Abstract: The development of an autonomous unmanned vehicle (AUV) presents both significant challenges and considerable benefits. This paper examines the issues encountered and the solutions implemented during the design and construction of the AUV, focusing specifically on the Kalypso robotic vehicle, which was developed for the inspection of fish farm nets. Kalypso is capable of detecting both overgrown and damaged nets through the application of advanced algorithms [1]. The robot is also equipped with a range of sensors that monitor key parameters, such as water ingress within the waterproof case, as well as its location, depth, temperature, and potential leaks. The waterproof section of the robot contains cables and connectors, some of which are detachable, allowing for the connection of various devices, including sensors and lights. Kalypso features two cameras, one positioned at the front and the other at the bottom, to provide comprehensive visual coverage. Additionally, the AUV is equipped with an underwater ultrasonic sensor that measures the distance to objects in its environment and an LED light at the front to enhance visibility in low-light conditions. A series of tests were conducted to assess the vehicle's performance in both aquaculture net environments and open-sea conditions.

Abstract: Permanent Magnet Synchronous Motors (PMSMs) require advanced control strategies to meet the high-performance demands of aerospace applications. Surface-mounted PMSMs (SPMSMs) have gained popularity in aeronautical systems due to their superior power density, efficiency, and reliability. In an SPMSM, the Iq current is directly proportional to torque, making Total Harmonic Distortion (THD) reduction essential for minimizing torque ripple and ensuring smooth operation. In this study, we propose a long prediction horizon Model Predictive Control (MPC) framework for PMSMs. Initial simulations using a one-norm cost function resulted in instability in switching frequency control, particularly due to the inherent limitations imposed by the sampling interval when no control effort was applied. To mitigate this, we reformulated the MPC framework using a two-norm cost function within a Sphere Decoding Algorithm (SDA), which at high sampling intervals (>40μs) resulted in an undershoot in the direct-quadrature axis. Extensive simulations were conducted over a range of sampling intervals (1–80μs), revealing that while a 10μs interval achieved the lowest THD, it also led to an increased switching frequency. To address this trade-off, a weighting factor tuning approach was employed, effectively reducing switching frequency while maintaining acceptable THD levels. Further investigations analyzed the effects of three-step and five-step prediction horizons, as well as parameter mismatches in the long prediction formulation, providing critical insights into controller robustness. These findings underscore the importance of norm selection, sampling interval optimization, and weighting factor adjustments in balancing THD reduction and switching frequency. The proposed approach enhances system efficiency, reliability, and overall performance, offering significant implications for high-performance aerospace PMSM applications.

Abstract: The growing demand for sustainable solutions in the automotive industry has prompted significant interest in the development of biodegradable materials for electric vehicle (EV) components. This project investigates a novel approach to enhance corrosion resistance in biodegradable materials, utilizing the Taguchi method to optimize manufacturing processes. Corrosion is a critical issue in automotive applications, affecting the durability and lifespan of components, particularly in harsh operating environments. As the automotive sector shifts towards eco-friendly alternatives, it becomes essential to understand and mitigate the corrosion behavior of new materials.This study aims to explore the interplay between advanced manufacturing techniques and the corrosion resistance of biodegradable materials, specifically focusing on their application in electric vehicles. The research begins with a comprehensive literature review that outlines current trends in EV production and the role of materials science in addressing corrosion challenges. Through the Taguchi method, which is designed to improve quality and efficiency, this project will systematically assess the factors influencing corrosion behavior in selected biodegradable materials.The experimental methodology will involve identifying key variables that impact corrosion resistance, designing a series of controlled experiments, and analyzing the results to establish optimal conditions for material performance. Additionally, advanced manufacturing techniques, such as additive manufacturing and other innovative processes, will be employed to enhance scalability and efficiency in producing these materials.Preliminary findings are expected to reveal significant correlations between manufacturing parameters and corrosion performance, providing insights into how biodegradable materials can be effectively utilized in EV applications. This research not only aims to contribute to the scientific understanding of corrosion in biodegradable materials but also seeks to offer practical solutions for the electric vehicle industry, enabling the production of more sustainable and durable components.Ultimately, this project underscores the importance of interdisciplinary approaches in addressing the challenges facing modern manufacturing and materials science. By integrating advanced manufacturing techniques with corrosion control strategies, this study aspires to pave the way for innovative solutions that align with the growing emphasis on sustainability in the automotive sector. The outcomes of this research could have far-reaching implications, promoting the adoption of biodegradable materials in electric vehicle production and contributing to a more sustainable future for transportation.

Abstract: Restrike frequently occurs in positive leader development of long air gap discharges. However, its detailed physical process and mechanism remains unclear. For the purpose of studying the physical mechanism of Restrike, experiments were conducted in a 6-meters rod-plane air gap under positive impulses with the crest time of 1000µs, and the process of Restrike was observed during discharge. Experimental results showed that significant luminescence appeared at the tip of the leader channel during a relatively longer time of discharge relaxation process before Restrike, and the luminescence became more and more intense with the increasing of applied voltage until Restrike occurred. By analyzing the charged particles composition inside the leader channel, this paper infers that during the relaxation process, the positive ions inside the leader channel migrate and concentrate towards the leader channel tip as the applied electrical field increasing, and the concentration of positive ions at the leader channel head distort and enhance the local field, which then induces the streamer corona discharge and lead to the luminescence of the leader channel. The observation evidence and analysis may provide valuable reference for a better understanding of the physical mechanism of Restrike.

Abstract: This study analyses the performance of a multi-user digital communication system aided by reflect intelligent surfaces (RIS) in terms of bit error probability and secrecy outage probability for a system sending symbols with M-QAM modulation passing through channels with Weibull fading where the RIS is employed to improve the signal to noise plus interference ratio (SINR) for each user. The performance analysis is conducted based on the statistical properties of the phase correction error of the transmitted signal, which follows a Von Mises distribution. Furthermore, this study demonstrates that the resulting SINR follows a Gamma distribution, with its parameters derived analytically. The RIS performance has increased the line of sight strength and reduced the secrecy outage probability and error probability when the number of reflectors is sufficiently large, even without direct links between the users and the transmitter.

Abstract: In this study, we developed core system software for a microcontroller-based, low-cost data acquisition system, along with the necessary environmental software. The system includes an analog-to-digital converter, digital inputs/outputs, pulse-width modulation, and a counter, supporting data transmission via USB 2.0 full-speed. We also designed flexible and easily configurable firmware to enable synchronization with external devices, making the system highly adaptable for applications such as laboratory research setups or scientific instrumentation. We created the schematic and printed circuit board for the core system, which is based on a 32-bit ARM Cortex-M4F microcontroller. In addition, we measure the signal-to-noise and distortion of the analog-to-digital converter to evaluate its conversion capability.

Abstract: FPGAs are becoming more popular for general-purpose computing, and AI/ML acceleration, but memory management on these reconfigurable hardware elements is still a challenge compared to fixed-logic architectures. We present a new, Machine Learning-driven Memory Management Unit (MMU) architecture for FPGAs, which employs intelligent algorithms (e.g., reinforcement learning and LSTM neural networks) to optimize memory response. In this work, we describe an ML-augmented MMU's architectural design and algorithmic framework, including virtual memory support, adaptive caching/prefetching, and dynamic allocation. Further showcasing latency, throughput, energy efficiency, and memory bandwidth benefits. We also show how we improve security mechanisms, relying on cache timing side-channels and speculative execution vulnerabilities, for cryptographic and ML algorithms. The design is adaptable across various applications (AI inference, high-performance computing, general workloads) and FPGA platforms. Finally, we describe the novelty when applied to a patent context with broad claims on machine learning applied to hardware memory management and security integration. This work is derived from the Provisional Patent Application #63/775,213, entitled "ML-Driven Memory Management Unit (MMU) in FPGA Architectures," filed on Mar 20, 2025, by Raj Sandip Parikh, with the United States Patent and Trademark Office (USPTO).

Abstract: Wireless Sensor Networks (WSNs) use distributed nodes for tasks such as environmental monitoring and surveillance. Existing anomaly detection methods fail to fully capture correlations in multi-node, multi-modal time series data, limiting their effectiveness. Additionally, they struggle with small sample scenarios, because they do not effectively map features to classes. To address these challenges, this paper presents an anomaly detection approach that integrates deep learning with metric learning. A framework incorporating a Graph Attention Network (GAT) and Transformer is developed to capture spatial and temporal features. A novel distance measurement module improves similarity learning by considering both intra-class and inter-class relationships. Joint metric-classification training improves model accuracy and generalization. Experiments conducted on public datasets demonstrate that the proposed approach achieves an F1 score of 0.89, outperforming existing approaches by 7%.

Abstract: This paper presents a systematic methodology for enhancing microstrip patch antenna (MPA) performance at 1.3 GHz for Phase Alternating Line (PAL) television broadcasting systems. Through the integration of slot-loaded patch geometries, substrate optimization, and array configurations, the proposed design achieves an 8.1 dBi gain and a 108 MHz bandwidth, representing improvements of 30% and 50% respectively, over conventional designs [1], [2]. The approach combines analytical modeling in MATLAB with full-wave electromagnetic simulations using CST Microwave Studio and is validated through precision measurements of fabricated prototypes. Key innovations include the implementation of a U-shaped slot for multi-resonant operation (∆f = 45 MHz per iteration), strategic selection of RT/Duroid 5880 substrate (ϵr = 2.2), and a 1 × 4 phased array configuration incorporating a defected ground structure (DGS) [5]. Experimental verification demonstrates 82% radiation efficiency and −22 dB cross-polarization isolation, fulfilling PAL TV specifications while maintaining compact dimensions (58 × 58 × 1.6 mm) [6], [7]. MATLAB and CST simulations analyze the antenna’s performance, including the reflection coefficient (S11) [11], [12]. The optimized MPA achieves a gain of 8 dBi with a bandwidth exceeding 100 MHz, aligning with the operational requirements of PAL TV applications [3], [13]. Future work will explore adaptive configurations and alternative substrate materials. [14], [15]

Abstract: This paper presents the design analysis of a low power wideband single-ended CMOS low-noise amplifier (LNA). The proposed topology is based on the modified current reuse circuit to achieve good performance and low power consumption. Two stage current source (CS) amplifiers consume the same DC current which are isolated with large MIMCAPs. The proposed circuit has 2.5 GHz bandwidth which can cover several wireless communication standards (GSM, WLAN and Bluetooth). In first stage a current reuse circuit with shunt feedback is used to satisfy input impedance matching and amplify the signal with minimal noise injection. A common source (CS) with a source follower circuit forms the second stage to improve NF, harmonic distortion and also output impedance matching. The proposed LNA is designed in 65-nm CMOS technology with 2.51 GHz bandwidth that covers frequency range of 0.17-2.68 GHz. The post-layout simulation results show a maximum S₂₁ of 17.24 dB, minimum NF of 2.67 dB, maximum IIP3 of -14.9 dBm, input and output return losses less (S₁₁, S₂₂) than -10 dB while power consumption is 3.52 mW from 1 V power supply. Excluding pads, the proposed circuit occupies 0.45 mm² silicon die area.

Abstract: Power efficiency has emerged as a significant constraint in the design of modern digital systems, requiring a holistic approach that includes architectural, register-transfer level, and physical implementation stages. This survey comprehensively reviews low-power design techniques for FPGA and ASIC technologies developed in the last five years. It addresses high-level synthesis optimizations, RTL power-aware methodologies, and dynamic power management techniques, including clock gating, power gating, and voltage scaling. The backend activities discussed in the paper are power-driven placement, multi-threshold and multi-voltage design, leakage minimization, and robust power grid architectures. The quantitative trends in junction scaling and process technology, such as FinFET and GAA transistors, are elaborated along with emerging paradigms of chipset-based integration and machine learning-driven design automation. It also addresses application-specific low-power techniques for application domains such as IoT, AI accelerators, and high-performance computing. The paper ends with directions toward adaptive, context-aware systems that minimize real-time power consumption (given the workload and changing conditions). This work presents a coherent reference for designers, researchers, and engineers to comprehend cutting-edge low-power design methodologies through a unified consolidation of the scientific and industrial body of knowledge.

Abstract: A new Common-Mode (CM) choke with asymmetrical winding capable of attenuating Differential-Mode (DM) and Common-Mode (CM) noise for AC to DC Power Converters is proposed in this paper. For a conventional CM choke with symmetrical winding, the CM and DM noise are also attenuated by the magnetizing inductance (LM) and very small leakage inductance (Lk) due to the higher coupling factor, respectively, of the CM choke. However, because the number of turns is limited according to the shape and size of the CM choke, it is very difficult to find the proper leakage inductance (Lk) to eliminate the DM noise thoroughly. For this reason, a bulky CM choke is required for a larger leakage inductance (Lk). Typically, another DM choke is used, for the sole purpose of DM noise attenuation. Therefore, the cost and size of the EMI filters for use with a conventional CM choke can increase. To solve these problems, the proposed CM choke with asymmetrical winding is introduced in this paper. Different from the conventional CM choke with symmetrical winding, the magnetizing inductance (LM) of the proposed CM choke with asymmetrical winding can suppress not only CM but also DM noise. Therefore, the proposed CM choke with asymmetrical winding can effectively attenuate CM and DM types of noise even with fewer and smaller turns. This method can be applied to all types of power converters which use two- or three-stage systems. To confirm the validity and superiority of the proposed CM choke, comparisons of the total noise as well as the DM and CM noise characteristics between conventional and proposed CM chokes are presented in this paper. Lastly, a practical approach to the design of an EMI filter and the design procedures are also presented and addressed.

Abstract: This paper proposes a methodology for the co-expansion planning of transmission lines and energy storage devices, considering unit commitment constraints and uncertainties in load demand and wind generation. The problem is formulated as a mixed-integer nonlinear program (MINLP) and solved using a decomposition-based approach that combines a genetic algorithm with mixed-integer linear programming (MILP). Uncertainties are modeled through representative day scenarios obtained via clustering. The proposed methodology is validated on a modified IEEE 24-bus system. Results show that co-planning reduces wind curtailment, fuel costs, and total investment costs compared to transmission-only expansion.

Abstract: This work proposes a High-Brightness LED (HB-LED) driver for Visible Light Communication (VLC) based on two converters, a high frequency Buck DC/DC converter and a low frequency Boost DC/DC converter, connected in series with respect to the LED load and connected in parallel at the input. This topology is called a Series/Parallel Boost/Buck DC/DC Converter. A VLC system needs to do two different tasks: biasing the HB-LED and generating the communication signal. These typically have different power requirements, the bias power is 3/4, while the communication power is 1/4 of the total power. The requirements of each are also different: the communication signal requires a high frequency, fast output response, while the biasing control requires a converter with a slow output voltage response. The proposed architecture takes advantage of the differences between the two tasks and achieves high efficiency and high communication performance by means of splitting the power between the two DC/DC converters. A high frequency Buck DC/DC converter generates the communication signal, while the low frequency Boost DC/DC converter is responsible for biasing the LEDs. This technique allows most of the DC biasing power to be processed by the low frequency converter (achieving high efficiency), keeping the high frequency converter delivering the communication power (achieving high communication performance). To provide experimental results, the proposed VLC HB-LED driver was built and validated by reproducing a 64-QAM with a bit rate up to 1.5 Mbps, reaching 91.5% overall efficiency.

Abstract: This review presents the band selective frequency technology of Electromagnetic Interference (EMI) noise spectrum spread in the DC-DC switching converter for communication devices. This technology generates notch characteristic spectrum bands with a low noise level in the received frequency band spectrum. It detects the received frequency and generates a notch band there using a switching pulse control technology. First, we introduce the conventional spread spectrum technology. By modulating the clock frequency, EMI noise is dispersed to avoid concentrating at specific frequency bands. There are both analog modulation techniques and digital modulation methods. Next, we explain the main technology of this review, the notch band generation technology. This technique involves modulating the phase or pulse width of clock to produce notch band characteristics in the EMI noise spectrum. Then we present its simulation results, theoretical analysis, and implementation results. Finally, we demonstrate a technique that tunes the notch band frequency to the received signal one automatically.

Abstract: A novel multi-terminal quantum transistor device is presented that leverages coherent scattering and multi-path interference to enable multi-input, multi-output signal processing beyond the conventional binary switching paradigm. A tight-binding framework is employed to model a disordered two-dimensional lattice, and quantum transport is analyzed using scattering matrix formalism. The device demonstrates energydependent transmission characteristics with channelresolved information capacities exceeding one bit per cycle. It is proposed that by scaling such devices into dense VLSI architectures, a single quantum transistor element may replace hundreds of classical transistors, thus paving the way for a new class of semiconductor devices with exponentially enhanced computational density.

Abstract: (1) Background and objective: Mobility is crucial for healthy aging, and its loss significantly impacts the quality of life, healthcare costs, and mortality among older adults. Clinical mobility assessment methods, though precise, are resource-intensive and economically impractical and most of the existing solutions for automatic detection of mobility anomalies are either obtrusive, or improper for long time monitoring. This study explores the feasibility of using non-intrusive, low-cost binary sensors for continuous, remote detection of mobility anomalies in older adults, aiming to identify both sudden mobility events and gradual mobility loss. (2) Method: The study utilized publicly available datasets (CASAS Aruba and HH120) containing annotated activity data recorded from binary sensors installed in residential environments. After data preprocessing—including filtering irrelevant sensor events and aggregation into behaviorally meaningful places (BMPs)—a time series forecasting model (Prophet) was used to predict normal mobility patterns. A fuzzy inference module analyzed deviations between observed and predicted sensor data to determine the probability of mobility anomalies. (3) Results: The system effectively identified periods of prolonged inactivity indicative of potential falls or other mobility disruptions. Preliminary evaluation indicated a detection rate of approximately 77–81% for point mobility anomalies, with a false positive rate ranging from 12–16%. Additionally, the approach successfully detected simulated gradual declines in mobility (1% per day reduction), evidenced by statistically significant regression trends in activity levels over time. (4) Conclusion: The study argues that non-intrusive binary sensors, combined with lightweight forecasting models and fuzzy inference, may provide a practical and scalable solution for detecting mobility anomalies in older adults. Although performance can be further enhanced through improved data preprocessing, predictive modeling, and anomaly threshold tuning, the proposed system effectively addresses key limitations of existing mobility assessment approaches.

Abstract: This review paper presents a comprehensive analysis of electroluminescence (EL) imaging techniques for photovoltaic (PV) module diagnostics, focusing on advancements from conventional indoor imaging to outdoor and daylight EL imaging. It examines key challenges, including ambient light interference and environmental variability, and highlights innovations such as infrared-sensitive indium gallium arsenide (InGaAs) cameras, optical filtering, and periodic current modulation to enhance defect detection. The review also explores the role of artificial intelligence (AI)-driven methodologies, including deep learning and generative adversarial networks (GANs), in automating defect classification and performance assessment. Additionally, the emergence of drone based EL imaging has facilitated large-scale PV inspections with improved efficiency. By synthesizing recent advancements, this paper underscores the critical role of EL imaging in ensuring PV module reliability, optimizing performance, and supporting the long-term sustainability of solar energy systems.

Abstract: The rapid development of electronic technology has generated the demand for low - power and high- performance circuit design. The key indicator, power supply rejection ratio (PSRR), is crucial for circuit design. Given the deficiencies of existing PSRR calculation methods in high-frequency applications and complex circuit designs, this pa-per proposes an innovative PSRR calculation theory. Based on a simplified circuit model, this theory uses Thevenin's equivalent principle to transform multi-stage operational amplifiers into a black-box model, simplifying the calculation process and enhancing the intuitiveness of calculation. In view of the characteristics of industrial design, we also proposed the PSR Gain Bandwidth (PGB) theory, which can more intuitively analyze the trade-off between PSRR and other performance within the target range. This paper further explores the impact of different circuit structures on PSRR characteristics, covering typical circuit structures such as PMOS/NMOS input two-stage operational amplifiers and folded cascode operational amplifiers, and deeply analyzes the key factors affecting PSRR characteristics. The effectiveness of the proposed theory is verified through case analysis, and its potential applications in circuit designs with high PSRR requirements are demonstrated.

Abstract:

The FPGA and ASIC debugging, boundary scan testing, and device coding owe vivid gratitude to JTAG Interfaces (Joint Test Action Group format adhering largely to IEEE 1149.1 standards). In this paper, we experiment with an AI-based method for JTAG log monitoring and performance trend forecasting. Using deep learning models such as LSTMs and Transformers, the system can find deviations from log patterns and predict potential failures in advance. This kind of closed-loop analysis enhances our reliability to unprecedented levels. The system described in this work is made for hybrid cloud deployment, providing secure, scalable, and real-time log analysis software. The paper further discusses the architectural integration of AI into existing JTAG frameworks for FPGAs and RISC-V ASICs, detailing security considerations, implementation challenges, and potential industry applications. This paper is based on a patent: AI-Driven Hybrid Cloud JTAG Log Monitoring System for FPGA Debugging and Failure Prediction (Patent Number 63/771,667).