responsible development and deployment practices for LLMs, ensuring fairness, transparency, and cultural sensitivity.

Intelligent System Designs For Hvac Energy Reduction In Buildings: Ai-Based Forecasting And Hybrid Active/Passive Approaches, Leena N. Alam, Rim M. Obaid, Thoraya Musa, Wegdan O. Alshateri, Passent M. Elkafrawy Prof

Effat Undergraduate Research Journal

The majority of building energy utilization worldwide is related to HVAC (Heating, Ventilation, and Air-Conditioning) systems. Eighty percent of the energy produced in Saudi Arabia is used by buildings, and since 70\% of that energy is used for ventilation, air conditioning accounts for roughly 50\% of the nation’s electrical use. This study reviewed and compared much research that used various AI-based forecasting algorithms. Specifically, the study explored the potential of passive and active cooling methods and intelligent system designs and used this analysis to develop a hybrid model that combined AI-based forecasting with active/passive approaches for optimal energy savings. The …

An Envisioned And Efficient Design Of Next Generation Decentralized Iot Bot Detection Model, Ahmed Abdullah Almalki

Doctoral Dissertations

The Industrial Internet of Things (IIoT) and Internet of Medical Things (IoMT) are revolutionizing critical infrastructures, but their expansion has also introduced severe cybersecurity vulnerabilities. Traditional IoT Bot Detection Systems (IBDS) struggle to scale in environments characterized by high-dimensional, large-scale, and redundant network traffic. These challenges hinder the development of reliable cloud-based intrusion detection systems. The limitations of static and rulebased methods in detecting evolving IoT botnet attacks—such as those launched by Mirai and Gafgyt—underscore the need for intelligent, adaptive approaches. To address this, the present study proposes a machine learning and deep learning-driven IoT Botnet Detection Model, validated through …

The Innovative Technique For Obtaining The Solutions Of Jeffery Hamel Nano-Fluids Flow Problem, Haedir Abd Alrazak Namoos, Abeer Majeed Jasim

Iraqi Journal for Computer Science and Mathematics

The study of heat transfer in nanofluid flows is increasingly important in many engineering, medical, and industrial applications. These fluids offer enhanced thermal cooling properties compared to conventional fluids. The research problem lies in the challenges of solving the Jeffrey-Hammel flow model for nanofluids, which includes coupled nonlinear differential equations that describe the thermal and hydrodynamic behavior of this type of flow, taking into account the influence of multiple factors such as the type, size, and concentration of nanoparticles. This research aims to propose a new hybrid analytical technique that combines the Laplace transform and the q-homotopy analysis technique with …

Optimized Hybrid Watermarking: Dual-Scheme Strategies For Enhanced Robustness, Ooi Jessie, Liew Siau Chuin, Syifak Izhar Bt Hisham, Khor Hui Liang, Khoo Bee Ee, Jasni Mohamad Zain

Iraqi Journal for Computer Science and Mathematics

Digital watermarking is crucial in content identification and copyright protection, particularly multimedia and medical imaging. This paper introduces two novel hybrid watermarking methods, Entropy-Guided Singular Embedding (EGSE) and Entropy-Guided Hybrid Embedding (EGHE), that improve upon existing techniques by integrating entropy-based adaptive block selection with Particle Swarm Optimization (PSO) for dynamic embedding strength determination. Unlike traditional methods, which rely on fixed embedding regions or manual parameter tuning, the proposed approaches automatically identify high-entropy regions to embed watermark signals, ensuring stronger resistance to distortion while maintaining image quality. EGSE employs Integer Wavelet Transform (IWT) and Singular Value Decomposition (SVD), whereas EGHE enhances …

Retracted: Efficient Multi-User Computation Offloading And Reducing Latency In Mobile-Edge Computing For Iot Applications, Sarmad T. Abdul-Samad, Osamah Al-Hwaidi, Ali Abd Al-Rasool Muslem

Iraqi Journal for Computer Science and Mathematics

Mobile Edge Computing (MEC) is an inventive paradigm for computing that has the potential to notably diminish latency and energy consumption by transferring computationally demanding jobs to edge clouds near intelligent mobile users. This investigation aims to reduce offloading and latency between multiple users and edge computing in the context of Internet of Things (IoT) applications in the fifth generation (5G) by utilizing an optimization algorithm called the Bald Eagle Search Optimization Algorithm. Although employing deep learning methods might increase time consumption and computational complexity, an edge computing system enables devices to transfer their demanding jobs to edge servers, decreasing …

Retracted: Iot Flow Parameters Classification Based On Machine Learning Techniques, El-Sayed M. El-Kenawy, Marwa M. Eid, Ban Salman Shukur, Amel Ali Alhussan, Doaa Sami Khafaga

Iraqi Journal for Computer Science and Mathematics

In recent years, there has been a highly remarkable convergence of artificial intelligence (AI) and the Internet of Things (IoT), which has made rapid progress in smart city initiatives by developing smart devices for such cities. Since these devices are increasingly diversified, they require a resilient communication network to demonstrate high performance in managing consistent traffic flows. A machine learning model intended for identifying network parameters from diverse devices, in addition to proposing modifications meant for network performance enhancement, is developed in this study. In relation to packet data as a network traffic parameter, employing gateway devices can facilitate its …

Retracted: Metaguard: A Federated Learning Approach To Hybrid Xgboost And Meta-Learning Models For Proactive Cyber Threat Hunting, Shatha H. Jafer Al-Khalisy, Ghada Al-Kateb

Iraqi Journal for Computer Science and Mathematics

In an increasingly interconnected world, cybersecurity threats have become more sophisticated, necessitating advanced, scalable, and privacy-preserving solutions. MetaGuard emerges as a novel framework that integrates federated learning with hybrid machine learning models, specifically XGBoost and meta-learning, to enhance proactive cyber threat detection. This framework offers a robust, distributed approach to cybersecurity, ensuring high detection accuracy while preserving user privacy through the implementation of differential privacy and homomorphic encryption. MetaGuard leverages distributed nodes to collaboratively train a global model, enabling rapid adaptation to new threats without the need for centralized data aggregation. Experimental evaluations using the CYBER-2024 dataset demonstrate that MetaGuard …

Detection Of Small Apple Targets Based On Improved Yolov5 In Natural Environments, Zilong Liu, Lei Zhang

Journal of System Simulation

Abstract: The distribution of apples usually features occlusion and small and dense targets. To address these issues, a target detection algorithm was proposed based on an improved YOLOv5 model. Specifically, this paper added the coordinate attention (CA) mechanism, receptive field block (RFB), and adaptively spatial feature fusion (ASFF) modules to the YOLOv5, enhancing the ability to detect small targets. Additionally, the proposed algorithm replaced the CIoU in YOLOv5 with SIoU to improve the target detection box's prediction accuracy. Finally, some normal convolutions were replaced with depthwise separable convolutions (DSC), effectively reducing the calculation burden. Experiment results show that the comprehensive …

Scene Generation Method For Maritime Target Recognition Based On Detection Parameters, Yuxuan Run, Dezhen Yang, Yeyang Liu, Wei Deng, Xiangyu Xing, Yi Ren

Journal of System Simulation

Abstract: Traditional scene generation methods for maritime target recognition consider only the effects of different environments on the generated scene data, while overlooking the changes in scene information caused by sensor detection parameters, resulting in a lack of accuracy and authenticity in generated scenes. To address this issue, a detection parameter-based scene generation method for maritime target recognition was proposed. For the task of maritime target recognition, key detection parameters affecting scene generation quality and essential scene features were analyzed. An association relationship modeling method based on Bayesian networks was proposed to construct a mapping relationship model between scene features …

Ship Fire Prediction Method Based On Evidence Theory With Fuzzy Reward, Chunyu Yang, Chuang Zhang, Xiaofan Zhang

Journal of System Simulation

Abstract: A multi-source information fusion approach based on the dempster-shafer (D-S) evidence theory with a fuzzy reward-penalty mechanism was proposed to address the issues of underreporting and false reporting in the early prediction of ship fires. PyroSim was utilized to construct a ship's laboratory model for fire simulation. Variations in carbon monoxide, temperature, and smoke concentration were recorded for data acquisition, followed by the application of a sigmf function for membership assignment. By leveraging the classical D-S theory, a reward-penalty mechanism was applied in weighted evidence fusion. Reward-penalty factors were utilized to differentiate various basic probability assignments, with unified belief …

Research On Joint Simulation Of Special Vehicle Engine Operation Characteristics Based On Virtual Driving Scenarios, Xueyuan Xie, Chen Lin, Han Wu, Qinglan Zhao, Junfei Gao, Qiangguo Hao, Xinqian Zheng

Journal of System Simulation

Abstract: The preliminary design of the overall operation performance of diesel engines cannot be guided by actual vehicle driving tests, which hinders the improvement of the power development level and efficiency of special vehicles. By using the virtual visual simulation engine Unity3D, two virtual driving scenario models were established: a flat road scenario and an undulating road scenario. Based on the speed characteristic parameters of the engine, a diesel engine's operation performance output model was constructed. Combined with the transmission system model and the longitudinal dynamics model of the vehicle's center of mass, a straight vehicle driving dynamics model was …

Simulation And Optimization Of Support Processes For Aircraft Fleet Launch Under Limited Resources, Feng Gong, Tao Jiang, Qin Zhang, Yu Liu

Journal of System Simulation

Abstract: To address the scheduling problem of aircraft fleet support processes under limited resources, a fleet support process optimization model that covered multiple aircraft, activities, and resource constraints was developed. An activity node graph model was used to establish the temporal logic, resource competition, and other constraints in the fleet support process, forming a "time – activity – resource" multidimensional optimization model. A genetic algorithm based on priority encoding was proposed, incorporating a serial decoding strategy and a dynamic penalty function to handle the complex constraints in the model, efficiently solving the optimization problem under complicated temporal and resource constraints. …

Short-Term Load Forecasting Based On Dual-Attention Temporal Convolutional Long Short-Term Memory Network, Lifen Li, Jinyue Zhang, Wangbin Cao, Huawei Mei

Journal of System Simulation

Abstract: In order to improve the accuracy of load forecasting and fully extract the hidden relationships between load and other characteristic factors, a load forecasting method based on dual-attention temporal convolutional LSTM network (DA-TCLSNet) was proposed. Correlation analysis was conducted on the dataset using the maximum information coefficient method to perform feature screening to reduce the computational cost of the model. The model input was constructed using a sliding window. The DATCLSNet forecasting model was constructed. The temporal convolutional layer extracted dependencies at different time scales and captured the nonlinear characteristics among variables such as load and weather. The multi-head …

Research On 3d Visualization Of Safety Monitoring And Early Warning For Steel Continuous Casting Scenarios, Wei Zhang, Wei Sheng, Yidan Cao, Tingsheng Zhao

Journal of System Simulation

Abstract: In order to improve the visualization and integration of production safety monitoring and fault warning, a three-dimensional (3D) visualization model architecture for whole-process industrial production safety monitoring and early warning for steel continuous casting scenarios was designed. By using 3ds Max and Unity3D, a multi-dimensional and multi-scale model was built, and functional modules such as visualization display and multi-level early warning for safety monitoring data were developed. By combining WebGL technology and Node. js runtime environment, the visualization of whole-process industrial production safety monitoring based on Web terminal was realized. The alarm threshold determination method for whole-process industrial production …

Optimal Scheduling Of An Integrated Energy System Considering Demand Response And Two-Stage P2g, Xinhui Duan, Zelong Cheng, Dongchao Zhang, Xiaochong Duan

Journal of System Simulation

Abstract: In the context of carbon peaking and carbon neutrality goals, this study aims to improve the energy utilization rate and further explore the role of user-side flexible loads and P2G equipment in energy saving and emission reduction. An optimal scheduling model for integrated energy systems considering demand response and two-stage P2G was proposed. A regional integrated energy system coupled with electricity, heating, cooling, gas, storage, and hydrogen was taken as the research object. Models for system equipment and two-stage P2G were established. Based on load characteristics, a multi-load demand response model for electricity, heating, and cooling was constructed using …

Storage Life Assessment Methods For Long-Term Storage Products Based On Multi-Scale Simulation: A Review, Hongmin Li, Xiao Han, Shuo Huang, Shengpeng Zhang, Shuanglong Rong, Hao Li, Cheng Qian

Journal of System Simulation

Abstract: Traditional experiment-based life assessment methods for long-term storage products suffer from drawbacks such as prolonged duration, high cost, and low prediction accuracy, greatly limiting the effectiveness of storage life assessment in practical applications. With the advancement of digital simulation technology, simulation analysis methods based on the physics of failure (PoF) have emerged as a research hotspot in the field of storage life assessment, as they can accurately characterize product aging behavior. The multi-scale characteristics of long-term storage products and their typical storage failure modes and mechanisms were analyzed. Multi-scale modeling and simulation analysis methods for storage failures of fundamental …

Research On Digital Simulation Method For Cognitive Load Evaluation Of Pilots, Zeng Fan, Mingjun He, Xiangyu Xing

Journal of System Simulation

Abstract: The operator's cognitive load constitutes a critical determinant of task performance. Pilots, as the primary operators of aircraft, must face an overwhelming volume of information during complex missions, which significantly heightens the risk of cognitive overload and operational errors. Evaluating cognitive load during tasks helps reduce human errors and improve system safety by optimizing design schemes. A simulation model was established to dynamically predict the pilots' cognitive load during tasks for multi-task scenarios. Based on the multiple resource theory, a method for quantifying cognitive load in multi-task conditions was established. By considering cognitive capacity, task time constraints, task priority, …