Electronics

The rapid integration of large-scale AI models into distributed systems, such as the Artificial Intelligence of Things (AIoT), has introduced critical security and privacy challenges. While configurable models enhance resource efficiency, their deployment in heterogeneous edge environments remains vulnerable to poisoning attacks, data leakage, and adversarial interference, threatening the integrity of collaborative learning and responsible AI deployment. To address these issues, this paper proposes a Hierarchical Federated Cross-domain Retrieval (FHCR) framework tailored for secure and privacy-preserving AIoT systems. By decoupling models into a shared retrieval layer (globally optimized via federated learning) and device-specific layers (locally personalized), FHCR minimizes communication overhead while enabling dynamic module selection. Crucially, we integrate a retrieval-layer mean inspection (RLMI) mechanism to detect and filter malicious gradient updates, effectively mitigating poisoning attacks and reducing attack success rates by 20% compared to conventional methods. Extensive evaluation on General-QA and IoT-Native datasets demonstrates the robustness of FHCR against adversarial threats, with FHCR maintaining global accuracy not lower than baseline levels while reducing communication costs by 14%.

Knitting, a cornerstone of textile manufacturing, is uniquely challenging to automate, particularly in terms of converting fabric designs into precise, machine-readable instructions. This research bridges the gap between textile production and robotic automation by proposing a novel deep learning-based pipeline for reverse knitting to integrate vision-based robotic systems into textile manufacturing. The pipeline employs a two-stage architecture, enabling robots to first identify front labels before inferring complete labels, ensuring accurate, scalable pattern generation. By incorporating diverse yarn structures, including single-yarn (sj) and multi-yarn (mj) patterns, this study demonstrates how our system can adapt to varying material complexities. Critical challenges in robotic textile manipulation, such as label imbalance, underrepresented stitch types, and the need for fine-grained control, are addressed by leveraging specialized deep-learning architectures. This work establishes a foundation for fully automated robotic knitting systems, enabling customizable, flexible production processes that integrate perception, planning, and actuation, thereby advancing textile manufacturing through intelligent robotic automation.

Small to medium-sized shipyards play a crucial role in the European naval industry. However, the globalization of technology has increased competition, posing significant challenges to shipyards, particularly in domestic markets for short sea, work, and inland vessels. Many shipyard operations still rely on manual, labor-intensive tasks performed by highly skilled operators. In response, the adoption of new tools is essential to enhance efficiency and competitiveness. This paper presents a methodology for developing a human-centric portfolio of advanced technologies tailored for shipyard environments, covering processes such as shipbuilding, retrofitting, outfitting, and maintenance. The proposed technological solutions, which have achieved high technology readiness levels, include 3D modeling and digitalization, robotics, augmented and virtual reality, and occupational exoskeletons. Key findings from real-scale demonstrations are discussed, along with major development and implementation challenges. Finally, best practices and recommendations are provided to support both technology developers seeking fully tested tools and end users aiming for seamless adoption.

As wireless communication technology advances towards faster and higher transmission rates such as Fifth Generation (5G) and beyond, the need for multiple access points increases. The growing demand for access points often results in them occupying any available surface area and potentially disrupting the existing scenery. In order to address this issue, Optically Transparent Antennas (OTAs) emerge as an optimal solution for balancing the aesthetics of a specific setting with the desired communication system requirements. These antennas can be integrated into various infrastructures without interfering with the design of the objects on which they are installed. Research on the techniques and materials for OTA fabrication, which is proposed as a solution to the 5G wireless communication demand for access points, is presented. This work will highlight key antenna characteristics such as gain, bandwidth, efficiency, and transparency, and how the materials used for OTA implementation influence these parameters. Techniques like Metal Mesh (MM), Transparent Conductive Film (TCF), and Transparent Conductive Oxide (TCO) will be explained. The performance of the OTAs will be analyzed based on gain, bandwidth, transparency, and efficiency. This paper also addresses the challenges and limitations associated with OTAs. Finally, it confirms that OTAs offer a compelling solution for this scenario by balancing aesthetics with high antenna performance, making them an innovation for future wireless networks.

Frozen gait (FG) is an increasingly prevalent concern in individuals with Parkinson’s disease (PD) that limits mobility and increases the risk of falls. Traditional FG detection and monitoring methods using clinical observations and wearable sensors face limitations, such as inflexibility, lack of portability, inaccessibility to individuals, and the inability to provide continuous monitoring in real-life environments. To address these challenges, this experimental study presents the development of a software-defined radio (SDR)-based radio frequency (RF) sensing platform for continuous FG monitoring. Data were collected through multiple experiments involving various physical activities, including FG episodes. The acquired data were processed using advanced signal-processing (ASP) techniques to extract relevant wireless channel state information (WCSI) patterns. The physical activities were classified using machine learning and deep learning models developed on the dataset prepared from the SDR-based RF sensing system. The results demonstrated that the deep learning models outperformed the machine learning models. The bidirectional gated recurrent unit (BiGRU) achieved the highest accuracy of 99.7%. This indicates that the developed system has the potential for accurate, real-time monitoring of FG and other PD symptoms. The proposed RF sensing platform using SDR technology and artificial intelligence (AI) offers an intelligent and continuous monitoring solution, addressing the limitations of traditional methods. This system provides portable, continuous detection of FG events, potentially improving patient care, safety, and early intervention.

Image steganalysis has been a key technology in information security in recent years. However, existing methods are mostly limited to the binary classification for detecting steganographic images used in digital watermarking, privacy protection, illicit data concealment, and security images, such as unaltered cover images or surveillance images. They cannot identify the steganography algorithms used in steganographic images, which restricts their practicality. To solve this problem, this paper proposes a general steganography algorithms recognition scheme based on image big data matching with improved ResNet50. The scheme first intercepts the image region with the highest complexity and focuses on the key features to improve the analysis efficiency; subsequently, the original image of the image to be detected is accurately located by the image big data matching technique and the steganographic difference feature image is generated; finally, the ResNet50 is improved by combining the pyramid attention mechanism and the joint loss function, which achieves the efficient recognition of the steganography algorithm. To verify the feasibility and effectiveness of the scheme, three experiments are designed in this paper: verification of the selection of the core analysis region, verification of the image similarity evaluation based on Peak Signal-to-Noise Ratio (PSNR), and performance verification of the improved ResNet50 model. The experimental results show that the scheme proposed in this paper outperforms the existing mainstream steganalysis models, such as ZhuNet and YeNet, with a detection accuracy of 96.11%, supports the recognition of six adaptive steganography algorithms, and adapts to the needs of analysis of multiple sizes and image formats, demonstrating excellent versatility and application value.

Applying the acoustic orbital angular momentum (AOAM) wave for underwater imaging can yield richer differential target echo information, a consequence of its spiral wavefront phase and multiple mutually orthogonal modes. In broadband AOAM wave imaging, the resolution of conventional beamforming is very low. Although sub-band processing can improve resolution, it cannot handle coherent signal sources. To further enhance the resolution of broadband AOAM wave underwater imaging and address the imaging issue of coherent signals in practice, this paper proposed a modal-domain focusing beamforming method. This paper initially established the echo signal model of broadband AOAM waves based on a uniform circular array. This was followed by the derivation of the beam output signal model. Finally, a new modal-domain focusing transformation matrix was constructed. Numerical results show that the proposed method reduces the background level of the beam pattern to −86dB in simple coherent target source imaging, compared with −40dB for sub-band methods and −70dB for plane wave focusing processing. Furthermore, under different noise conditions, the proposed method achieves high-resolution imaging of complex structures and good imaging of details.

The increasing number of power electronic devices in power networks causes a significant increase in supraharmonics in these networks. Supraharmonics are spectral components in the 2–150kHz bandwidth that cause high-frequency signal distortions that can disturb the operation of other supplied loads, including in the field of communication or control. In the case of an increase in the occurrence of supraharmonics, it is necessary to identify the source of the disturbance, taking into account, among others, the indication of its supply point. This article presents the results of observations of supraharmonics in modern power networks. Based on results of long-term research carried out in controlled laboratory conditions and under a real power network in industrial conditions, significant diagnostic problems in the identification of supraharmonic sources related to the influence of typical loads in a low-voltage network are indicated. For the presented cases, the propagation of selected spectral components in a low-voltage network with a branched radial topology is presented. The influence of typical loads in low-voltage networks on the diagnosis of supraharmonics in modern power systems is presented. The possibilities of amplification or supression of supraharmonics by loads that are not their source are demonstrated, depending on their supply point in the power network.

The development of deep neural networks, although demonstrating astounding capabilities, leads to more complex models, high energy consumption, and expensive hardware costs. While network quantization is a widely used method to address this problem, the typical binary neural networks often require the batch normalization (batchnorm) layer to preserve their classification performances. The batchnorm layer contains full-precision multiplication and the addition operation that requires extra hardware and memory access. To address this issue, we present a batch normalization-free binarized deep spiking neural network (B-SNN). We combine spike-based backpropagation in a spiking neural network with weight binarization to further reduce the memory and computation overhead while maintaining comparable accuracy. Weight binarization reduces the huge amount of memory storage for a large number of parameters by replacing the full-precision weights (32 bit) with binary weights (1 bit). Moreover, the proposed B-SNN employs the stochastic input encoding scheme together with a spiking neuron model, thereby enabling networks to perform efficient bitwise computations without the necessity of using a batchnorm layer. As a result, our experimental results demonstrate that the efficacy of the proposed binarization scheme on deep SNNs outperforms the conventional binarized convolutional neural network.

The rapid integration of renewable energy and electric vehicles is challenging modern distribution networks with increased demand volatility and overload risks. To address these issues, we propose an integrated, multi-timescale battery dispatch framework that unifies long-term capacity planning, day-ahead/intra-day scheduling, and sub-minute real-time control. The framework combines HOMER Pro-based capacity sizing, a MISOCP model for economic scheduling, and an agent-based simulation for immediate overload mitigation. In a case study of a Danish distribution network projected to reach full EV penetration by 2034, our approach reduced moderate-to-severe overload incidents by 82.7%. Furthermore, a price-sensitive variant achieved a 27.4% reduction in operational costs, with only a 12.5% increase in minor overload events. These quantitative improvements, alongside qualitative enhancements in grid stability and battery longevity, provide actionable insights for distribution system operators.

Industrial visual inspection plays a crucial role in intelligent manufacturing. However, existing anomaly-detection methods based on unsupervised learning paradigms often struggle with issues such as overlooking minor defects and blurring component edges in confidence maps. To address these challenges, this paper proposes an industrial anomaly-detection method based on component-level feature enhancement. This method introduces a component-level feature-enhancement module, which optimizes feature matching by calculating the structural similarity between global coarse-grained confidence features and local fine-grained confidence features, thereby generating enhanced feature maps to improve the model’s detection accuracy for minor defects and local anomalies. Additionally, we propose a region-segmentation method based on multi-layer piecewise thresholds, which effectively distinguishes between foreground and background in confidence maps, circumvents background interference and ensures the integrity of structural information of foreground components. Experimental results demonstrate that the proposed method surpasses comparative methods in both logical and structural defect detection tasks, showing significant advantages, especially in fine-grained anomaly detection, with stronger robustness and accuracy.

Mobile-assisted language learning (MALL) has emerged as a powerful tool for language education, offering flexibility, multimedia integration, and personalized learning experiences. Despite its growing adoption, most studies have focused on user perceptions and learning outcomes, with limited attention given to systematically evaluating the design, content, and pedagogical efficacy of mobile language learning apps (MLLAs). To address this gap in the effective design of MALL tools, this study developed an evaluation framework by integrating and refining elements from three established models in the field. The framework is organized into four dimensions: background and characteristics, app design, app content, and app pedagogy. It incorporates objective criteria alongside a standardized scoring system (0–2) to ensure consistent and systematic evaluations. The resulting framework provides researchers and educators with a tool to analyze and compare MLLAs based on their alignment with effective teaching and learning principles. This study contributes to the advancement of MALL app evaluation, supporting their development and improving teaching practices and learner outcomes.

This study proposes a force–position hybrid control method for quadruped robots based on the Model Predictive Control (MPC) algorithm, aiming to address the challenges of stability and adaptability in complex terrain environments. Traditional control methods for quadruped robots are often based on simplified models, neglecting the impact of complex terrains and unstructured environments on control performance. To enhance the real-world performance of quadruped robots, this paper employs the MPC algorithm to integrate force and position control to achieve precise force–position hybrid regulation. By transforming foot-end forces into joint torques and optimizing them using kinematic inverse solutions, the robot’s stability and motion accuracy in challenging terrains is further enhanced. In this study, a Kalman filter-based state estimation method is adopted to estimate the robot’s state in real time, enabling closed-loop control through the MPC framework, combined with kinematic inverse solutions for hybrid control. The experimental results demonstrate that the proposed MPC algorithm significantly improves the robot’s adaptability and control accuracy across various terrains. In particular, it exhibits superior performance and robustness in multi-contact and uneven terrain scenarios. This research provides a novel approach for deploying quadruped robots in dynamic and complex environments and offers strong support for further optimization of motion control strategies.

Post-stroke finger dysfunction severely impacts patients’ daily living abilities and quality of life. Traditional rehabilitation assessment methods face challenges such as high subjectivity, insufficient precision, and difficulty in capturing subtle changes. These challenges are particularly pronounced in small-sample data scenarios, where the accuracy and robustness of assessment models are limited. This study proposes an intelligent rehabilitation assessment method tailored for small-sample scenarios, combining the rehabilitation matching value (RMV) with machine learning to address the challenges of rehabilitation assessment in such contexts. A rehabilitation matching value calculation model is constructed based on existing data, and interpolation methods are employed to expand the small-sample dataset. Machine learning models are then utilized for validation. Experimental results demonstrate that the proposed method effectively captures subtle changes in finger function, significantly improving the sensitivity and accuracy of rehabilitation assessments. This provides a scientific basis for the development of personalized rehabilitation training plans. Compared to traditional methods, the proposed approach exhibits significant advantages in flexibility, practicality, and adaptability to small-sample scenarios.

A precision resistance measurement method based on synchronous sampling is proposed to enable accurate resistance measurement under low-current testing conditions. This method utilizes a single current source input, connecting the standard resistor and the test resistor in series. The setup is simple, easy to operate, and allows for the convenient and rapid comparison of the voltage across the resistors. The resistance value of the test resistor is determined through calculation, achieving high measurement accuracy. An experimental comparison of two standard resistors under a low current of 1 μA demonstrated that the data dispersion reached the 10−6 level, and the measurement error was within the 10−6 range. This study also employed a resistance ratio measurement method based on room-temperature Direct Current Comparator (DCC) technology to validate the superiority of the proposed synchronous sampling approach under low-current measurement conditions. The comparative analysis demonstrated that our method exhibits significant advantages over conventional DCC techniques under a low current of 1 μA. This paper presents the first demonstration of high-accuracy resistance ratio measurement using synchronous sampling under weak current conditions. Experimental results verify that at 1 μA current level, the proposed method achieves superior measurement accuracy and lower uncertainty compared to conventional precision DC current comparator techniques under identical test conditions.

Selecting appropriate locations of emergency centers is an important issue in avoiding probable damages by natural disasters. Emergency rescue sites are constructed to provide emergency supplies swiftly for people in affected areas. Factors of transportation fluency and road damage degrees should be considered, which largely affect rescue efficiency. In order to find appropriate sites accurately, we proposed a redesigned method Variable Butterfly Optimization Algorithm (VBOA), based on the Butterfly Optimization Algorithm, by adding the Variation Operator mechanism to avoid the limitations of local optimum problems present in other optimization algorithms. The Variation Operator effectively combines both global and local search strategies to improve the performance of global searching, and it accelerates the convergence speed of the algorithm. We conducted our experiment on selected candidate sites with multiple optimization methods; the experiment results demonstrate that our proposed method maintains the balance between conditions of coverage area and expenditure. Our proposed method relieved the reliance of local optimum results and achieved better convergence accuracy in our selected samples in comparison with other methods both in initial and later siting phases.

The adoption of kiosks in public spaces is steadily increasing, with a trend toward providing more natural user experiences through embodied conversational agents (ECAs). To achieve human-like interactions, ECAs should be able to appropriately gaze at the speaker. However, kiosks in public spaces often face challenges, such as ambient noise and overlapping speech from multiple people, making it difficult to accurately identify the speaker and direct the ECA’s gaze accordingly. In this paper, we propose a lightweight gaze control system that is designed to operate effectively within the resource constraints of kiosks and the noisy conditions common in public spaces. We first developed a speaker detection model that identifies the active speaker in challenging noise conditions using only a single camera and microphone. The proposed model achieved a 91.6% mean Average Precision (mAP) in active speaker detection and a 0.6% improvement over the state-of-the-art lightweight model (Light ASD) (as evaluated on the noise-augmented AVA-Speaker Detection dataset), while maintaining real-time performance. Building on this, we developed a gaze control system for ECAs that detects the dominant speaker in a group and directs the ECA’s gaze toward them using an algorithm inspired by real human turn-taking behavior. To evaluate the system’s performance, we conducted a user study with 30 participants, comparing the system to a baseline condition (i.e., a fixed forward gaze) and a human-controlled gaze. The results showed statistically significant improvements in social/co-presence and gaze naturalness compared to the baseline, with no significant difference between the system and human-controlled gazes. This suggests that our system achieves a level of social presence and gaze naturalness comparable to a human-controlled gaze. The participants’ feedback, which indicated no clear distinction between human- and model-controlled conditions, further supports the effectiveness of our approach.

The development of vehicle-to-grid (V2G) technique and the growth of battery swapping stations are expected to enhance the resilience of power networks. However, V2G battery swapping stations exhibit inconsistencies among internal battery packs, where the power capacity is significantly affected by the battery swapping behavior of electric vehicle (EV) users. To address this issue, this paper proposes a secondary frequency control strategy for V2G battery swapping stations that accounts for battery pack heterogeneity. First, a user behavioral model is developed through quantitative analysis of key factors such as economic incentives, time costs, and battery degradation, which is then used to optimize the operation of V2G battery swapping stations. Moreover, active balancing of EV battery energy levels is achieved by incorporating penalty terms into the objective function. Finally, a distributed secondary frequency control strategy based on the consensus algorithm is established to minimize total frequency control loss. Simulation results demonstrate that the proposed strategy effectively meets the secondary frequency control requirements of the power grid.

Human activity recognition (HAR) is vital for applications in fields such as smart homes, health monitoring, and navigation, particularly in GNSS-denied environments where satellite signals are obstructed. Wi-Fi channel state information (CSI) has emerged as a key technology for HAR due to its wide coverage, low cost, and non-reliance on wearable devices. However, existing methods face challenges including significant data fluctuations, limited feature extraction capabilities, and difficulties in recognizing complex movements. This study presents a novel solution by integrating a multi-sensor array of Wi-Fi CSI with deep learning techniques to overcome these challenges. We propose a 2 × 2 array of Wi-Fi CSI sensors, which collects synchronized data from all channels within the CSI receivable range, improving data stability and providing reliable positioning in GNSS-denied environments. Using the CNN-LSTM-attention (C-L-A) framework, this method combines short- and long-term motion features, enhancing recognition accuracy. Experimental results show 98.2% accuracy, demonstrating superior recognition performance compared to single Wi-Fi receivers and traditional deep learning models. Our multi-sensor Wi-Fi CSI and deep learning approach significantly improve HAR accuracy, generalization, and adaptability, making it an ideal solution for GNSS-denied environments in applications such as autonomous navigation and smart cities.

In this paper, a wideband OAM antenna array for wireless communication is proposed, which has a wide impedance bandwidth and can cover the S-C band with a relative bandwidth of 61.58%. The measured gain can reach 7.81 dBi and the radiation efficiency can reach 74.7%. Compared with similar antennas, the antenna array has a metal back cavity as the supporting structure, which further improves the structural stability of the array. The array adopts Z-shaped parasitic radiation units, a ring-shaped stepped metal reflection back cavity, and other structures. These can be verified to improve the performance of the array after design analysis and testing. In addition, the performance enhancement of a conventional Wilkinson divider by adding the S-shaped parasitic radiation patch is analysed by parameter scanning. The array is robust, simple to process, and easy to integrate. It can maximise its value in the crowded retrofit space of wireless antennas.

The widespread adoption of variable frequency air conditioners (VFACs) in household appliances is primarily driven by their energy-saving qualities. However, extremely high ambient temperatures and limited space affect the heat dissipation of the electronic control module of a VFAC, resulting in a substantial increase in the temperature of its electronic chips. Its reliability and working performance will be largely compromised. To address this issue, we propose a feasible thermal management design based on thermoelectric coolers (TECs) that can cool electronic control modules working in an extremely high ambient temperature of 55 °C. Firstly, we designed four cooling schemes and established simulation models via Ansys Icepak. Then, we compared the chip temperatures across different schemes. The results indicate that the average temperatures of IPM, IGBT, FRD, and Rectifier Bridge were reduced by 13.58 °C, 14.03 °C, 15.88 °C, and 15.56 °C, respectively, in the scheme incorporating TECs, indicating that TECs have a significant impact on the thermal management of electronic control modules. This enables VFACs to operate at their full potential in extremely high ambient temperatures. This study explores the potential of using TECs to cool the electronic control modules of VFACs in extremely high ambient temperatures, suggesting that TECs can be effectively utilized at a large scale in the commercial VFAC field.

A three-stage low-noise amplifier (LNA) operating over the 6–18 GHz frequency range is designed and implemented, featuring a flat noise figure (NF) and enhanced output 1 dB compression point (OP1dB). To improve linearity and minimize distortion, a power high-electron-mobility transistor (HEMT) is employed in the final stage. Additionally, resistive feedback and self-biasing techniques are integrated to extend the amplifier’s bandwidth. The proposed LNA exhibits a high and flat power gain of 25 ± 1 dB, with an input return loss of more than 10 dB. The measured NF remains stable at 2.6 ± 0.3 dB over the 6–18 GHz range. Furthermore, the OP1dB exceeds 19.5 dBm across the entire 3 dB gain bandwidth (BW). The circuit is fabricated using a 0.15 μm GaAs pHEMT process, occupying a compact chip area of 1.2 × 1.8 mm2.

Researchers are actively exploring advanced algorithms to enhance robots’ ability to navigate complex environments while avoiding obstacles. Four different environments were designed in the Webots simulator, including a mobile robot, a goal, a static obstacle, and one or two dynamic obstacles. The robot’s state vector was determined based on its position, the goal, and sensor variables, with all elements randomly placed in each learning and test step. A multi-layer perceptron (MLP) agent was trained for 1000 episodes in these environments using classical and fuzzy logic-based reward functions. After the training process was completed, the agents trained with the fuzzy logic-based reward function were tested for each environment. As a result of the test, while the robot’s arrival rate was 100% in the first three environments, it was measured as 91% in the fourth environment. In the last environment, the rate of crashing into a wall or dynamic obstacle was observed to be 7%. In addition, the agent trained in the fourth environment was found to successfully reach the target in multi-robot environments. The agent trained fuzzy logic-based reward function obtained the best result for four different environments. Based on these results, a fuzzy logic-based reward function was proposed to address the tuning problem of the classical reward function. It was demonstrated that a robust fuzzy logic-based reward function was successfully designed. This study contributed to the literature by presenting a reinforcement learning-based safe navigation algorithm incorporating a fuzzy logic-based reward function.

The planar parallel robots are widely employed in industrial applications due to simple geometry, few linkage interferences, and a large, reachable workspace. The symmetric geometry can bring significant convenience to parallel robots. The complexity of the mathematic models can be simplified since only one calculation method can be proposed to deal with various kinematic limbs in a parallel manipulator. The symmetric geometry can ease the assembly and maintenance procedures due to the modular design of linkages/joints. A novel 2-translation and 1-rotation (2T1R) parallel robot with symmetric geometry is proposed in this research. There is one closed loop in each kinematic limb, and 18 revolute joints are applied in its planar structure. Both the inverse and direct kinematic models are explored. The first-order relationship between robot inputs and outputs are constructed. Various singularity configurations are obtained based on the Jacobian matrix. The reachable workspace is resolved by the discrete spatial searching methodology, followed by the impacts originating from various linkages. The dexterity analysis of the parallel robot is conducted.

In this study, GaN devices are implemented in low-power three-phase inverters to achieve high-frequency operation and a compact design. A 500 W power-rated prototype operating at a switching frequency of 200 kHz was designed. A linear active disturbance rejection control (LADRC) strategy was introduced to solve the issue of high-frequency inverter oscillation. Additionally, the driving circuit was optimized to address the issue of voltage spikes in the drive voltage. Subsequently, the feasibility and effectiveness of the control strategy were validated through a system simulation model built in Simulink. Lastly, a digital control algorithm was developed by leveraging the capabilities of DSP28034. During testing at an operating frequency of 200 kHz, it was found that the drive voltage waveform was excellent and that the three-phase inverter output remained stable, with no signs of oscillation. The implemented design effectively suppresses peak drive voltage levels, demonstrating optimized gate drive circuitry performance. The output waveform of three-phase inverter proves the effectiveness of the control strategy.