We present a hierarchical framework aimed at decentralizing the distribution systems market operations using localized peer-to-peer energy markets. A hierarchically designed decision-making algorithm approaches the power systems market operations from a bottom-up perspective. The three layers of the hierarchical framework operate in orchestration to enable prosumers (the grass-root ac- tors) to maximize their revenues - hence, a prosumer-centric framework. The design of the framework incorporates existing smart grid technologies (Virtual Power Plants, Microgrids, Distributed Energy Resources) and redefines their functional objectives to align them with the decentralization paradigm focused on empowering the bottom-up grid operations approach. On one hand, the framework is enabling prosumers with simultaneous access to the buy-sell choices that help them maximize their cost savings while ensuring their consumption patterns and preferences are not being traded off as a result of top-down operational decisions. On the other hand, it is designed to operate in harmony with the existing top-down grid operations mechanisms - thereby reducing the potential friction in its adaptation. This marriage of the top-down and bottom-up operational approaches is facilitated through meticulous orchestration of operational timescales. The frameworkâ\euro™s novel design also incorporates scalability and interoperability considerations, thereby tackling the challenge of decentralization holistically.

In the past 20 years, improvements in night-time light (NTL) remote sensing have spurred a resurgence of interest in the mapping of human economic activity. Nevertheless, the full potential of NTL data for urban research is constrained by a relatively coarse spatial resolution and the blooming effect. Downscaling NTL data is a potential solution, aiming to obtain fine-resolution nocturnal data with high accuracy. Most existing remotely sensed image fusion techniques were developed for optical remote sensing images taken during the day. When NTL images are compared to optical images, they exhibit a greater quantity of dark (low value) pixels, higher levels of background noise, and a more obvious blooming effect. In this paper, we proposed a spatially non-stationary, geostatistical-based downscaling technique (random forest area-to-point Kriging) to downscale NTL data while accounting explicitly for the point spread function, thus, dealing with the blooming effect specific to NTL data. We compared several image fusion algorithms for downscaling while reducing the blooming effect. Numerical experiments on two megacities showed that downscaling was improved both numerically and visually by taking the PSF into consideration. During the RF regression, the R 2 increased and the RMSE decreased for both study regions, when accounting for the PSF. For the ATPK-based residual part, considering the PSF led to increased accuracy of prediction. The suggested methodology has the potential to increase the detail and accuracy of the NTL data available for modeling socioeconomic phenomena at the city scale, with wide potential for application in future socioeconomic research.

This paper proposes a new approach for convergence analysis of continuous action iterative dilemma (CAID) to reach a stable consensus outcome within the prescribed time. Unlike usual game theory, where players can only choose between two options, i.e., cooperation or defection, CAID lets players pick from varying options and make more nuanced decisions. In the proposed strategy, a prescribed function, which is tunable by the user, adapts the learning rate of the player's strategy. This new method guarantees that players will eventually agree on a single strategy, regardless of where they started initially, and it achieves this agreement in a predefined time set by the user. The Lyapunov analysis guarantees the convergence of strategy to a consensus within a prescribed time. The simulation results of the proposed scheme with two evolutionary game examples demonstrate faster convergence and fewer iterations compared to the state-of-the-art method.

The rapid escalation of ransomware attacks has become a significant threat to individuals, organizations, and critical infrastructures, resulting in substantial financial losses and operational disruptions. A novel approach was developed through the application of machine learning techniques combined with file entropy analysis to provide an early detection system capable of identifying ransomware infections before critical files are encrypted. The method leverages entropy as a statistical measure of file randomness, allowing for the detection of encrypted or disordered files that are indicative of ransomware activity. Machine learning models, such as Random Forests and Support Vector Machines, were trained to classify files based on entropy and additional metadata, demonstrating high accuracy, precision, and recall across a diverse dataset of benign and ransomware-infected files. The results indicated that entropybased detection outperforms traditional signature-based methods, particularly in identifying zero-day ransomware variants. Future work includes the development of real-time monitoring systems for continuous entropy assessment, as well as exploring the use of deep learning to enhance detection performance. Integrating these advancements into practical, real-time environments holds significant promise for mitigating ransomware attacks before substantial damage occurs.

For many emerging digital economies in the world, the decentralization of digital services offers a way to incentivize the development and operations of the Web3 infrastructures for these services by the private sector while avoiding the concentration of power in the hands of a few. Indonesia, as one of the fastest-growing economies in the global south, can benefit from utilizing Web3 infrastructures as a form of the Digital Public Infrastructure (DPI) concept. Key to implementing the DPI is the standardization of open protocols and APIs in the architecture, enabling various market players to develop basic services with a high degree of interoperability. The tokenization of real-world assets built on decentralized Web3 infrastructures could be a promising new sub-sector of Indonesia's economy, enabling local communities to actively participate in the coding, development, and running of the nodes that constitute these new Web3 infrastructures.

Frame coalescing is a well-established technique which manages the low power idle (LPI) mode supported by energy efficient Ethernet (EEE) interfaces. Generally, this technique enables EEE interfaces to remain in the LPI mode for a certain amount of time upon the arrival of the first frame (timer-based coalescing), or until a predefined amount of traffic accumulates in the transmission buffer (size-based coalescing). In this paper, we propose a novel open-loop dynamic coalescing technique that is based on model predictive control (MPC) and queuing theory. In contrast with the conventional timer-based coalescing, the proposed method enables the update of the timer parameter repeatedly throughout the duration of the LPI mode of a single coalescing cycle by taking into account the arrival instant and size of the frames waiting in the buffer. Two different methods, namely MPC-mean and MPC-tail, are proposed which attempt to minimize the energy consumption of the Ethernet link under constraints on mean and tail of the queue waiting time, respectively. The effectiveness of the proposed dynamic MPC-based coalescing algorithms are validated using simulations with synthetic and actual traffic traces.

Neural Network (NN) technology is revolutionizing the modeling paradigm in the engineering arena, enhancing simultaneously model precision and process speed. Nevertheless, the deficiency of physical interpretability in current NN models causes an unattainable demand for the quantity of training data, especially when dealing with high-dimension physical behaviours. In this work, a Hierarchical Physics-Informed Neural Network (HPINN) framework is proposed to address the 3D magnetic modeling issue of Medium Frequency Transformers (MFTs). By establishing the knowledge transfer channel between the cross section and the equivalent space, the inferior 2D model parameter search space, which is restricted by Dowell’s equation and accessible 2D magnetic data information, can be mapped into the superior 3D search space. Accordingly, with limited 3D training data, the performance of the HPINN framework is still guaranteed around the optimal state. The structural embedding of physics knowledge reduces the model's reliance on 3D training data, thereby relieving the strict computing requirement at the data generation stage. Finally, according to the training results and experiment verifications, with only one-third computational burden used in the classic NN model, the average error of this proposed HPINN framework is reduced to 1%, and the maximum error does not exceed 10%.

1. Introduction:The recent improvement of machine learning (ML) has impacted the development of various fields and enabled computers to learn complex data patterns and make predictions without prior instructions. Especially in the natural language processing (NLP) application. A crucial phase in NLP is the representation of words [23]. It significantly impacts the performance of ML algorithms in tasks such as Language translation, Sentiment Analysis, and Text Classification. While deep learning techniques have led to remarkable developments in NLP, traditional models still hold relevance in specific instances of sentiment analysis, where they provide more interpretable results. This interpretability becomes especially valuable in domains where explainability is crucial. Moreover, in scenarios with limited computational resources or time constraints, traditional models offer simplicity and efficiency, making them a preferred choice. Additionally, when only a limited amount of labeled data is available, traditional models can still achieve reasonable performance, making them suitable for situations where data scarcity is a concernThe first introduced application of word representation is a one-hot encoded vector, which represents the word as a sparse vector. This representation, however, lacks essential features of the data [7]. The one-hot encoding technique is unable to extract the semantic similarity between words. Additionally, the sparse vector representation is heavy to compute and requires more processing resources with massive data.More recent word representation has resolved the one-hot encoding representation limitation. These techniques have successfully demonstrated higher performance in various tasks. The new word embeddings extract the semantic similarity of words using a dense low dimensional space vector which requires minor resources to compute and displays a better word representation.The field of text-based classification on ML has seen significant advancements, with a growing technique, newly introduced model, and enhanced word representation to attempt to solve various challenges. Especially in sentiment analysis which extracts abstract information from the data. However, there is further research in understanding machine learning classifiers and word representation to optimize the model’s performance. Which we address in the following research questions:Q1: Which word representations yield the highest performance in sentiment analysis when used with traditional machine learning and deep learning models?Q2: Is there a word representation technique that can output comparable performance in both traditional machine learning models and deep learning models for sentiment analysis?

Head injuries pose a significant risk in industries such as construction and manufacturing, making helmet use essential for safety. This paper presents a comparative analysis of six YOLO (You Only Look Once) models-YOLOv5mu, YOLOv6m, YOLOv7, YOLOv8m, YOLOv9 (m, GELAN-c, GELAN-e), and YOLOv10m-for helmet detection using a public dataset. The models are evaluated for mean Average Precision (mAP), frames per second (FPS), training time, and model size. YOLOv9 (GELAN-e) achieves the highest accuracy with 96.17% mAP at 0.1% confidence and 82.71% at 70% confidence. YOLOv8m is the fastest, achieving 45.37 FPS, while YOLOv10m has the highest mAP (85.10%) at 70% confidence and is the most compact (33.5 MB). YOLOv9 (GELAN-c) offers the shortest training time (1.32 hours), while YOLOv9 strikes the best balance between accuracy, speed, and size. These findings suggest YOLOv9 as an excellent choice for applications requiring both accuracy and efficiency, while YOLOv8m is optimal for real-time scenarios.

This paper explores the problems caused by spam messages on the internet. It talks about different types of spam, like email spam and fake advertisements on social media. The paper also looks at how spam can make things difficult for people, like draining resources and tricking them into clicking on harmful links. The paper suggests fighting against spam by enabling filters in emails, using third-party spam filtering, avoiding clicking suspicious links, reporting, and blocking and avoiding using personal or professional emails. Lastly, the paper discusses what might happen with spam in the future and the challenges we may face. Overall, the paper aims to help and understand then find ways to deal with it for a safer and smoother experience.

Pulsatile physiological signals, characterized by periodic waveform traits, are essential for patient monitoring. However, artifacts considerably compromise signal fidelity, thereby rendering the signal cleansing process essential. Although previous studies have focused on algorithmic advancements, domainspecific factors in this field are yet to be comprehensively understood. Therefore, this study conducts an empirical investigation into four critical signal cleansing aspects in the medical field. First, we verified the performance enhancement of incorporating domain knowledge into cleansing techniques. Second, we emulated real-time cleansing requirements from clinical environments. Third, we explored the similarities and differences among three evaluation approaches: quantitative assessment of the cleansing performance, qualitative assessment of the same by clinical experts, and a cleansing effect assessment on downstream tasks. Finally, we evaluated whether the willingness to accept these techniques varied according to the knowledge incorporation mode and professional groups comprising clinicians and engineers. The results revealed that domain knowledge incorporation enhances cleansing performance, visually appealing cleansing results can mislead experts, and the willingness to accept cleansing techniques varies depending on the knowledge incorporation method, professional group, and experience level.

This paper presents a novel framework for real-time object detection in Wireless Mesh Networks (WMNs) using AI driven drones. The system integrates YOLOv5 for object detection, a hybrid greedy and reinforcement learning-based routing algorithm for efficient data communication, and a scalable multihop mesh topology. MATLAB-based simulations evaluate energy consumption, detection accuracy, and network connectivity, showcasing significant performance improvements. The results demonstrate the system's ability to detect objects accurately, maintain robust connectivity, and optimize energy consumption, even in dynamic environments. Future work includes expanding the system to integrate 5G and satellite networks and conducting field tests.

We propose an integrated behavior and motion planning framework for the automated lane-merging problem. The behavior planner combines search-based planning with game theory to model the interaction between vehicles and select multivehicle trajectories. Inspired by human drivers, we model the lane-merging problem as a gap selection process. To overcome the challenge of multi-modal driving behavior exhibited by the surrounding vehicles, we formulate the trajectory selection as a matrix game and compute some equilibrium solutions. In practice, however, the surrounding vehicles might deviate from the computed equilibrium trajectories. Thus, we introduce a branch model predictive control (BMPC) framework to account for the uncertain behavior modes of the surrounding vehicles. A tailored numerical solver is developed to enhance computational efficiency by leveraging the tree structure inherent in BMPC. Finally, we validate our proposed integrated planner using real traffic data and demonstrate its effectiveness in handling interactions in dense traffic scenarios.

Federated Learning (FL) in load forecasting improves predictive accuracy by leveraging data from distributed load networks while preserving data privacy. However, the heterogeneous nature of smart grid load forecasting poses significant challenges to conventional FL methods, which are also unsuitable for resource-constrained devices. To address these issues, we propose a novel Lightweight Single Layer Aggregation (LSLA) framework tailored for smart grid networks. The LSLA framework mitigates the issue of data heterogeneity in load forecasting by emphasizing local learning and partially using updates from local devices for model aggregation. Additionally, the framework is optimized for resource-constrained devices by introducing a stopping criterion during model training and weight quantization. Our results show that after quantization, there is an acceptable degradation of 0.01% in Mean Absolute Error (MAE). Compared to traditional FL, up to a 1000fold reduction in communication overhead is achieved with LSLA. Moreover, with an 8-bit fixed-point data representation of neural network weights, a 75% reduction in storage/memory requirement is achieved, and computational cost is reduced due to the replacement of complex floating-point computational units with simpler fixed-point counterparts. By addressing data heterogeneity and minimizing data storage, computation, and communication overheads, our novel framework is well-suited for resource-constrained devices in smart grid networks.

This paper presents an algorithm for a team of heterogeneous mobile robots to estimate and adaptively sample a stationary, isotropic, Gaussian process. An estimation framework is proposed to assimilate measurements from robots with differing sensing capabilities (i.e., measurement noise variance). To improve computational efficiency of the Gaussian process regression, the survey area is divided into regions that each use a common semivariogram matrix constructed with a truncated measurement set determined by an adaptive selector. As the mission proceeds, a Voronoi-based algorithm periodically partitions a density function-representing time-varying sampling priority-to identify high-value sampling locations. The path for each robot is modeled as an mechanical system: a sequence of masses (waypoints) are interconnected by springs and dampers and pulled towards Voronoi cell centroids. At each path planning cycle the robots are iteratively simulated with heterogeneous dynamics (e.g., speed, turn radius) following their respective waypoint paths as stiffness/damping parameters are adjusted to satisfy mission time constraints. Numerical simulations show that the proposed approach reduces mapping error when compared to non-adaptive lawnmower coverage of a survey area. The algorithm is demonstrated experimentally using two cooperating autonomous surface vessels to map the bathymetry in a section of Lake Norman near Charlotte, NC.

The inpainting problem is addressed in this work through a very simplified version of the approach based on low-dimensional manifold model (LDMM), in which the actual working principle of the LDMM is put into evidence, namely, the simulated diffusion of image pixels that takes place in a manifold from which patches are drawn to form a given image. The simplicity of this principle is translated into a straightforward algorithm that borrows ideas from the Locally Linear Embedding (LLE) method. The equivalence between this much simpler algorithm and the LDMM is illustrated through visual inspection and experimental measurements of peak signal-to-noise ratios. Additionally, a (U-shaped) multi-scale use of the proposed algorithm is presented as a significantly better initializer for missing pixels, thus reducing the number of algorithmic iterations.

Deep brain stimulation of the subthalamic nucleus (STN-DBS) is an established treatment for motor impairment due to Parkinson's disease (PD) progression. While treated subjects mostly experience significant amelioration of symptoms, some still report adverse effects. In particular, changes in gait patterns due to the electrical stimulation have shown mixed results across studies, with overall gait velocity improvement described as the core positive outcome. This retrospective study investigates changes in the gait parameters of 50 PD patients before and 6 months after STN-DBS, by exploiting a purely data-driven approach. First, unsupervised learning identifies clusters of subjects with similar variations in the gait parameters after STN-DBS. This analysis highlights two dominant clusters (Silhouette score: 0.45, Dunn index: 0.18), with one of them associated to a worsening in walking. Then, supervised machine learning models (i.e., Support Vector Machine and Ensemble Boosting models) are trained using pre-surgery gait parameters, clinical scores, and demographic information to predict the two gait change clusters. In a Leave-One-Subject-Out validation, the best model achieves balanced accuracy over 82%, denoting moderate predictability of both clusters. Moreover, feature importance analysis reveals the variability in the step width and in the step length asymmetry during the preoperative gait test as promising biomarkers to predict gait response to STN-DBS.

Existing end-to-end congestion control algorithms, in Transmission Control Protocol (TCP), use packet loss and queueing delay for congestion detection, and use static control laws to adjust the sending rate and to control the congestion. This approach presupposes that the network, and its interaction with the congestion control mechanism, is static or quasi-static. In practice, the state of the network continuously changes over time, resulting in suboptimal performance of existing algorithms. The article proposes an optimal control approach that solves a dynamical function derived using queuing theory, based on Little's law. The proposed control depends on the rates of change of the average occupancy (DataInFlight) and average response time (round-trip-time), respectively, with respect to the average arrival rate (sending rate). This results in damped harmonic oscillating trajectory. In terms of the average goodput, the proposed mechanism performs better than widely used algorithms, such as TcpCubic and TcpBbr. Similar to TcpBbr, the scheme satisfies Kleinrock's optimality condition for an optimal queuing system, by keeping the average occupancy close to one bandwidth-delay-product and the queueing delay low.