We present a method for real-time parallel compression of neural signals, achieving a 400:1 compression ratio while preserving temporal and spike count accuracy. This approach utilizes a monostable multivibrator-based circuit model that compresses neural activity within a 20 ms time frame into a single voltage point, effectively representing the entire signal before digitization. Using patch-clamp recordings from pyramidal cells in layer 4 of the visual cortex in young guinea pigs, this technique compresses a 499.95 ms dataset sampled at 20 kHz into 25 data points. Decompressed signals retained all 10 action potentials (APs) from the original dataset, with a worst-case timing deviation of 4.15 ms. Spike timing accuracy was further validated using the Victor-Purpura distance, resulting in a distance of 2.217 and a normalized value of 0.2217, respectively. By significantly reducing the data size required for transmission, this method demonstrates the potential to support multi-channel microelectrode arrays (MEAs) in wireless systems with limited throughput, enabling scalable and efficient neural recording technologies.

Diffusion models have revolutionized generative modeling by offering a robust probabilistic framework capable of producing highquality, diverse outputs across various domains. Rooted in stochastic processes and equipped with rigorous mathematical foundations, these models have demonstrated exceptional performance in tasks ranging from image and video synthesis to natural language processing, healthcare, and scientific research. However, their iterative nature imposes substantial computational demands, hindering their scalability and real-time applicability. This paper provides a comprehensive overview of diffusion models, focusing on their theoretical principles, recent advancements in efficiency, and real-world applications. Key innovations such as optimized noise scheduling, streamlined architectures, accelerated sampling strategies, and hardware-aware optimizations have significantly mitigated computational overhead while maintaining performance. Furthermore, we explore the transformative impact of these models across disciplines and discuss emerging challenges, including domain-specific adaptation, ethical considerations, and scalability constraints. By synthesizing state-of-the-art techniques and applications, this work aims to equip researchers and practitioners with a nuanced understanding of diffusion models and inspire future developments. The findings underscore the potential of these models to redefine generative AI, driving innovation and enabling practical solutions across a wide array of fields.

Voltage Source Converters (VSCs) operating as active rectifiers inherently lack currentlimiting capabilities for faults occurring on the DC side due to the presence of freewheeling diodes in IGBT or MOSFET structures. This limitation leads to uncontrolled fault currents, flowing from the AC to the DC network, that can jeopardize the safety of power electronic components. Additionally, the challenge is compounded in DC networks, where DC circuit breakers must interrupt high fault currents, fed by the AC side, without the benefit of current zero crossings. To address these issues, this paper presents a novel topology that integrates a classical buck DC/DC converter into a VSC to regulate the currents of faults occurring in DC network, thereby improving the protection of the converter and aiding DC circuit breakers in interrupting the fault. The advantage of the proposed topology lies in that under normal operating conditions, the buck converter is totally bypassed, thus improving the efficiency of the topology. When a fault is detected within the DC microgrid, the buck converter is connected in series with the VSC to control the current flowing from the AC to DC network. Simulation results using MATLAB/Simulink validate the effectiveness of the proposed topology in completely controlling the current at the DC side of converter, thus demonstrating significant improvements in fault management, system reliability, and converter protection.

Potential for subsynchronous interactions (SSI) involving inverter-based resources is often screened for using small-signal impedance scans. However, these impedance scans are often slow to compute and it is often desired to study a wide range of operating conditions. This paper presents modeling practices and a new scanning approach for performing impedance-based SSI screening studies that can reduce the required computation time by orders of magnitude with negligible effect on accuracy. Relevant principles are illustrated using a type3 wind turbine model, using two different reactive power control strategies that have significantly different SSI behavior. The discussion of modeling practices includes per-unitization, mechanical dynamics, timestep selection, PLL design, and negative-sequence behavior and covers both time-saving simplifications and pitfalls or unsuitable simplifications to avoid. The new scanning approach involves identifying sets of input frequencies that do not interfere with one another’s outputs and using crest factor minimization to develop multisine test signals.

Cyberbullying has become a major problem nowadays especially on social networking sites. Training a system to detect the negative comments to remove them from the sites can solve this problem. Analyzing the attitude of the comments or posts on social media using word embeddings can identify it. word embeddings with a simple dimensionality reduction operation called Hellinger PCA (Principal Component Analysis) is used in this paper. The dataset is created by collecting comments from popular pages of Facebook using Graph API. On detecting cyberbullying and negative comments, accuracy of 72% is obtained for our corpus of 10500 with this approach. The performance is improving for larger corpus.

In the face of current modular robots that rely on manual coding to generate deformation instructions and are limited by binary instructions, this research proposes a Cognitive Modular Control (CMC) architecture inspired by ACT-R theory. Build simulation tool ReoForm to conduct experiments through simulation-based experimental design. The data results show that each module of the architecture demonstrates feasibility and universality in terms of time-to-completion, energy consumption, manipulability index and robustness indicators.

To improve the cognitive autonomy of humanoid robots, this research proposes a multi-scenario reasoning architecture to solve the technical shortcomings of multi-modal understanding in this field. It draws on simulation based experimental design that adopts multi-modal synthesis (visual, auditory, tactile) and builds a simulator "Mahā" to perform the experiment. The findings demonstrate the feasibility of this architecture in multimodal data.

The radiation pattern and efficiency of a leaky-wave antenna is dependent on its aperture field distribution. The required field magnitude is achieved by modulating the leakage factor along the antenna, for which an approximate formula derived in the last century continues to be widely used nowadays. This expression assumes the field transverse profile inside the guiding structure does not change along its length. This contribution revisits this well-known formula in order to obtain it from a rigorous approach assuming a leaky rectangular waveguide. Thus, the origin of the limitation of the maximum leakage factor value is explained, and its influence is quantified by establishing a numerical criterion. Finally, some practical examples are discussed, showing how the approximate leakage factor differs from the exact numerically-computed one in limit cases, and how the radiation patterns are affected.

Deep learning methods have been applied in many domains with remarkable performance for the past decade, including in medical imaging. There is currently a growing interest in the research community for deep learning in Wireless Capsule Endoscopy (WCE). However, the availability of quality annotated dataset, and robust performing methods are still challenging in WCE. To fill this gap, MISAHUB provided an annotated dataset for bleeding in WCE, and launched the "AutoWCEBleedGen challenge Version V1" which aims at investigating deep learning methods for classification and detection of bleeding in WCE. As a response to the challenge, our work presents a method to classify bleeding and non-bleeding frames which is based on ResNet50 using cross validation and data augmentation techniques. Furthermore, we used YOLOv5 model to detect and draw the bounding boxes of the bleeding pixels in the bleeding images. Our solution achieves respectively 99.57% accuracy, 99.58% F1-score, and 99.53% Recall for the classification task on the validation set. For the detection task we achieve 70% AP, 68.20% mAP and 45% IoU. Our code is available at https://github.com/agossouema2011/ WCEBleedGenChallenge_Colorlab_Team.

In Indiana, roadside mowing operations span over 11,000 miles and are performed by contracted crews using tractors and hand-held trimmers, exposing workers to significant safety risks, including reported casualties. Autonomous mowers offer a promising solution to enhance safety, but large-scale deployment requires rigorous testing to ensure reliability. This study addresses this need by developing digital-twin roadside environments to evaluate autonomous mowing systems. The method generates realistic roadways that accurately replicate real-world conditions, achieving single-float rounding planar accuracy over tens of meters, enabling risk-free and comprehensive testing. Leveraging remote sensing data, model libraries, and a georeferenced data mapping tool, the approach overcomes challenges in scalability and fidelity, even in locations with low-density source data. Digital roadway construction accelerates testing, enhances the realism of virtual deployments, and supports the development of autonomous mowing technologies. Furthermore, this approach has broader applications for generating test environments for various roadway-related systems.

Federated Learning (FL) is a distributed machine learning paradigm that enables collaborative model training across decentralized devices while preserving data privacy. It addresses critical challenges in privacy, scalability, and data ownership, making it a promising approach for applications in healthcare, IoT, and finance. However, practical implementation of FL faces several efficiency bottlenecks, including communication overhead, system and data heterogeneity, and security vulnerabilities. This paper provides a comprehensive survey of state-of-the-art techniques aimed at enhancing the efficiency of FL. Key methods such as model compression, including pruning, quantization, and tensor decomposition, are explored to address communication constraints. Strategies to mitigate data and system heterogeneity, including personalized FL and resource-aware training, are discussed alongside advancements in privacy-preserving mechanisms like differential privacy and secure aggregation. We also examine scalability solutions, including hierarchical and decentralized FL, to enable large-scale deployment. The survey highlights open challenges and emerging opportunities in FL, offering insights into future research directions for building efficient and robust federated systems.

The rapid evolution of software systems, particularly multi-component architectures, has intensified the need for efficient and context-aware testing solutions. Traditional methods of test case generation often struggle to account for the dynamic interdependencies between system components, which can lead to incomplete or redundant test coverage. Leveraging Large Language Models (LLMs), such as GPT-based architectures, presents a novel approach to generating context-aware test cases that better reflect real-world usage scenarios and system interactions. By fine-tuning LLMs on domain-specific knowledge and historical test data, we can enhance their ability to generate meaningful, diverse, and context-sensitive test cases. This abstract discusses a methodology for utilizing LLMs to automatically produce test cases tailored to multi-component systems, accounting for both individual component behaviors and their interactions. We highlight the benefits of this approach, such as improved test coverage, reduced manual effort, and faster development cycles, while also addressing challenges like ensuring test case quality and managing model biases. Ultimately, this research aims to establish a robust framework for integrating LLMs into the software testing lifecycle, with the goal of enhancing software reliability and reducing time-to-market for complex systems.

This paper explores the integration of WordPress and associated technologies in developing a custom web-based HR system, focused on recruitment, staff deployment, and document management. Using Delorecarescreeningapp.com as a case study, we highlight WordPress's role in delivering scalable, secure, and user-friendly HR solutions. By incorporating middleware, API integrations, and automation tools, WordPress proves to be an effective solution for streamlining HR operations and optimizing the employee lifecycle from recruitment to document management.Purpose of the Research: This research aims to introduce a web-based solution for optimizing HR processes-focusing on recruitment, staff deployment, and document management-using modern technologies. By leveraging WordPress and middleware integrations, the system automates HR workflows, enhancing efficiency, scalability, and compliance in organizations with large, dynamic workforces.

As minimum area SRAM bit-cells are obtained when using cell ratio and pull-up ratio of 1, we analyze the possibility of decreasing the cell ratio from the conventional values comprised between 1.5-2.5 to 1. The impact of this option on area, power, performance and stability is analyzed showing that the most affected parameter is read stability, although this impact can be overcome using some of the read assist circuits proposed in the literature. The main benefits are layout regularity enhancement, with its consequent higher tolerance to variability, cell area reduction by 25% (with respect to a cell having a cell ratio of 2), leakage current improvement by a 35%, as well as energy dissipation reduction and a soft error rate per bit improvement of around 30%.

The spread of fake news and misinformation has become a global challenge, undermining public trust, causing political polarization, and facilitating the dissemination of harmful ideologies. This study explores the use of advanced AI models, specifically transformers such as BERT and GPT, for the automatic detection of fake news. Leveraging natural language processing (NLP) techniques like Named Entity Recognition (NER), Sentiment Analysis, and Topic Modeling, we aim to identify patterns unique to misinformation. Our model demonstrates high accuracy in experimental trials on benchmark datasets, highlighting the potential of AI to combat disinformation and improve media literacy.

Ransomware, a malicious software type leveraging cryptography, has evolved into a significant global cybersecurity threat. This paper explores the use of symmetric, asymmetric, and hybrid encryption techniques in ransomware attacks and its impacts. Symmetric cryptography is favored for its speed, while asymmetric systems provide enhanced key security. Hybrid systems, combining the strengths of both, offer the most effective encryption method for ransomware operations. Through analysis of notable ransomware like Jigsaw, Archiveus, and WannaCry, this paper highlights their cryptographic methods, impact, and operational strategies. The study underscores the importance of robust countermeasures, including updated cybersecurity protocols and awareness, to mitigate ransomware risks. Additionally, the paper evaluates the role of evolution in ransomware attacks, including the importance of cryptocurrencies enabling anonymous ransom payments. The findings emphasize the need for comprehensive strategies that incorporate both technological advancements and human-centered approaches to cybersecurity. By addressing these challenges, individuals and organizations can reduce the frequency and severity of ransomware attacks.

This paper explores the complex behavior of advanced persistent threat (APT) attacks, characterized by a dual threat: the sophisticated manipulation of adversarial disturbance inputs and the exacerbation of system vulnerabilities due to environmental uncertainties. To address these security concerns in large-scale multi-agent industrial cyber-physical systems (CPSs), we develop a decentralized control framework using mean-field game (MFG) theory with multiplicative noise in the dynamics. Our approach effectively tackles the scalability challenges inherent in large-scale environments while countering both intelligent adversarial disturbances and operational uncertainties. By designing resilient and robust decentralized controllers, we ensure system stability and convergence, even under worst-case disturbance inputs. We prove that the mean-field approximation accurately captures the system's collective behavior, and the proposed decentralized controllers achieve ϵ-Nash equilibrium. Numerical experiments, inspired by the Ukraine power grid attack, demonstrate the effectiveness of the proposed control strategy.

Ferroresonant systems involving power system transformers and stray capacitances is relatively rare but produces dangerous and damaging overvoltages. The proliferation of distributed energy resources (DERs) has altered the way distribution systems are operated and introduced new ferroresonance risks, as supported by both first principles and real-world ferroresonance events. DERs that are inverter-based are particularly in need of study, as the role in and response of inverters to ferroresonance events differs drastically from that of conventional generators. In this paper, ferroresonances involving the distribution transformer used to connect the DER to the grid are studied using electromagnetic transient analysis. Ferroresonance scenarios involving single-pole open (SPO) events, fault-induced overvoltages, and energization from ungrounded sources are studied. In each, the role of the inverter is assessed in detail, providing new insights into how ferroresonance near inverter-based DERs should be modeled, what the consequences of ferresonance are to inverter operation, and how the use of inverters instead of traditional generators affects the overall risk of ferroresonance.