Master of Science in Electrical Engineering, Cleveland State University, 2025, Washkewicz College of Engineering

With the ever-increasing demand for high-speed, reliable data transmission, the development of efficient optical communication systems has become a critical area of research. Optical communication offers several advantages over traditional radio frequency (RF) and acoustic methods, including higher bandwidth, lower latency, higher data rate, and greater immunity to electromagnetic interference. These benefits make optical systems particularly appealing for underwater communication, where achieving robust and highdata-rate transmission remains a significant challenge due to the unique limitations added by the underwater channel. This research aims to determine the optimal combination of parameters for efficiently transmitting video frame data through the underwater medium, using results from simulated channel models as a baseline. Utilizing an arbitrary waveform generator, oscilloscope, along with a 520 nm green laser as an optical source, the system is designed to test the effectiveness of different modulation schemes and error correction techniques. MATLAB is employed for video processing, encoding/decoding, modulation/demodulation and interfacing with the waveform generator and oscilloscope, ensuring precise control and analysis of the transmitted and received signals. On-off keying (OOK) Intensity Modulation as well as 4 and 8 order Quadrature Amplitude Modulation (QAM) are the modulation techniques under consideration, comparing spectral efficiency and resilience to channel impairments between them. Another focus of this research is evaluating the performance of Reed-Solomon and Low-Density Parity-Check (LDPC) error correction codes to identify the most effective strategy for mitigating transmission errors. By systematically testing these variables, this research seeks to provide insights into the practical implementation of underwater optical communication systems and identify configurations that maximize data throughput and reliability.

Master of Science in Engineering, Youngstown State University, 2025, Rayen School of Engineering

Electricity is an integral part of everyday life in both essential and non essential uses. As much as it can be our friend, it can also be our enemy if not used and controlled properly. An electrical design of any kind is done in the form of a circuit. Ensuring that a circuit is both designed and operating correctly all boils down to one thing: math. Kirchoff's laws state that all amperages and voltages in a loop must sum to zero. If either does not, that means the circuit is not working as intended. Similarly, Ohm's law states that the current and voltage of a circuit are proportional in terms of the impedance, or the circuit sum of resistance, inductance and capacitance. If the current increases above or the voltage drops below than expected, the circuit is not working as intended. The previous two cases are examples of an electrical fault, where the source needs to be removed immediately as serious damage and/or injuries could result if left connected. A safe way to open and close a circuit while also monitoring for any fault that occurs can be done with a circuit breaker. This paper discusses methods for testing & calibrating circuit breakers to ensure a circuit can be protected from faults.

Master of Science in Electrical Engineering, University of Dayton, 2025, Electrical and Computer Engineering

Lung cancer is the leading cause of cancer-related deaths. Which is why early and accurate detection of pulmonary nodules in computed tomography (CT) scans is necessary. Manual segmentation of nodules by radiologists is time consuming and prone to variability, fueling the need for automated solutions. This thesis introduces a novel Difference-over Union (DoU) Combo Loss function for 3D lung nodule segmentation using a 3D U-Net architecture, trained on the LIDC-IDRI dataset. The proposed loss function combines dice loss, binary cross-entropy, and boundary DoU loss to address region, pixel, and boundary level segmentation challenges, enhancing performance on imbalanced data. Quantitative results show that the DoU Combo Loss slightly outperformed the standard Combo Loss and other baselines, like the unified focal loss. Qualitative analysis showed reduced hallucination in complex CT scans with DoU Combo Loss, improving reliability. Although the quantitative gains are modest, the proposed method demonstrates potential for robust automated nodule segmentation, offering a simpler alternative to complex architectures and paving the way for enhanced clinical diagnostics.

Master of Science in Electrical Engineering, University of Dayton, 2025, Electrical Engineering

Wearable electronics depend heavily on robust and responsive sensors, yet conventional invasive methods and rigid structures have long limited their potential. While miniaturization has improved sensor integration, these systems often struggle under mechanical stress such as bending or stretching. Recent advances in soft, stretchable materials with high conductivity and self-healing capabilities offer a forward, promising path. Among these, Eutectic Gallium–Indium (EGaIn) stands out due to its high conductivity and minimal resistance change under strain. This thesis presents the design and development of a soft, stretchable sensing system that leverages EGaIn-based electronics for multimodal sensing. Capacitive sensing is used for touch localization, while resistive EGaIn traces enable strain measurement. Also, gesture recognition is implemented using a convolutional neural network (CNN), which classifies touch patterns captured from the sensor surface. The system demonstrates the potential of combining capacitive and resistive sensing to form a highly responsive, deformable interface. Through extensive testing—including uniaxial and biaxial stretch experiments—this work highlights the challenges of two-dimensional strain sensing and the importance of sensor layout in performance.

Master of Science, Miami University, 2025, Electrical and Computer Engineering

First-order iterative algorithms are widely employed to solve large-scale optimization problems, many of which often appear in science or engineering. A relevant example of such a problem is in the field of machine learning, where huge amounts of computational power are spent optimizing model errors. Consequently, the ability to systematically compare algorithm performance across any given application to find the most efficient method can save time and energy. However, existing methods for algorithm analysis often have shortcomings ranging from the rigorousness of the result to user accessibility. In this thesis, we introduce the AlgorithmAnalysis.jl Julia package, a Lyapunov function-based framework for robust algorithm analysis accessible to a broad user base.

Master of Science in Engineering, University of Akron, 2025, Electrical Engineering

With the development of microgrids as an important part of electrical distribution systems, and particularly with an increased recognition of the benefits of DC microgrids for some applications, it is desirable to minimize their cost of operation. An approach using loads capable of operating at variable power levels is proposed, to maximize the use of energy at times of relatively low cost. This method is evaluated in a simulated microgrid. Models of energy sources and load are constructed, with control systems that allow the contribution of each component to be adjusted. These items are assembled into the simulated microgrid. A cost function for the system is developed, and evaluated using a Genetic Algorithm (GA). This optimization method considers contributions from all sources and loads connected to the microgrid, and provides optimal operating parameters for each of them. The system is simulated in Matlab/Simulink, for a variety of scenarios representing various expected operating conditions. The calculated optimal operation for each scenario is compared to a benchmark to determine cost savings. Some savings were found to be achievable by the proposed method, with the amounts of financial benefits varying somewhat depending on the specific conditions of each scenario.

Master of Science in Engineering, Youngstown State University, 2025, Rayen School of Engineering

Internet of Things (IoT) devices, such as thermostats, lighting systems, and fitness trackers, have revolutionized both residential and industrial environments, enabling users to remotely control and manage them. Although many IoT devices are often manageable through Original Equipment Manufacturer (OEM) software applications, overseeing devices from various OEM origins simultaneously, or even customized hardware, can be complex and tedious. To address this challenge, an IoT gateway serves as a centralized hub that supports wireless connectivity across various protocols. Bluetooth Low Energy (BLE), a widely adopted wireless communication protocol in low-power-consuming devices such as sensors, is therefore employed in many IoT gateways. Large-scale IoT networks significantly benefit from an IoT gateway, as it provides a unified management point for all connected devices. OEMs of IoT gateways may offer a software development kit (SDK) to facilitate application customization, enabling the attainment of specific design requirements. This thesis presents the design, development, and implementation of two software applications to acquire real-time sensor data from custom BLE-enabled printed circuit boards (PCBs). Leveraging an IoT gateway, its compatible SDK, and a library of sample programs, two applications are developed to monitor sensor data: a serial terminal interface and a dynamic web-based dashboard.

Master of Science, Miami University, 2025, Electrical and Computer Engineering

Modern technological growth has led to an increasing demand for bandwidth usage. This necessitates greater bandwidth efficiency and effort to reduce signal clutter. Through this research, we suggest a method of combined radar and communication signal design as a potential solution to these needs. This paper presents an analysis of “chaotic” pseudo-random signals, Orthogonal Frequency Division Multiplexed (OFDM) signals and their performance in Synthetic Aperture Radar (SAR) scenarios. A Matlab based SAR simulation was used to evaluate each signal's performance against a benchmark LFM Chirp signal. Tests were performed at varying bandwidths and SNRs. A novel method of Random Signal Encoding (RSE) with randomized subchannel assignment was also utilized to improve secure communications capabilities for the OFDM signal. The results indicate that both the chaotic signal and OFDM signal perform sufficiently in the combined radar and communication scenario. Each is able to accurately image multiple targets of varying reflectivity and perform comparably to a standard LFM chirp. Each signal also has additional benefits, with the chaotic signal being immune to repeat jamming and the OFDM signal providing greater security for communications data through the novel method of encoding.

Master of Science, The Ohio State University, 2025, Electrical and Computer Engineering

Analog time encoding is a promising technique for processing analog signals in the time domain, especially in applications where traditional analog circuits are difficult to implement. This approach leverages phase-locked loops (PLLs) and sigma-delta modulation to encode analog information into time-based signal characteristics such as phase or frequency. A programmable and reconfigurable PLL can serve as a flexible hardware platform for testing and prototyping various analog time-encoded applications. This thesis presents the Register Transfer Level (RTL) design of an All-Digital PLL (ADPLL) using Verilog-HDL. Time-domain simulations were conducted to evaluate its lock time and jitter performance.

Master of Science, The Ohio State University, 2025, Electrical and Computer Engineering

Disability has become a global issue. About 1.3 billion global population were living with significant disability as of 2023, which accounted for 16% of the earth population. Paralyzed patients live with significant disabilities because they cannot control their arm(s) or finger(s), which is caused by permanent diseases like spinal cord injury (SCI). While passive treatment methods to SCI like occupational therapy suffer from very long-term commitment for patients and potentially poor recovery, active treatment methods like neuromuscular stimulation help to recover patients' sensation and controlling ability fundamentally, which is a better solution. Active engineering methods like neuromuscular stimulation are relatively mature, well-studied, certain and promising. The digital wireless data collection system developed in this work serves as the foundation for a novel neuromuscular stimulation system currently under development. The ultimate goal of this work is to develop a real-time force sampling and data collection system with a microcontroller for a capacitive application-specific integrated circuit (ASIC) force sensor (IAM implant) proposed and manufactured by several previously published works. Experiments in this work start from quantitatively fitting the force-capacitance linear regression relationship for the IAM implant and a previously manufactured standard Printed Circuit Board (PCB) LC transceiver using an oscilloscope (𝑐𝑎𝑝 [𝑝𝐹]=0.16594×𝑓𝑜𝑟𝑐𝑒 [𝑁] + 5.0897). The initial MATLAB program performs the basic functionality to extract bits and calculate capacitance values based on an analog data processing scheme. To improve the initial sampling method and data processing algorithm, a better system that involves a logic analyzer and a newly designed flexible PCB (fPCB) LC transceiver is implemented, which automates the force sampling process with continuous-time recording and abundant data samples. The improved MATLAB program features a transitional trigger/st (open full item for complete abstract)

Master of Science, The Ohio State University, 2025, Electrical and Computer Engineering

This thesis introduces the development of an implantable system designed for optogenetic applications. The work is based on the development of DNA editing technology, enabling the transition from electrical stimulation to photonic stimulation. This work primarily reduces the size of the implantable system while maintaining its function. The proposed system achieves a compact form factor with 6 mm × 6 mm compared to previous works typically exceeding 10mm. This work mainly uses PSIM for power circuit simulation and Altium Design for PCB level design. By multiple iterations of design in coils, power management unit and control unit, this project develops a miniaturized device that meets the implantation requirements. This work improves the implantability of photonic stimulation platforms, making optogenetic tools feasible in broader medical applications.

Doctor of Philosophy, The Ohio State University, 2025, Electrical and Computer Engineering

Direct-current (DC) power distribution has gained increasing attention in recent years due to its potential to enhance energy efficiency, integrate renewable energy sources seamlessly, and better accommodate modern electronic loads. Unlike traditional alternating-current (AC) systems, which dominate power distribution, DC systems offer several advantages, including reduced conversion losses, improved compatibility with energy storage technologies, and simplified power electronics interfaces. These benefits make DC microgrids particularly attractive for applications such as data centres, electric vehicle charging infrastructure, and renewable energy systems. However, despite these advantages, the widespread adoption of DC distribution remains hindered by critical technical challenges. Managing and mitigating fault currents in DC networks is inherently more complex due to the absence of natural current zero-crossing, simplifying fault interruption in AC systems. Additionally, precise power flow control and voltage regulation are essential to ensure stable operation, especially in multi-source, multi-load power grids. Maintaining power quality and preventing instability in interconnected DC grids further complicate the design and operation of the system. To overcome these challenges, this work proposes a flexible DC energy router (FeDER)—a modular, scalable power management unit built using wide-band-gap (WBG) semiconductors for interconnected DC microgrids. The FeDER integrates local energy storage and is designed to provide an all-in-one solution for DC microgrid power management requirements: fault management, stability enhancement, power flow regulation, and improved power quality. The dissertation thoroughly examines FeDER, addressing its key operating modes and functionalities within interconnected DC microgrids, providing an in-depth analysis of the FeDER topology and its underlying operating principles. In addition, the proposal explores strategies to im (open full item for complete abstract)

Master of Science, The Ohio State University, 2025, Electrical and Computer Engineering

In this thesis, two papers that I have published on the topics of power amplifier identification and power amplifier linearization are presented. The first paper discusses quasi-identical power amplifiers that can be differentiated through the unique non-linearities that are inherent to each individual power amplifier (PA). In this paper it is experimentally verified that first linearizing the PA outputs using the same digital post distortion (DPoD) technique to correct for the non-linearities of the reference PA, magnifies the differences observed in the EVM and NMSE metrics. As the signal-to-noise ratio (SNR) of the distorted signal output increases, the differences between the EVM/NMSE measurements of the two PAs also increases. At a high enough SNR, two quasi-identical PAs can be reliably differentiated using the EVM/NMSE metrics once DPoD has been applied. In the second paper which is a continuation of the first, a generalized cubic spline basis (GCSB) with selective deep memory is used to perform an enhanced DPoD. It is experimentally verified that the use of deep memory in the GCSB model can not only increase the performance of DPoD but also greatly magnify the differences observed in the normalized mean squared error (NMSE) of the linearized PA output signal relative to the reference input signal. This new technique can thus be used to reliably differentiate between two quasi-identical power amplifiers from the same device's manufacturer. Lastly, a signal visualization tool developed with the goal to aid in signal analysis and the understanding of power amplifier linearization is presented.

Doctor of Philosophy (Ph.D.), University of Dayton, 2025, Electrical Engineering

Lithium-ion batteries (LIBs) are a cornerstone of next-generation electric aircraft, offering high energy density, power density, and long cycle life. However, these attributes introduce safety challenges, particularly the risk of thermal runaway (TR), which poses significant concerns in high-energy, high-power applications. This study addresses these challenges by investigating LIB systems' structural, thermal, and mechanical under extreme conditions and proposing engineering solutions to enhance safety and performance by improving resiliency against internal electrical shorting. A key focus of this work is the detection and mitigation of internal short circuits (ISC), a primary trigger for TR. ISCs can result from mechanical abuse or the formation of lithium dendrites in which a metallic object internally connects anodes and cathodes, leading to a fast battery discharge and consequent heat generation. Advanced sensing layers that detect early dendrite growth are demonstrated, providing a critical safety feature. Techniques to distinguish dendrite-induced changes from routine operational behaviors are discussed, emphasizing the need for continued algorithm development for practical application. These innovations hold promise for integration into battery management systems (BMS), enhancing overall system safety while maintaining performance. To mitigate mechanical abuse risks, this study explores materials (battery enclosure) and configurations to improve LIB safety and functionality for a wide range of applications, including aviation. Titanium (Ti) is an optimal material for battery enclosures due to its superior thermal and mechanical properties and moderate gravimetric density. Among three widely used LIB formats, viz. cylindrical, prismatic, and pouch, pouch cell formats offer advantages in weight and compactness; they require enhanced protection to withstand mechanical stress and prevent TR. Modular designs and optimized secondary enclosures are proposed to (open full item for complete abstract)

Master of Science, The Ohio State University, 2025, Electrical and Computer Engineering

Accurate segmentation of brain tumors, particularly adult gliomas, is critical for effective diagnosis, treatment planning, and prognostication. Gliomas exhibit signifi- cant spatial and scale heterogeneity along with an infiltrative growth pattern, posing major challenges to conventional segmentation techniques. Conventional methods in brain tumor segmentation often fall short in capturing both fine local details and the broader contextual information needed for precise delineation of tumor bound- aries. Additionally, many current deep learning approaches lack robust uncertainty estimation, which is essential for reliably assessing prediction confidence in clinical decision-making. There remains a critical need for resource-efficient architectures that not only achieve high segmentation accuracy but also reduce computational costs, especially when addressing the complex, heterogeneous, and infiltrative characteristics of gliomas. In this thesis, we propose novel deep learning frameworks that have efficient receptive field expansion and uncertainty-guided segmentation to address these challenges. Our approach combines multi-scale attention mechanisms with Atrous Spatial Pyramid Pooling (ASPP) to capture both fine-grained local details and broader contextual information, thereby enhancing segmentation accuracy. Additionally, by incorporating Monte Carlo dropout and designing an uncertainty-aware loss function, our model quantifies per-pixel prediction confidence, ultimately improving the reliability and interpretability of the segmentation outcomes. Extensive experiments on the BraTS2023 and BraTS2024 datasets demonstrate that the proposed method outperforms traditional U-Net variants in terms of both segmentation performance and computational efficiency. The integration of uncertainty estimation not only offers valuable insights for clinical decision-making but also paves the way for more personalized treatment strategies in managing adult glioma. Followin (open full item for complete abstract)

Doctor of Philosophy, The Ohio State University, 2025, Electrical and Computer Engineering

To solve linear inverse problems, plug-and-play (PnP) methods replace the proximal step in a convex optimization algorithm with a call to an application-specific denoiser, often implemented using a deep neural network (DNN). Although such methods yield accurate solutions, they can be improved. For example, denoisers are usually designed/trained to remove white Gaussian noise, but the denoiser input error in PnP algorithms is usually far from white or Gaussian. Approximate message passing (AMP) methods provide white and Gaussian denoiser input error, but only when the forward operator is sufficiently random. In the first work of the dissertation, for Fourier operator based linear inverse problem, we propose a PnP algorithm based on generalized expectation-consistent (GEC) approximation---a close cousin of AMP---that offers predictable error statistics at each iteration, as well as a new DNN denoiser that leverages those statistics. We apply our approach to magnetic resonance (MR) image recovery and demonstrate its advantages over existing PnP and AMP methods. In the second work of the dissertation, we expand our focus to address both linear and non-linear inverse problems within the generalized linear model (GLM) framework. We propose a novel variant of the expectation-consistent (EC) approximation that iteratively leverages DNNs to solve GLM inverse problems. Unlike traditional EC implementations, our proposed framework does not require random forward operators. As a case study, we focus on a popular non-linear inverse problem of phase retrieval, which involves accurately recovering images from noisy phaseless measurements. In addition to applying EC in a non-traditional manner, we also propose a novel stochastic damping scheme that is inspired by recent diffusion methods. Like existing phase-retrieval methods based on PnP priors, regularization by denoising, or diffusion, our approach iterates a denoising stage with a measurement-exploitation stage. B (open full item for complete abstract)

Doctor of Philosophy, The Ohio State University, 2025, Electrical and Computer Engineering

Electromagnetic (EM) waves propagating through the lower atmosphere are significantly influenced by the environmental factors. Variations in the index of refraction can lead to ducting events, altering the expected spherical spreading of transmitted waves and resulting in trapping of these waves. Developing methods to characterize, estimate, and predict these occurrences is crucial for understanding their impact and mitigating performance degradation in systems such as naval and land-over-sea based radar and communication systems. Traditionally, sensing atmospheric ducts involves sounding launches or applying Monin-Obukhov Similarity Theory (MOST) to meteorological sensor data (humidity/temperature/wind), often in costly and non-real-time processing scenarios. Given that ducting enhances clutter through signal trapping, this dissertation explores radar systems' sensitivity to range-dependent clutter and how radar power returns from sea and land clutter can be utilized to estimate oceanic ducts. Refractivity-From-Clutter (RFC) involves estimating the atmospheric index of refraction variation with altitude using radar clutter data. The inversion process in RFC utilizes a best-fit algorithm to align with a known refractivity profile. The Parabolic Wave Equation (PWE) model is widely used for simulating path loss in RFC scenarios. Accurate land and sea surface modeling for both forward and backward propagation is the propelling force that enables RFC to have an accurate real-time estimation of lower tropospheric refractivity profiles. Previous RFC work primarily used only sea clutter with a single launch angle. The motivation behind this research is to enhance the RFC method by exploring the advantages of multiple launch angles and by incorporating land clutter. This dissertation is structured in three main sections, each aimed at advancing the precision and applicability of the RFC technique. The first section extends prior work by incorporating the Multip (open full item for complete abstract)

Doctor of Philosophy, The Ohio State University, 2025, Electrical and Computer Engineering

The proliferation of counterfeit integrated circuits (ICs) poses a critical threat to the global electronics supply chain, with ramifications ranging from product failures to national security risks. This dissertation presents the design, implementation, and evaluation of a rapid, non-destructive counterfeit detection method leveraging a microwave resonant cavity, termed the Resonant Cavity (ResCav) System. The system captures electromagnetic (EM) signatures of ICs by measuring shifts in reflection coefficient values when components are placed inside a rectangular microwave cavity. These signatures are influenced by the IC's impedance, internal structure and material properties, enabling differentiation between authentic and counterfeit devices. The ResCav System was engineered through multiple design iterations. Coupled with a vector network analyzer (VNA), the cavity records EM data across a broad frequency band. These signatures are then analyzed using supervised, semi-supervised, and unsupervised machine learning techniques. Experimental evaluations demonstrate the system's ability to distinguish ICs from different manufacturers, families, and production lots, as well as to detect a range of counterfeit types, including recycled, remarked, tampered, and cloned devices. A specialized machine learning model employing a modified radial basis function kernel identified certain types of tampered ICs where modifications were made solely to the silicon die. Additionally, tests incorporating total ionizing dose (TID) exposure revealed the system's sensitivity to radiation-induced internal changes in some IC types, providing further validation of the method's robustness. The dissertation also addresses intra-system measurement variability, showing consistent results across multiple environmental and setup conditions. While certain limitations exist—such as the requirement to test chips outside of circuit boards and difficulty detecting counterfeits that are phy (open full item for complete abstract)

Doctor of Philosophy (Ph.D.), University of Dayton, 2025, Electrical Engineering

The growing aging population requires innovative solutions in the healthcare industry. Telemedicine is one such innovation that can improve healthcare access and delivery to diverse and aging populations. It uses various sensors to facilitate remote monitoring of physiological measures of people, such as heart rate, oxygen saturation, blood glucose, and blood pressure. Similarly, it is capable of monitoring critical events, such as falls. The key challenges in telemonitoring are ensuring accurate remote monitoring of physical activity or falls by preserving privacy and avoiding excessive reliance on expensive and/or obtrusive devices. Our approach initially addressed the need for secure, portable, and low-cost solutions specifically for fall detection. Our proposed system integrates a low-power millimeter-wave (mmWave) sensor with a NVIDIA Jetson Nano system and uses machine learning to accurately and remotely detect falls. Our initial work focused on processing the mmWave sensor's output by using neural network models, mainly employing Doppler signatures and a Long Short-Term Memory (LSTM) architecture. The proposed system achieved 79% accuracy in detecting three classes of human activities. In addition to reasonable accuracy, the system protected privacy by not recording camera images, ensuring real-time fall detection and Human Activity Recognition (HAR) for both single and multiple individuals at the same time. Building on this foundation, we developed an advanced system to enhance accuracy and robustness in continuous monitoring of human activities. This enhanced system also utilized a mmWave radar sensor (IWR6843ISK-ODS) connected to a NVIDIA Jetson Nano board, and focused on improving the accuracy and robustness of the monitoring process. This integration facilitated effective data processing and inference at the edge, making it suitable for telemedicine systems in both residential and institutional settings. By developing a PointNet neural network for (open full item for complete abstract)

Doctor of Philosophy, The Ohio State University, 2025, Electrical and Computer Engineering

Aviation electrification has become a key trend in the development of next-generation aircraft, driven by the growing demand for high power density, improved fuel efficiency, decarbonization, and quieter propulsion systems. In response, the concept of More Electric Aircraft (MEA) has emerged and been implemented in the aviation industry, with prominent examples including the Boeing 787 and Airbus A350 and A380. These aircraft employ onboard power generation systems to supply electricity to various subsystems, such as the de-icing system, environmental control system, and engine starting system. The prevailing architecture typically uses an engine or turbine as the prime mover for electrical generators. As the shaft speed fluctuates during operation, there is increasing interest in high-speed, variable-frequency power generation systems. This dissertation focuses on a proposed brushless doubly-fed machine (BDFM) featuring a flux-guide rotor. This topology presents several advantages over traditional machine types: 1. It eliminates permanent magnets and brushes or slip rings, enhancing fault tolerance as well as mechanical and thermal robustness. 2. The output frequency remains constant regardless of shaft speed, which simplifies the drivetrain by removing bulky speed conversion components. 3. The rotor's specialized design improves its mechanical performance for high-speed operation. 4. The presence of two independently measurable and controllable stator windings increases the degrees of freedom for advanced control strategies. Despite these merits, the low power factor behavior of BDFMs has received limited attention in the literature. As power factor is a critical indicator of machine efficiency, this research aims to explore the underlying causes of low power factor and its impact on the practical deployment of BDFMs. An analytical method is employed to derive the steady-state equivalent circuit of the prototype machine, facilitating a deeper investigation o (open full item for complete abstract)