Intelligent Torque Control Of Switched Reluctance Motors Based On Ditc And Wavelet Neural Network, Ameer L. Saleh, László Számel

Mansoura Engineering Journal

Switched Reluctance Motors (SRMs) have been considered a high-performance and environmentally friendly solution for Electric Vehicle (EV) applications owing to their simpler construction, robust design, and high efficiency. However, it suffers from high torque ripple and acoustic noise due to its highly nonlinear magnetic characteristics and double-salient structure.  This paper introduces a nonlinear torque control based on the Direct Instantaneous Torque Control (DITC) scheme and a Wavelet Neural Network (WNN)  to achieve better dynamic response and mitigate torque ripple.  The proposed WNN is employed as a nonlinear torque controller, inserted between the torque error and the hysteresis torque controller within …

State-Dependent Queueing For Adaptive Signal Control: A Simulation-Based Performance Evaluation, Shaimaa Alseddiek, Usama Elrawy Shahdah, Hala B. Nafea, Hossam El-Din Moustafa, El-Said Ahmed Marzouk, Mohamed M. Ashour

Mansoura Engineering Journal

Urban traffic congestion persists as a critical challenge to transportation system efficiency, sustainability, and safety. Traditional queuing models utilizing fixed service rates inadequately represent the dynamic feedback between congestion and capacity in real vehicular flow. State-Dependent Queuing Models (SDQMs) address this limitation by modelling service rate as a function of queue length or density. This research advances SDQM application for adaptive traffic signal control through development of a calibrated state-dependent departure rate implemented within a microscopic simulation environment using SUMO and TraCI. Six control strategies including fixed-time, actuated, and two SDQM variants were evaluated across traffic demands ranging from undersaturated …

Psychological Impact Of Cultural Heritage Vr Experience Using Well-Mind Framework. (Case Study: Mahmoud Sami Palace In Mansoura)., Sara Ahmed El-Bayoumi, Hager Ahmed El-Sayed El-Ayouti, Mohamed Mohamed Shawki Abo Leila

Mansoura Engineering Journal

This study examines the psychological effects of immersive virtual reality (VR) experiences designed to preserve and disseminate intangible cultural heritage. It addresses the growing challenge of declining cultural engagement and the need for innovative tools that support psychological well-being, emotional engagement, and cultural connection. A VR-based reconstruction was developed within the historic Mahmoud Sami Al-Baroudi Palace in Mansoura, Egypt, focusing on traditional Egyptian practices and oral storytelling. The evaluation employed the WELL-MIND framework alongside participant feedback to assess both structured well-being indicators and subjective user experiences. The WELL-MIND-based assessment indicated improvements in cognitive stimulation, emotional engagement, and environmental perception. In …

Control Of Steel Corrosion Using Natural Corrosion Inhibitor, Rasha A. Nasef, O. Hamed, M. M. El-Halwany, M. H. Mahmoud

Mansoura Engineering Journal

This study investigates corrosion control of AISI 430 stainless steel in 1M HCl medium using an eco-friendly inhibitor extracted from pomegranate peel (PPE). The inhibition efficiency was evaluated at different PPE concentrations, with the highest applied concentration of 13g/L. Corrosion behavior was evaluated through weight-loss measurements and potentiodynamic polarization techniques conducted at room temperature.

The experimental results showed that corrosion resistance improved as the inhibitor  concentration increased. Lower corrosion rates were observed in the presence of PPE, along with reduced anodic and cathodic activities, indicating that the extract functions as a mixed-type inhibitor. Surface analysis confirmed the formation of …

The Multiwalled Carbon Nanotube-Incorporated Cfrp Laminates Interlaminar Fracture Toughness, Nikhil Janardan Rathod, Markala Karthik, Harshvardhan P. Ghongade, Javed Sikandar Shaikh, Rahul Soma Deshmukh

Mansoura Engineering Journal

The structural integrity of carbon fibre reinforced polymers (CFRPs) suffers from damage caused by interlaminar cracking and delamination which diminishes their performance capabilities. This study provides an experimental analysis of the impact of carboxyl-functionalized multi-walled carbon nanotubes (MWCNTs) on the Mode I interlaminar fracture toughness of CFRP laminates. The researchers created composite specimens by using epoxy systems which included MWCNT loadings of 0.1 wt% 0.2 wt% and 0.3 wt% together with an unmodified control. The researchers conducted Double Cantilever Beam (DCB) testing according to ASTM standards to ascertain the rate of critical strain energy release (G_IC) of the material. The …

Effectiveness Of Talcum Powder In Reducing Swelling Potential And Enhancing Strength Of Expansive Soil, Hesham E. Eleraky, Ashraf K. Nazir, Wasiem R. Azzam, Mohamed A. Sakr

Mansoura Engineering Journal

Expansive soils are clayey soils that undergo large volume changes as a result of moisture content variations. These soils have the potential to make foundations and other structures unstable and require costly repairs. This study examined the impact of adding talcum powder as a new stabilizing component to the matrix of an expansive soil in order to enhance the swelling pressure, Atterberg limits, modified Proctor, unconfined compressive strength at different periods of curing, X-ray diffraction and scanning electron microscopy were all used in a series of laboratory tests to examine how adding talcum powder to expansive soil affected the soil's …

Performance Of Foamed Concrete Before And After Fire Conditions: A Review, Poles Bahgat, Mohamed Kohail, A. Fathy, A. Afifi

Mansoura Engineering Journal

Foamed concrete is defined as a type of concrete distinguished by its workability, low density, and insulating properties. It is used as building material. The objective of this review is to discuss the physical and mechanical characteristics and fire sensitivity of foamed concrete. Density influences the strength of concrete. The strength of concrete is between 2 and 3.5 MPa, which increased with the addition of gold mining waste, reaching approximately 35 MPa at optimal replacement levels.  There is a 60-70% reduction in strength for lightweight concrete with a density of 400-1200 kg/m³ compared to conventional concrete. The fire sensitivity of …

Effect Of Load Variations On Power Quality And Ev Battery Performance In Vehicle-To-Grid System, Jitender Kaushal, Rahul Paswan

Mansoura Engineering Journal

Vehicle-to-Grid (V2G) technology enables the exchange of electricity between electric vehicles (EVs) and the utility grid, helping improve the integration of renewable energy sources and the grid's flexibility. Load-dependent power quality disturbances can affect the performance and reliability of V2G systems, which in turn can negatively impact system and battery operation. This study investigates the effect of resistive, inductive and nonlinear loads on the electrical performance and battery behaviour of a bidirectional V2G system. A complete MATLAB/Simulink model was developed to investigate the grid voltage, current, frequency, THD and the battery voltage, current and state of charge. The model uses …

Circulating Biomarker Results From A Phase 2 Study Of Seralutinib In Pulmonary Arterial Hypertension, Robin Osterhout, Athénaïs Boucly, Raymond L. Benza, Richard N. Channick, Kelly M. Chin, Robert P. Frantz, Anna R. Hemnes, Luke S. Howard, Vallerie V. Mclaughlin, Olivier Sitbon, Jean-Luc Vachiéry, Rotham T. Zamanian, Richard Aranda, Matt Cravets, Zhaoqing Ding, Thao Duong-Verlé, David Mattola, Robert F. Roscigno, Ravikumar Sitapara, Lawrence S. Zisman, Jean-Marie Bruey, Hossein-Ardeschir Ghofrani

Department of Medicine Faculty Publications

[Introduction] To the Editor: Pulmonary arterial hypertension (PAH) is a progressive disease characterized by obstructive pulmonary arterial remodeling (1). Seralutinib, an investigational inhaled tyrosine kinase inhibitor, potently and selectively targets kinases relevant to PAH pathobiology including platelet-derived growth factor receptors (PDGFR) α and β, colony stimulating factor 1 receptor (CSF1R), and mast/stem cell growth factor receptor kit (c-KIT) (2). In preclinical models, seralutinib improved cardiopulmonary hemodynamics, reversed pulmonary vascular pathology and decreased right ventricular hypertrophy (3). In TORREY, a phase 2, multicenter, double-blind, randomized, placebo-controlled trial in PAH, seralutinib significantly reduced pulmonary vascular resistance (PVR) after 24 weeks with good …

Micro- And Nanoplastics: A Paradigm Shift In The Pathogenesis Of Inflammatory Bowel Disease, Michael Saadeh, Priyata Dutta, Gordon Hong, Edward Oldfield, David A. Johnson

Department of Medicine Faculty Publications

Micro- and nanoplastics (MNPs) are pervasive environmental contaminants with growing recognition as potential contributors to human disease. Widespread human exposure occurs primarily through ingestion of contaminated food and water, and MNPs have been detected in multiple human tissues, including the gastrointestinal tract. Experimental evidence provides a plausible biological basis for disease associations, including impairment of intestinal barrier integrity, activation of mucosal immune pathways, and alteration of gut microbial communities caused by MNP exposure. Although human data remain limited, early studies demonstrate MNP detection in stool and suggest potential correlations with inflammatory biomarkers such as fecal calprotectin. These findings, together with …

A Comprehensive Survey Of Prompt Engineering Techniques In Large Language Models, Tonmoy Debnath, Md Nurul Absar Siddiky, Muhammad Enayetur Rahman, Prosenjit Das, Antu Kumar Guha, Muhammad Rezaur Rahman, H. M. Dipu Kabir

Electrical & Computer Engineering Faculty Publications

Prompt engineering has arisen as a pivotal discipline in optimizing the performance of Large Language Models (LLMs) by structuring inputs to enhance coherence, accuracy, and task alignment. This paper comprehensively surveys various prompting techniques, systematically categorizing them according to their application domains and methodological foundations. Fundamental approaches like zero-shot and few-shot prompting are examined along with advanced strategies, including chain-of-thought reasoning, retrieval-augmented generation, and self-consistency mechanisms. A rigorous qualitative analysis is conducted to evaluate each technique's strengths, limitations, and optimal use cases, offering a structured framework for selecting the most effective prompting strategies. Theoretical insights and empirical findings are consolidated …

Physics-Informed Temperature Prediction Of Lithium-Ion Batteries Using Decomposition-Enhanced Lstm And Bilstm Models, Seyed Saeed Madani, Yasmin Shabeer, Michael Fowler, Satyam Panchal, Carlos Ziebert, Hicham Chaoui, François Allard

Electrical & Computer Engineering Faculty Publications

Accurately forecasting the operating temperature of lithium-ion batteries (LIBs) is essential for preventing thermal runaway, extending service life, and ensuring the safe operation of electric vehicles and stationary energy-storage systems. This work introduces a unified, physics-informed, and data-driven temperature-prediction framework that integrates mathematically governed preprocessing, electrothermal decomposition, and sequential deep learning architectures. The methodology systematically applies the governing relations to convert raw temperature measurements into trend, seasonal, and residual components, thereby isolating long-term thermal accumulation, reversible entropy-driven oscillations, and irreversible resistive heating. These physically interpretable signatures serve as structured inputs to machine learning and deep learning models trained on temporally …

Machine Learning-Based Lifetime Prediction Of Lithium Batteries: A Comparative Assessment For Electric Vehicle Applications, Abdelilah Hammou, Raffaele Petrone, Demba Diallo, Boubekeur Tala-Ighil, Philippe Makany Boussiengue, Hicham Chaoui, Hamid Gualous

Electrical & Computer Engineering Faculty Publications

This paper evaluates and compares four data-driven methods (Gaussian Process Regression (GPR), echo state network (ESN), gated recurrent unit (GRU), and long short-term memory (LSTM)) for lithium-ion capacity prognostics adapted to electric vehicle conditions. This comparison aims to find the most efficient prognosis method considering two constraints: the limitation of computational power and the unavailability of on-board capacity measurement that requires full charge and discharge conditions. The machine learning models are trained using capacity values estimated under vehicle conditions. The ageing data is collected from cycling tests of two battery chemistries, Lithium Fer Phosphate (LFP) and Nickel Manganese Cobalt (NMC), …

Recent Insight And Perspective Of Marine-Inspired Biopolymers For Wound Healing Applications - A Review, Muhammad Umar Aslam Khan, Saiqa Yousaf, Abdalla Abdal-Hay, Mohd Faizal Bin Abdullah, Sahar Madani, Muhammad Shahzad Zafar, Goran M. Stojanović, Lobat Tayebi

Electrical & Computer Engineering Faculty Publications

There is an increasing necessity for advanced, sustainable, and biocompatible materials for wound healing as therapeutic and diagnostic products. Marine environments, characterized by high biodiversity, offer an underutilized source of natural resources with enormous potential for creating novel materials for dressings. This review highlights the revolutionary nature of polymeric biomaterials of marine origin, with a focus on polysaccharides, like alginate, chitosan, and carrageenan; proteins, such as collagen and gelatin. These biopolymers are outstanding in their physicochemical properties, such as biodegradability, bioactivity, and modifiable mechanical strength, which enable their use in wound-healing systems. Besides, these biomaterials may be easily chemically and …

Pin-Plane Electrical Discharge Driven By A Mosfet Dc Current Source, Myles Perry, Sidmar Holoman, Daniel Wozniak, Shirshak Kumar Dhali

Electrical & Computer Engineering Faculty Publications

The generation of atmospheric pressure nonequilibrium plasma using electrical discharges is an active area of research due to its significance in a wide spectrum of applications including medicine, combustion, and manufacturing. In our attempt to create a helium plasma jet in a pin-plane discharge with a constant current source, we observed self-pulsating behavior. We present the results of the electrical, optical, and spectroscopic measurements carried out to characterize the discharge. The duration of the discharge is a few tens of nanoseconds, and the repetition rate is in the few tens of kHz. The effect of the gap distance and gas …

Healthcare Digital Twins: A Methodological Literature Review On Integrating Iot And Ai For Personalized Medicine And Predictive Care, Sara Shahnazinia, Mahsa Tavasoli, Abdolhossein Sarrafzadeh, Ali Karimoddini

Electrical & Computer Engineering Faculty Publications

Digital Twin (DT) technology has the potential to revolutionize healthcare delivery and enhance patient outcomes through personalized and precision medicine, simulation models for operations and interventions, and drug discovery. However, successful implementation of DTs in Internet of Things (IoT) and artificial intelligence (AI) healthcare is contingent upon addressing key challenges such as privacy, ethics, and robust data security. This paper presents a methodological literature review of DT applications in healthcare, systematically analyzing the current state of research, key enabling technologies, and implementation challenges. The review summarizes DT categorization approaches (application-based, technology-based, and real-time function-based); delineates core DT components such as …

An Explainable Cs-Mitigation Triangular (Ecsmt) Framework To Secure Graph Neural Networks, Sabah Ettahri, Sergio Pallas Enguita, Chung-Hao Chen, Wen-Chao Yang

Electrical & Computer Engineering Faculty Publications

This research addresses cyber risk by defending against backdoor attacks on Graph Neural Networks (GNNs). We propose the Explainable Complex System-Mitigation Triangular (ECSMT) Framework, which integrates Robust Training, Graph Regularization, and Data Sanitization into a lightweight, hardware-efficient defense layer. To evaluate structural generalizability, we conducted empirical evaluations across three distinct benchmark domains (AIDS, MUTAG, and PROTEINS) using a Graph Isomorphism Network (GIN) backbone. Under a baseline 5% backdoor subgraph trigger injection ratio, ECSMT achieves excellent utility retention, securing a Clean Accuracy (CA) of 97.33% (±0.62%) while reducing the Attack Success Rate (ASR) from 97.00% down to 69.45% on the primary …

Application Of A Hyaluronic Acid-Based Lotions Containing Hydroxyapatite Nano-Particles Approach For Treatment Of Initial Dental Caries, Zahra Gharavi, Zahra Namazi, Fatemehsadat Pishbin, Helia Givian, Maryam Torshabi, Farhood Najafi, Parisa Amdjadi

Electrical & Computer Engineering Faculty Publications

This study evaluated the remineralization potential of a ceramic–polyelectrolyte system, based on a novel combination of hydroxyapatite nanoparticles (HAp NPs) and a hyaluronic acid (HY) matrix, for dental enamel, a tissue that remains challenging to repair in clinical dentistry. Nano-Hydroxyapatite (nanoHAp) powder was synthesized and characterized using X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR), field-emission scanning electron microscopy (FE-SEM), dynamic light scattering (DLS), and zeta potential analysis to confirm their crystalline structure, functional groups, morphology, particle size distribution, and colloidal stability. HY-based suspensions (remineralizing lotions) containing 5, 10, and 12 wt% of the synthesized nanoHAp particles were formulated via a …

Skin Type Diversity In Image Datasets, Neda Alipour

Doctoral

Image-based AI systems that analyse human skin are increasingly used in healthcare and computer vision applications. However, many human skin-based image datasets do not provide reliable information about skin type, making it difficult to assess whether these systems perform consistently across the full spectrum of skin colour. The objective of this thesis is to examine how skin type diversity is represented and measured in image datasets, and to evaluate the reliability of image-based skin type measurement methods under different imaging conditions. Using publicly available skin lesion image datasets as a well-defined and widely used sub-class of skin image datasets, this …

Automation And Habitat Development For A Space Based Marine Life Environment, Logan Trimmer, Alexander Hang

Harrisburg University Other Works

This project was a cross-collaboration between the Environmental Sciences and Advanced Manufacturing and Robotics programs for the company Monolith Space.

The goal of this project was to design an autonomous system to be able to track qualities of water in an aquaculture system designed to be sent to space.

The Efficacy Of Hybrid Manufacturing For High Stress Automotive Components, Logan Trimmer

Harrisburg University Other Works

The goal of this research was to establish the viability of using hybrid manufacturing for automotive applications. By verifying that high-stress components can be created, it can be assumed that any other lower stress part could be made to match the strength requirements. A limiting factor of adoption for hybrid manufacturing is how new the technology is. Studies on time and cost were performed allowing for comparisons with traditional manufacturing technologies (casting, forging, milling) used in automotive applications. This research utilized a Haas Automation UMC750 5-axis CNC mill with a Meltio laser wire direct energy deposition attachment. Fusion 360 was …

Bridging Mission And Execution: Integrating Participatory Design In Early-Phase Mission Engineering For Stakeholder Alignment And Mission Clarity, Rafi Soule

Knowledge and Creativity Expo

This research examines mission framing during the early phase of Mission Engineering. Stakeholder interpretations diverge under ambiguity. Interoperability constraints are often not surfaced early. These conditions reduce mission clarity and weaken mission-to-system mapping readiness. The study integrates a participatory design-inspired, artifact-first workflow with RAG-enabled retrieval from a closed corpus to support evidence-grounded reasoning and traceable citations.

Phase 1 uses an online survey to establish baseline patterns in practice (N = 86). Shared understanding is positively associated with mission clarity (r = 0.60, p < 0.001). Phase 2 uses a time-bounded comparative workshop with two conditions. Expert reviewers rate mission statement quality higher for the participatory design condition (mean 3.5) than the traditional condition (mean 2.8). Technical feasibility ratings are similar across conditions. Phase 3 demonstrates RAG-enabled, closed-corpus, retrieval-supported traceability using the Referencer tool. It is reported as a proof-of-concept for evidence-grounded rationale and auditability, and as a pathway …

Strontium Titanate For Capacitor And Energy Storage Applications At Cryogenic Temperatures, Hung Trinh

Doctoral Dissertations

This study investigates the dielectric properties of single crystal and ceramic strontium titanate (SrTiO₃) for cryogenic capacitor applications from room temperature to 4 K. Permittivity (k) and loss tangent are dependent on temperature, frequency, mechanical stress, and applied DC electric field. Accordingly, the dielectric constant and loss tangent were measured at various frequencies and DC bias levels. Loss tangent data are also presented as equivalent series resistance (ESR). For single crystal SrTiO₃, an impurity level of ≈500 ppm barium resulted in an increase of the maximum permittivity to approximately 50,000 at 6 K. This relatively high …

Rapid Urbanization Reduces Genetic Diversity And Increases Genetic Differentiation Of A Lynx Spider Oxyopes Sertatus In Central Taiwan, Ying-Yuan Lo, Chi Wei, Wan-Jyun Chen, Chung-Ping Lin

Biological Sciences Faculty Publications

Urbanization is a dominant force driving destructive and irreversible changes of natural habitats in modern times. While the effects of urbanization on community composition and phenotypic responses are well-documented, its influence on genetic diversity and population structure remains understudied, particularly for invertebrates in subtropical regions. This study tested the hypothesis that urbanization reduces genetic diversity and increases population differentiation in the lynx spider Oxyopes sertatus, a common foliage-dwelling spider in Taiwan. We sampled 245 individuals from 17 sites distributed along an urban-rural gradient and quantified urbanization intensity using land-use composition at both landscape (4 km²) and local (0.25 km²) …

Multi-Objective Hydro-Thermal Optimization Of Spiral Conformal Cooling Channels Using A Kriging-Cfd Framework, Soroush Masoudi, Barun K. Das, Majid Tolouei-Rad

Research outputs 2022 to 2026

This study presents a surrogate-based computational framework for the thermo-hydraulic optimization of spiral conformal cooling channels (CCCs) used in injection moulding applications. Sixteen design configurations were generated using Latin Hypercube Sampling and evaluated through CFD simulations to determine cooling time and pressure drop. Based on the simulation data, a Universal Kriging surrogate model was developed to describe the nonlinear relationships between channel diameter, helix pitch, channel-to-surface distance, and the resulting thermo-fluid performance. The predictive capability of the surrogate model was evaluated using leave-one-out cross-validation. The model showed excellent agreement for pressure drop prediction (R² ≈ 0.99) and good predictive accuracy …

Optimizing Ev Battery Charging Using Fuzzy Logic In The Presence Of Uncertainties And Unknown Parameters, Minhaz Uddin Ahmed, Md Ohirul Qays, Stefan Lachowicz, Parvez Mahmud

Research outputs 2022 to 2026

The growing use of electric vehicles (EVs) creates challenges in designing charging systems that are smart, dependable, and efficient, especially when environmental conditions change. This research proposes a fuzzy-logic-based PID control strategy integrated into a photovoltaic (PV) powered EV charging system to address uncertainties such as fluctuating solar irradiance, grid instability, and dynamic load demands. A MATLAB-R2023a/Simulink-R2023a model was developed to simulate the charging process using real-time adaptive control. The fuzzy logic controller (FLC) automatically updates the PID gains by evaluating the error and how quickly the error is changing. This adaptive approach enables efficient voltage regulation and improved system …

Numerically Evaluating The Effect Of Extrusion Angle On Material Flow And Thermal Behaviour During Additive Friction Extrusion Deposition (Afed), Numan Habib, Ferdinando Guzzomi, Ana Vafadar

Research outputs 2022 to 2026

Additive Friction Extrusion Deposition (AFED), also known as “SoftTouch”, is an emerging friction-based Additive Manufacturing (AM) technology allowing material to soften before the deposition, increasing the printing speed and reducing cost [1]. However, high power is required during the process, as excessive force is needed to extrude enough material through the printing head. This study investigates how extrusion angle and tool rotational speed affect thermal distribution and material flow in AFED to minimise power consumption. A three-dimensional computational fluid dynamics (CFD) model is proposed, and ANSYS ^® Workbench CFD code (Fluent) is used to discretise the CFD model. A User-Defined …

Choice-Based Crowdshipping For Next-Day Delivery Services: A Dynamic Task Display Problem, Alp Arslan, Firat Kilci, Shih-Fen Cheng, Archan Misra

Research Collection School Of Computing and Information Systems

This paper studies integrating the crowd workforce into next-day home delivery services. In this setting, both crowd drivers and contract drivers collaborate in making deliveries. Crowd drivers have limited capacity and can choose not to deliver if the presented tasks do not align with their preferences. The central question addressed is: How can the platform minimize the total task fulfilment cost, which includes payouts to crowd drivers and additional payouts to contract drivers for delivering the unselected tasks by customizing task displays to crowd drivers? To tackle this problem, we formulate it as a finite-horizon Stochastic Decision Problem, capturing crowd …

Learning-Based Graph Shrinking For Quantum Optimization Of Constrained Combinatorial Problems, Monit Sharma, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Graph shrinking has recently emerged as a powerful preprocessing technique for hybrid classical–quantum optimization, enabling variable and constraint reduction before quantum solving. Conventional approaches rely on Semi-Definite Programming (SDP) relaxations to compute vertex correlations, but these methods suffer from high computational overhead, instance-specific tuning, and limited generalizability. In this work, we replace the handcrafted SDP correlation stage with a reinforcement learning (RL) based correlation estimator, trained to predict merge quality directly from graph structure. We reformulate the graph shrinking process as a Markov Decision Process (MDP), design a Graph Neural Network (GNN) policy to guide vertex merging, and integrate the …

Dystop: Dynamic Staleness Control And Topology Construction For Asynchronous Decentralized Federated Learning, Yizhou Shi, Qianpiao Ma, Yan Xu, Junlong Zhou, Ming Hu, Yunming Liao

Research Collection School Of Computing and Information Systems

Federated Learning (FL) has emerged as a potential distributed learning paradigm that enables model training on edge devices (i.e., workers) while preserving data privacy. However, its reliance on a centralized server leads to limited scalability. Decentralized federated learning (DFL) eliminates the dependency on a centralized server by enabling peer-to-peer model exchange. Existing DFL mechanisms mainly employ synchronous communication, which may result in training inefficiencies under heterogeneous and dynamic edge environments. Although a few recent asynchronous DFL (ADFL) mechanisms have been proposed to address these issues, they typically yield stale model aggregation and frequent model transmission, leading to degraded training performance …