Search

In recent years, the use of renewable energy sources by many power grid companies worldwide has increased significantly. The trend towards the use of renewable energy sources is mainly due to environmental issues and rising fuel prices associated with conventional electricity generation. Distributed generation units are power generation plants that are very important for the grid architecture of today’s power system. The benefit of adding these Distributed Generation (DG) units is to increase the power supply to the grid. However, the installation of DG units can cause a negative impact if not properly allocated and/or sized. Therefore, there is a need for their optimal sizing and allocation to avoid situations such as voltage instability and high investment cost. In this paper, four heuristic based algorithms, namely Particle Swarm Optimization (PSO) algorithm, Whale Optimization Algorithm (WOA), Dolphin Echolocation Optimization (DEO), and Slime Mould Algorithm (SMA) are applied to solve the optimal placement and sizing of DG units in distribution network planning. Three cases were used to address the network problems, which are represented by adding photovoltaic cells and wind turbines individually. In the last case, both were used, and the feasibility of algorithms was confirmed for two systems, IEEE 33-bus and 69-bus test systems. The comparison results showed that the SMA algorithm produces good solutions. In general, the SMA algorithm was able to reduce the two system losses. The reduction of real power losses in SMA, taking into account the technical and economic constraints in the IEEE 33 system, is reduced to a minimum of 66.31%, 67.3%, and 81.1%, While in 69 bus, reduced to a minimum of 90.7%,91%,and 97.30% for three cases respectively, as well as improving the voltage profile, thus obtaining a more efficient system.

Given the rising demand for low-power sensing, integrating additional devices into an existing wireless infrastructure calls for innovative energy- and spectrum-efficient wireless connectivity strategies. In this respect, wireless-powered or energy-harvesting symbiotic radio (EHSR) is gaining attention for establishing the secondary relationship with the primary wireless systems in terms of RF EH and opportunistically sharing the spectrum or schedule. In this paper, assuming the commensalistic relationship with the primary system, we consider the energy-efficient optimization of such an EHSR by intelligently making EH and transmission decisions under the inherent nonlinearity of the EH circuitry and dynamics of pre-scheduled primary devices. We present a state-of-the-art deep reinforcement learning (DRL)-engineered, energy-efficient transmission strategy, which intelligently orchestrates EHSR’s uplink transmissions, leveraging the cognitive radio-inspired non-orthogonal multiple access (CR-NOMA) scheme. We first formulate the energy efficiency (EE) optimization metric for EHSR considering the nonlinear EH model, and then we decompose the inherently complex, non-convex problem into two optimization layers. The strategy first derives the optimal transmit power and time-sharing coefficient parameters, using convex optimization. Subsequently, these inferred parameters are substituted in the subsequent layer, where the optimization problem with continuous action space is addressed via a DRL framework, named modified deep deterministic policy gradient (MDDPG). Simulation results reveal that, compared to the baseline DDPG algorithm, our proposed solution provides a 6% EE gain with the linear EH model and approximately a 7% EE gain with the non-linear EH model.

The end of Moore’s Law and Dennard Scaling has slowed processor improvements in the past decade. While multi-core processors have improved performance, they are limited by the application’s level of parallelism, as prescribed by Amdahl’s Law. This has led to the emergence of domain-specific processors that specialize in a narrow range of functions. Smart Network Interface Cards (SmartNICs) can be seen as a revolutionary technology that combines heterogeneous domain-specific processors and general-purpose cores to offload infrastructure tasks. Despite the impressive advantages of SmartNICs and their importance in modern networks, the literature has been missing a comprehensive survey. To this end, this paper provides a background encompassing an overview of the evolution of NICs from basic to SmartNICs, describing their architectures, development environments, and advantages over legacy NICs. The paper then presents a comprehensive taxonomy of applications offloaded to SmartNICs, covering network, security, storage, and compute functions. Challenges associated with SmartNIC development and deployment are discussed, along with current initiatives and open research issues.

Entropy is a general term for measuring or determining the uncertainty of a system. Because of the advantageous concept of entropy, it has gained interest and attraction from the standpoint of graph theory, as it has been used as applied mathematical chemistry. The first application of graph-based entropy was to determine the structural information of chemical networks or graphs. Networks, like mathematical chemistry, are constructed in the form of nodes and edges to form a network or system graph. Carbon nanotubes, on the other hand, have a long history of use in memory devices and tissue engineering. Carbon nanotubes, a productive chemical structure, are studied here by affixing one-end and both-end caps to them. The entropy of three different carbon nanotube topologies is calculated. Some comparative work is provided in addition to the analytical study of the entropy measure of armchair carbon nanotubes.

The use of optoelectronic technology in Optical Transposition Interconnection Systems (OTIS) offers an effective solution to the ongoing problem of storing and sending data with comprehensive information. This is due to the reduced power requirements and broad bandwidth capabilities of optoelectronic systems, which make them well-suited for this task. The integration of radio communication and electrical technology has transformed OTIS into a highly valued network, enhancing the efficiency of existing optoelectronic computers. OTIS is characterized by the swapped network that is formed with the help of path graph $\mathfrak {P}_{m}$ and denoted as $B(\mathfrak {P}_{m})$ . This research work focused on certain topological invariants (TIs) about the number of connections between nodes of the graph $B(\mathfrak {P}_{m})$ and its largest subgraph that preserves twin nodes ( $M(B(\mathfrak {P}_{m}))$ ). First Zagreb connection index, geometric-arithmetic connection index, Randić connection index, reduced reciprocal Randić connection index, sum-connectivity connection index, first, second and third redefined Zagreb connection indices are considered in this work two types of biswapped and OTIS networks.

Recently, the thermal management of power electronic converters has gained significant attention due to the continuous trend of developing very compact power electronic converters with high power density. With the evolution of power semiconductor devices, high operating temperatures and large thermal cycles have become possible, necessitating a significant improvement in thermal system designs. Researchers have made significant efforts to develop effective thermal management systems to improve the reliability and lifetime of power electronic converters. This article intends to present a thorough review of thermal management systems employed in power electronics cooling. The applied thermal management techniques have been reviewed from the perspective of electrical parameter regulation and heat dissipation control. Regulation of electrical parameters involves active thermal control, which is a method for controlling junction temperature and thermal cycling of power semiconductor devices. The active thermal control implementation processes reviewed in this article consist of increasing overload capacity, manipulating switching and conduction losses, employing modified modulation processes, balancing thermal stress at the converter level, and controlling thermal stress at the system level. Control of heat dissipation can be achieved through direct and indirect cooling of power electronic converters with air or liquid as the coolant. The effectiveness and implementation methods of these cooling techniques, such as channel cooling, phase change material-based cooling, immersion cooling, jet impingement and spray cooling, are reviewed in this paper. Moreover, performance-enhancing ideas and challenges for these techniques are discussed. The primary objective of this review paper is to bridge the existing gap in the literature by offering a comprehensive comparison of commonly employed thermal management techniques.

In an era characterized by ever-increasing data exchange across different networks, the importance of robust network security measures cannot be overemphasized. Cryptographic algorithms, on the other hand, perform a vital role in imparting security to the sensitive information during transmission, with S-Boxes (Substitution Boxes) serving as a fundamental component of symmetric-key ciphers. This study has endeavored to engineer a novel S-Box algorithm to bolster the network security. Moreover, the core focus of the suggested algorithm lies in enhancing the confidentiality, defiance to nefarious designs of cryptanalysis savvy, and overall cryptographic strength of data encryption in network communication. Random numbers spawned by the 5D multi-wing hyperchaotic system have been employed to design the proposed box. Apart from that, Rook —a chess piece has been used in this regard. Particularly, the random walk of Rook on the $16 \times 16$ hypothetical chessboard has been capitalized to sufficiently permute the data of the potential S-Box. Through a comprehensive performance evaluation and cryptographic analyses, the proposed S-Box algorithm exhibits superior resilience against diverse threats thus fortifying network security effectively. In particular, the state of the art security parameters like bijectivity, non-linearity (NL), strict-avalanche criterion (SAC), bit-independence criterion (BIC), linear probability (LP) and differential probability (DP) have been used. This work, no doubt, contributes significantly to the ever-evolving field of network security.

Threat monitoring in cybersecurity systems is often jeopardized by alarm flooding, which frequently occurs in Security Information and Event Management (SIEM) solutions due to the unnecessary annunciation of numerous logs and event data from a variety of sources, including applications, network devices, firewall logs, and other sources. Cybersecurity operators may become alert fatigued and less able to respond to important occurrences due to the high volume of unnecessary alarms. Hence, for security operators to monitor threats effectively and consistently, alarm rationalization using formal and efficient techniques is needed. This study proposes a new framework for efficiently prioritizing cybersecurity alarms based on the integration of penetration testing findings with well-established industrial standards in alarm management. This is efficiently done by developing a Natural Language Processing (NLP) model to automatically map penetration testing findings to the Adversarial Tactics, Techniques, and Common Knowledge (MITRE ATT&CK) relationships, followed by alarm rationalization based on the industrial alarm management standard ISA 18.2. Verification of the effectiveness of the new integrated approach is demonstrated on a real system, which showed a reduction in the number of critical and high alarms by 38% while conforming to industrial alarm management standards in terms of peak alarm rates, average alarm rates, and healthy alarm priority distribution.

The rapid progression of Cloud Computing (CC) technology has ushered in innovative ecosystem concepts such as Mobile Cloud Computing (MCC). In this context, the incorporation of Unmanned Aerial Vehicles (UAVs) into these cloud ecosystems has unlocked new avenues for use cases such as delivery services, disaster response, and surveillance. However, this integration presents challenges in resource management and service selection due to the unique constraints of drones and variations in service quality. This paper proposes a Game Theory-based UAV-cloud of Service Selection Architecture (GT-SSA) to address resource management and service selection challenges. By leveraging game theory in our proposal, GT-SSA optimizes decision-making for Client Drones and Provider Drones, enhancing service selection efficiency. GT-SSA proved its resilience to scalability concerns, as evidenced in Discovery Delay, Consumption Delay, End-to-End Delay, and Energy consumption. Moreover, when GT-SSA is compared with the Game Theory approach for Cloud Services in MEC- and UAV-enabled networks (GTCS), GT-SSA outperforms GTCS in terms of Successful Execution Rate, Average Execution Time, and Energy consumption. Our research also reveals that game theory surpasses fuzzy logic in terms of service selection efficiency.

The field of animated portrait generation, driven by audio cues, has seen remarkable advancements in creating lifelike visuals. Despite these strides, current methodologies struggle with the challenge of balancing stability and naturalism. Audio-driven techniques often encounter difficulties due to the subtlety of audio signals, leading to inconsistencies. In contrast, methods that rely solely on facial landmarks, while more stable, can produce artificial-looking results due to excessive manipulation of key point data. This paper introduces RealMock, an innovative approach that harmonizes audio inputs with facial landmarks during the training phase. RealMock pioneers a dual-driven training regimen, enabling the generation of animated portraits from audio, facial landmarks, or a combination of both. This novel methodology results in a more stable and authentic animation process. Extensive testing against existing algorithms across various public datasets, as well as our proprietary dataset, has highlighted RealMock’s superiority in both quantitative metrics and qualitative assessments. The RealMock framework has broad practical implications, including revolutionizing media production with realistic animated characters and improving online education through engaging avatar-based learning experiences.

In Natural Language Processing, creating training data for question answering (QA) systems typically requires significant effort and expertise. This challenge is amplified in few-shot scenarios where only a limited number of training samples are available. This paper proposes a novel pretraining framework to enhance few-shot question answering (FSQA) capabilities. It begins with the selection of the Discrete Reasoning Over the Content of Paragraphs (DROP) dataset, designed for English reading comprehension tasks involving various reasoning types. Data preprocessing converts question-answer pairs into a predefined template, consisting of a concatenated sequence of the question, a mask token with a prefix, and the context, forming the input sequence, while the target sequence includes the question and answer. The Question-Answer Replacement and Removal (QARR) technique augments the dataset by integrating the answer into the question and selectively removing words. Various templates for question-answer pairs are introduced. Models like BART, T5, and LED are then used to evaluate the framework’s performance, undergoing further pretraining on the augmented dataset with their respective architectures and optimization objectives. The study also investigates the impact of different templates on model performance in few-shot QA tasks. Evaluated on three datasets in few-shot scenarios, the QARR-T5 method outperforms state-of-the-art FSQA techniques, achieving the highest F1 scores of 81.7% in 16-shot and 32-shot, 82.7% in 64-shot, and 84.5% in 128-shot on the SQuAD dataset. This demonstrates the framework’s effectiveness in improving models’ generalization and performance on new datasets with limited samples, advancing few-shot QA.

In this paper an analytical non-quasi-static (NQS) model for long-channel symmetric double-gate junctionless field-effect transistors (JLFETs) operating in depletion mode is proposed for the first time. The model addresses the limitations of existing DC and AC models by incorporating time-dependent current continuity equations which are essentials to predict JLFETs behavior at high frequencies. Leveraging charge-based equations, the NQS model captures the delay between current and applied potentials arising beyond the quasi-static regime. Analytical solutions for small-signal perturbations allow the calculation of key transistor small signal parameters such as the gate transadmittance. The model’s validity is tested against TCAD simulations for various device parameters, including doping concentration and channel thickness. Good agreement between the model and TCAD simulations is observed across a wide frequency range, up to highly non-static transport conditions. This work lays the foundation for a comprehensive RF model of JLFETs for high-frequency applications.

One fundamental maneuver in vehicle platooning is merging, where a new vehicle joins the platoon to benefit from its advantages. To ensure safe and efficient merging, this operation typically follows a systematic framework. This paper presents a novel merging algorithm designed for heterogeneous platoon systems utilizing a decentralized protocol on multi-lane highways. Compared to existing approaches, the proposed design adopts a synergistic strategy that combines vehicle heterogeneity management with a fast, low-communication decentralized protocol, while achieving key performance objectives such as maintaining safety, minimizing spacing errors in platoon formation, and reducing merging times. The control schemes are designed to effectively regulate and stabilize platoon movements both longitudinally and laterally, before, during, and after the merging process. The merging algorithm accounts for various scenarios involving the longitudinal position and speed of the merging vehicle relative to the existing platoon, as well as situations where the merging request may be either accepted or rejected. Simulation results demonstrate the algorithm’s performance across multiple runs, achieving a merging duration of under 30 seconds, a reduction in inter-vehicular distance by no more than 50% of the desired spacing in extreme cases, and a maximum acceleration threshold at a comfortable level for no longer than 2.5 seconds. The results show satisfactory performance in maintaining platoon safety and formation, highlighting its potential application for sustainable transportation systems.

Community detection is a very important research topic in the field of Social Network Analysis. Lots of researchers are working in this field due to its applications in various fields like medicine, social, Business, Marketing, and research. Researchers are proposing new algorithms to detect the communities having better performance as compared to the existing techniques. Initially, Newman and Girvan proposed traditional algorithms for community detection from social networks in 2004, but with the growth of social networks, Convolutional Neural Network (CNN) based algorithms are proposed by different researchers in recent years due to the inefficiency of traditional methods. After reviewing the state-of-the-art algorithms based on CNN learned that MRFasGCN is having the best performance compared to any other state-of-the-art algorithms for large data sets. In this algorithm, researchers have integrated the technique of Graph Convolutional Neural Network (GCN) with the statistical model Markov Random Field (MRF) to get better results and after implementing it on large datasets comparison is done on its results with other state-of-the-art algorithms and got to know its performance is far better than any other algorithm. While MRFasGCN is performing well on social networks and provides ground truth communities, there is a possibility available for improvement due to the sparsity problem. This paper proposes a new algorithm called Modified MRFasGCN. Two modifications are done to the existing algorithm, 1. In pre-processing, rather than passing the adjacency matrix with the normalized adjacency matrix, it will pass the reconstructed adjacency matrix and normalized reconstructed adjacency matrix, which resolves the sparsity problem, 2. GCN layer output will be fed to the MRF layer and refined results will be passed to the Adam optimizer without subtraction. Our experimental analysis shows that the modified algorithm provides better ground truth communities than the MRFasGCN and solves the problem of sparsity as passes a reconstructed adjacency matrix. In this paper, the proposed algorithm is executed on different datasets having different sizes like CORA (2708 Nodes), Flicker (80513 Nodes) and DBLP (317080 Nodes) and compared on different Community Detection metrics like accuracy, NMI, F1 Score, and execution time with other algorithms. After Experiments NMI value for MRFasGCN on DBLP data set is 0.662 while for Modified MRFasGCN it is 0.672, Modified MRFasGCN algorithm provides significant improvement of 2.9% in performance as compared to MRFasGCN. F1-Score of proposed algorithm is 0.511 on DBLP dataset which is better than MRFasGCN.

The role of inter-satellite communication (ISC) is of great significance in the contemporary era of satellite technology. In this context, visible light communication (VLC) has emerged as a highly promising alternative to conventional radio frequency (RF) communication. This is especially true for small satellites like CubeSats, where constraints related to size, weight, and power (SWaP) are of utmost importance. However, VLC systems’ limited bandwidth makes it challenging to achieve large data rates and capacity. This can be evaded with the utilization of non-orthogonal multiple access (NOMA) as a fundamental technology to augment ISC based on VLC. Such schemes have already been evaluated and published in the public domain literature. There is, nevertheless, lack of information on the efficacy of utilizing cooperative NOMA (Co-NOMA) scheme in conjunction with VLC for a constellation of satellites, with limited power resources. This research study investigates the utilization of NOMA as the access technique for inter-satellite connections when communication is done through VLC under power restrictions, with a particular emphasis on reducing the bit error rate (BER) while enhancing the average sum rate (ASR) for distant satellites through Co-NOMA. The efficacy of the proposed methodology is verified through simulation results.

Collaborative perception exploits exchanging perception data across multiple agents to enhance situational awareness. To overcome the constraints created by limited network resources, it is necessary to have a reliable collaborative perception that can sustain the trade-off between performance and bandwidth. This research proposes a method to address this issue through the development of a graph attention network (GAT) for intermediate collaborative perception. The proposed approach aims to enhance object detection accuracy by exchanging information among multiple agents, addressing issues such as sensor limitations, and occluded objects, and expanding the sensing range. Additionally, the proposed approach addresses the challenges caused by limited communication bandwidth and inconsistencies in the data shared by different agents by maintaining a balance between performance and bandwidth, fitting real-life application requirements. Our approach aggregates the intermediate features obtained from multiple neighboring agents and introduces a novel attention mechanism to selectively emphasize significant regions within the intermediate feature maps. This attention mechanism operates on both the channel and spatial levels to direct the data aggregation process. This research presents a new method for aggregating features utilizing various attention architectures. We perform quantitative and qualitative assessments of the final results of this method and compare them to other state-of-the-art collaborative perception techniques. The results of this work are evaluated using the V2XSim and OPV2V datasets, and our quantitative and qualitative experiments in multi-agent object detection show that our approach achieves a better performance than the state-of-the-art collaborative perception methods.

The Versatile Video Coding (VVC) standard, finalized in 2020 by the Joint Video Experts Team (JVET) and the Video Coding Experts Group (VCEG), marks a major advancement in video compression technology, offering a 50% efficiency improvement over its predecessor, the High-Efficiency Video Coding (HEVC) standard. A key innovation in the VVC standard is the Quad Tree with nested Multi-Type Tree (QTMTT) structure, essential for the partitioning process. However, this enhancement has led to increased coding complexity, posing challenges for real-time applications. To address this, our paper focuses on optimizing the partitioning process in the VVC encoder under the Random Access (RA) configuration. We propose a novel approach that leverages inter-prediction by integrating both coding and motion information across inter-frames to enhance coding efficiency. This solution is implemented on the Fraunhofer Versatile Video Encoder (VVenC). It utilizes a set of lightweight Light Gradient Boosting Machine (LightGBM) binary classifiers to accurately predict the optimal split mode for each Coding Unit (CU). Consequently, our approach significantly accelerates the VVenC encoding process. Experimental results show that our method reduces the runtime of the slower preset by 43.21%, with only a slight bitrate increase of 2.9%. These improvements not only significantly reduce computational complexity but also outperform several existing state-of-the-art methods.

Drawing from established models of technology acceptance, this study presents a comprehensive model that explores the influence of both Innovation-related factors and Citizen-related factors. Furthermore, it examines how trust in Internet of Things (IoT) technology moderates the relationship between innovation-related factors and intention to adopt Smart Home Technologies (SHTs). Data were collected from 401 participants and analyzed using Partial least squares (PLS). The results revealed that citizen-related factors generally have a positive and significant influence on behavioral intention to adopt SHTs, with the exception of perceived security risk and service quality. On the other hand, innovation-related factors have a significant overall effect. Moreover, trust was found to moderate the relationship between citizen-related factors (such as compatibility and complexity) and the adoption of SHTs. This suggests that improving trust in SHTs providers and focusing on technological advancements are crucial for promoting acceptance of SHTs. To conclude, this research contributes to a deeper understanding of IoT acceptance among Jordanians, particularly regarding SHTs. It provides insights for decision-makers to prioritize technological improvements and foster trust in service offerings, ultimately driving the adoption of SHTs in Jordan.

Energy Harvesting Wireless Sensor Networks (EH-WSNs) main goal is to increase efficiency in settings where Energy Harvesting (EH) is restricted by environmental resources. To solve the drawbacks of conventional Wireless Sensor Networks (WSNs) routing methods that usually ignore EH, this work presents a multi-hop clustering and renewable energy-based routing protocol designed for EH-WSNs. The suggested method performs clustering both centralized and decentralized using energy circumstances and the quantity of captured energy. The protocol operates in three phases: cluster formation, data transmission, and centralized management. To evaluate the effectiveness of the proposed approach, we analyze three distinct scenarios with different settings. The findings show that the suggested approach greatly lowers the total network energy usage while allowing a higher number of nodes to stay operational. We find that our approach outperforms AEHAC, CRBS, HUCL, and EADUC in terms of average energy levels, overall efficiency, network stability, and number of live nodes during the simulation. The results taken together show that the suggested method continuously improves network efficiency and stability in all assessed situations.

In this work, we propose a novel Gate and Drain Engineered Schottky Barrier (SB) FET (GDE-SBFET) for biosensing application with significant sensitivity improvement. Two different gate materials are employed by the proposed SB device having work functions of 3.9 eV (Al) and 4.72 eV (Cu) and electrostatically doped drain (work function of 3.9 eV). Etching out the oxide on both sides of the gate creates a nano-gap cavity that is used to detect the biomolecule. The biomolecule electrical characteristics such as charge density and dielectric constant can modify the SB on either side of the gate, which can alter the device‘s driving current. The Drain Current sensitivity ( $S_{drain}$ ) parameter are extensively analyzed at $V_{DS} = V_{GS} = 0.5$ V and a comparison between state of the art devices and conventional devices has been carried out. From the obtained results, it can be concluded that proposed device sensitivity is much superior for negatively charged and neutral biomolecules (maximum of 460 for biomolecules with negative charge, maximum of 128 for neutral biomolecule and maximum of 35 for positively charged biomolecules, at K = 12). These are the highest values of sensitivity observed for SB-FETs.

In recent years, process monitoring structures utilize big data analytics to offer a more realistic interpretation of systems. Nevertheless, managing large datasets and providing affirmative responses are common obstacles of using such monitoring frameworks. Practically, faulty conditions are less prevalent than normal situations. Thereby, coping with an imbalanced data distribution is another challenge studied here. This paper presents an innovative fault detection framework that addresses the challenges of imbalanced data distribution and big data complexities for wastewater treatment plants (WWTPs). The fault scenarios implemented for the WWTP in this research include distortions in both process and equipment, individually as well as together. For this purpose, an advanced preprocessing stage is designed, including a measurement selection method and an under-sampling algorithm. First, the measurements that convey a fair amount of information in terms of different fault scenarios are selected. Subsequently, a novel under-sampling approach is implemented to remove a number of data points from the majority class (normal conditions). The down-sampling strategy is designed in a way that trades off the amount of data elimination and information loss. The extracted features are then inserted into a typical neural network classifier for decision making. The Area Under Curve and Geometric Mean serve as effective indicators in investigating the fault detection capability of handling imbalanced big datasets. When applying the proposed fault detection framework, the average AUC and Gmean for individual faults and faults simulation scenarios are over 98% while without implementing the advanced preprocessing stage the obtained indicator values are below 79%.

Autonomous electric vehicles provide benefits to both drivers and the environment compared to conventional vehicles; however, they are burdened with an increase in potential pathways for cyber-attacks. Therefore, reliable cyber-security strategies for these vehicles must be pursued. This paper addresses this concern by implementing a threat detection strategy that utilizes an observer and a neural network. These tools monitor discrepancies between the vehicle’s lateral metrics, which are produced via sensor data, neural network output, and an observer. Previous literature focuses on physics-based analytics to create the threat decision, but here, a data based approach is utilized. The vehicle used in this study is a four-motor-drive autonomous electric vehicle that is propelled with brushless DC motors. The motors are controlled by direct torque control. In this study, three forms of cyber-attacks are implemented. These include data integrity attacks, replay attacks, and denial-of-service attacks. A performance metric is also created, which indicates the data-driven approach outperforms the physics-based approaches. All modeling and simulation were conducted in the MATLAB/Simulink environment.

The secrecy of various forms of multimedia data constitutes a significant aspect of the cybersecurity field. In this case, chaos-based image ciphers can be adapted. Chaotic image encryption has been investigated in the literature using different transformation domains, such as spatial, discrete Fourier transform, wavelet transform, etc. Although some frequency-based cryptosystems exhibit favorable confusion properties, they may suffer in terms of diffusion properties and fail in terms of some statistical characteristics, which make them vulnerable to various statistical, analytical, and differential attacks. In this regard, we adopt a dual-domain (wavelet and spatial domains) image cryptosystem and propose a novel diffusion process in the wavelet domain to address the problem of weak resilience against the aforementioned attacks. The proposed diffusion process in the wavelet domain is applied only on 1/16 of the pixels of the plain image, which makes it computationally more efficient compared to the existing wavelet domain-based works. In addition, the proposed cryptosystem solves the bell shape histogram problem associated with some frequency-based cryptosystems, which has been verified using different performance metrics in our simulation results. We also introduce a novel key-dependent chaotic variable generator to generate the required initial conditions and control parameters for the adopted enhanced chaotic map. The superiority of the proposed algorithm compared to some of the existing state-of-the-art has been verified through various performance metrics. These include different types of correlation coefficients, histogram visualization, histogram deviation, irregular deviation, mean square error, chi-square test, entropy test, and differential analysis.

This paper introduces an innovative predictive caching strategy tailored to a real-world dataset, specifically the Facebook video dataset. Making caching decisions for the dataset is challenging due to its dynamic nature, where users’ content requests vary over time without fitting into any known models. Traditional caching strategies, which often rely on a constant pool of files, do not suit this dataset as content is requested by users, and then its popularity fades over time; furthermore, the list of available content changes. We propose a two-stage predictive caching strategy. Initially, it forecasts the number of user requests using content features and historical request data, achieved through training a long short-term memory (LSTM) network. Then, we employ our proposed extended Cox proportional hazard (E-CPH) model to predict the survival probability of content. This facilitates proactive content caching. Caching new content is made possible by the timely eviction of content unlikely to be requested again. To incorporate the predicted content popularity and its life cycle into the caching decision, we introduce a partially observable Markov decision process (POMDP)-based caching strategy. Here, the survival probability of content contributes to the belief state of the associated content which leads to our believed predicted reward - a cache hit. The caching algorithm then stores the files based on their predicted believed reward taking into account both the popularity and survival probability predictions. Simulation results validate the efficacy of our proposed predictive caching method in enhancing the cache hit rate compared to conventional recurrent neural network (RNN)-based caching and policy-based caching approaches, such as least frequently used caching and its variants.

As the frequency and intensity of power line-induced wildfires increase due to climate-, human-, and infrastructure-related risk drivers, maintaining power system resilience and reducing environmental impacts become increasingly crucial. This paper presents a comprehensive methodology to assess the susceptibility, vulnerability, and risk of power line-induced wildfires for lines and nodes in an electric grid. The methodology integrates a well-established wildfire spread simulator into power flow analysis through a set of analytical steps. The proposed approach is applied to a case study using the IEEE 30-bus test system mapped on a region in the Yosemite-Ritter section of the Sierra Nevada in California. The main findings include the identification of high-risk lines and high-impact nodes and quantification of their vulnerability. These insights can inform the implementation of microgrids, virtual power plants, and distributed energy resources (DERs) to increase grid resilience and guide risk mitigation efforts such as line undergrounding, vegetation management, and maintenance procedures. The proposed methodology intends to provide an effective tool for power system planners and operators to assess the risk exposure of their grid to power line-induced wildfires, enabling them to make informed decisions for allocating capital to their resilience building and risk mitigation strategies.

We propose a framework for contact-rich path following with reinforcement learning based on a mixture of visual and tactile feedback to achieve path following on unknown environments. We employ a curriculum-based domain randomisation approach with a time-varying sampling distribution, rendering our approach is robust to parametric uncertainties in the robot-environment system. Based on evaluation in simulation for compliant path-following case studies with a random uncertain environment, and comparison with LBMPC and FDM methods, the robustness of the obtained policy over a stiffness range $10^{4}$ – $10^{9}$ N/m and friction range 0.1–1.2 is demonstrated. We extend this concept to unknown surfaces with various surface curvatures to enhance the robustness of the trained policy in terms of changes in surfaces. We demonstrate $\sim 15\times $ improvement in trajectory accuracy compared to the previous LBMPC method and $\sim 18\times $ improvement compared to using the FDM approach. We suggest the applications of the proposed method for learning more challenging tasks such as milling, which are difficult to model and dependent on a wide range of process variables.

Accurate node localization in Wireless Sensor Networks (WSNs) is crucial for applications like environmental monitoring and military surveillance. Traditional localization methods, such as trilateration, often struggle with accuracy due to signal attenuation and environmental obstructions, leading to significant localization errors in practical scenarios. This paper aims to address the limitations of traditional localization methods by developing and evaluating a hybrid localization method that enhances accuracy and robustness in node localization within WSNs. The proposed method integrates the Bounding Box approach with the Harmony Search optimization algorithm, resulting in the Bounding Box Harmony Search (BBHS) method. The BBHS method utilizes the initial geometric constraints provided by the Bounding Box approach and refines these estimates using the global optimization capabilities of the Harmony Search algorithm. Simulation results, obtained using a custom-developed WSN localization simulator, demonstrate that the BBHS method significantly reduces localization errors compared to traditional trilateration and the standalone Bounding Box method. The BBHS method consistently provides enhanced accuracy and robustness across varying network conditions, highlighting its effectiveness in practical deployment scenarios. The advancements presented in this paper suggest that hybrid methods like BBHS represent a significant step forward in WSN localization technologies. By combining geometric constraints with optimization processes, the BBHS method overcomes the drawbacks of earlier techniques, paving the way for more reliable and resilient WSN operations.

Brain tumors pose a significant threat, especially when not detected early. The Inception v3 machine learning model has found extensive applications in computer vision and related fields. This study aims to develop a robust transfer learning model for classification, adaptable to various data modalities through neural networks. However, the training process for these neural networks is complex, being both demanding and computationally intensive. To tackle this challenge, we introduce an innovative training approach for Inception v3 referred to as ‘PSO-INCEPT’ (Particle Swarm Optimization-based Inception v3 training). In this method, the weight vectors for each Inception v3 model are analogized to particle positions in a phase space. The PSO cooperates with the ADAM optimizer in achieving the purpose of training with the best performance and generalization. This research is composed of two main parts, the first stage is being performed by the model independently using the ADAM optimizer. In the following stage, PSO-INCEPT models share the latest weight vectors or particle coordinates and loss function approximations via training. The optimization function then uses them to improve the validation accuracy. The effectiveness of PSO-INCEPT was evaluated through experiments that were conducted on Kaggle datasets that provide a common base ground by having four distinct classes. Experimental studies have proven the extraordinary ability of the proposed model by providing 99.33% validation accuracy and 99.95% training accuracy which shows exceptional performance.

Software-Defined Network (SDN) controller needs comprehensive visibility of the whole network to provide effective routing and forwarding decisions in the data layer. However, the topology discovery service in the SDN controller is vulnerable to the Topology Poisoning Attack (TPA), which targets corrupting the controller’s view on the connected devices (e.g., switches or hosts) to the network and inter-switch link connections. The attack could cause dramatic impacts on the network’s forwarding policy by changing the traffic path and even opening doors for Man-in-the-Middle (MitM) and Denial of Service (DoS) attacks. Recent studies presented sophisticated types of TPA, which could successfully bypass several well-known defence mechanisms for SDN. However, the scientific literature lacks a comprehensive review and survey of existing TPAs against topology discovery services and corresponding defence mechanisms. This paper provides a thorough survey to review and analyse existing threats against topology discovery services and a security assessment of the current countermeasures. For this aim, first, we propose a taxonomy for TPAs and categorise the attacks based on different parameters, including the attack aim, exploited vulnerability, location of the adversary, and communication channel. In addition, we provide a detailed root cause analysis per attack. Second, we perform a security assessment on the state-of-the-art security measurements that mitigate the risk of TPAs in SDN and discuss the advantages and disadvantages of each defence concerning the detection capability. Finally, we figure out several open security issues and outline possible future research directions to motivate security research on SDN. The rapid growth of the SDN market and the evolution of mobile networks, including components like the RAN Intelligent Controller (RIC) acting like SDN controller, highlight the critical need for SDN security in the future.

Transactive energy (TE) as a market-based mechanism provides a practical framework to fully manage and control local energy networks with a high penetration of distributed energy resources (DERs). However, the existing designed TE frameworks can rarely encourage prosumers to participate fairly owing to neglecting competition and cooperation. This paper proposes a tri-layer hybrid game-based TE framework, wherein the cooperation of prosumers to trade energy in a peer-to-peer (P2P) fashion is considered in the first layer using the Nash Bargaining Game (NBG) Theory. The competition among prosumers to trade with the most affordable aggregator is modeled at the second layer using the evolutionary game (EG). The third layer also models the competition among aggregators and the competition between aggregators and the cooperation of prosumers by developing a non-cooperative game. Besides, a scale-independent distributionally robust optimization (DRO) is developed based on the Wasserstein ambiguity set to allow prosumers to manage their uncertainty using all potential historical data while ensuring tractability. Finally, a new adaptive proximal alternative direction method of multipliers (ADMM) is introduced to develop a distributed-decentralized decision-making scheme for satisfying network constraints and energy trading in a P2P manner to accelerate the solution procedure and preserve privacy. The tests and implementations demonstrate that the proposed tri-layer TE framework lowered the overall costs for prosumers by 11 % and 2.85 % compared to the total costs in non-cooperative TE and cooperative TE, respectively.

Facial Expression Recognition (FER) is a machine learning problem that deals with recognizing human facial expressions. While existing work has achieved performance improvements in recent years, FER in the wild and under challenging conditions remains a challenge. In this paper, a lightweight patch and attention network based on MobileNetV1, referred to as PAtt-Lite, is proposed to improve FER performance under challenging conditions. A truncated ImageNet-pre-trained MobileNetV1 is utilized as the backbone feature extractor of the proposed method. In place of the truncated layers is a patch extraction block that is proposed for extracting significant local facial features to enhance the representation from MobileNetV1, especially under challenging conditions. An attention classifier is also proposed to improve the learning of these patched feature maps from the extremely lightweight feature extractor. The experimental results on public benchmark databases proved the effectiveness of the proposed method. PAtt-Lite achieved state-of-the-art results on CK+, RAF-DB, FER2013, FERPlus, and the challenging conditions subsets for RAF-DB and FERPlus.

The aim of this research is to enhance Oil and Gas (O&G) construction risk assessment using Fuzzy-based Failure Model Effect Analysis (FMEA) through the lens of O&G project managers in the U.S. A mixed-method approach was adopted for data collection, analysis, and processing, including semi-structured interviews with project managers to identify the key risks facing O&G construction projects; a Fuzzy-based FMEA to quantitatively analyse the level of significance of O&G risks; surveys to rank the assessment dimensions of the developed model and their components; and open-ended surveys to validate and verify the assessment model and its outputs, further expanding on the root causes of significant risks based on the assessment outputs, and to propose mitigation strategies for these risks. The research identified 41 risk factors classified under six categories, namely: management, technical and quality, financial and economic, health, safety, environmental, legal, and stakeholders’ risks. In addition, the risk assessment revealed that non-compliance with PPE regulations emerged as the most significant risk factor across all categories of O&G risks. This study offers valuable insights by assisting practitioners in better understanding the significant O&G risks that need to be addressed to ensure the successful execution and completion of O&G projects.

People are preoccupied with their own duties in today’s environment, making it tough to oversee and care for disabled people. Paralyzed and disabled people, on the other hand, have a powerful desire to move around freely. Different initiatives have been taken in the past to improve and preserve the self-esteem of such people, and various technologies have also been created to assist them in a better way. The primary advantage of a vision-based autonomous wheelchair is that it can provide greater independence to users with limited mobility. The methodology of other smart wheelchairs varies depending on the specific technology used. Some smart wheelchairs may use sensors to detect obstacles and provide feedback to the user, while others may incorporate GPS and other navigation technologies to assist with indoor and outdoor navigation. Our study proposes a revolutionary implementation of an autonomous system for fully handicapped people, allowing them to drive wheelchairs using just their eyes. The wheelchair’s orientation will be determined by the orientation of the eye. The camera will follow the movement of the eyes. Furthermore, sensors will be installed on the front side of the wheelchair to identify any obstacles along its path. The calibration of the camera with the human eye, without obstructing its vision, is the most challenging task in this research, besides controlling the wheels of the wheelchair.

Forest height (FH) is an important driver for aboveground biomass (AGB) that can be obtained using interferometric synthetic aperture radar (InSAR). However, the limited access to the quad-polarimetric data or high-accuracy terrain model makes FH retrieval a challenging task. This study aimed to retrieve FH and further predict AGB by combining TanDEM-X InSAR coherence, Sentinel-1 (S-1), and Sentinel-2 (S-2) data. A total of 125 sample plots with a size of 900 m2 were established in a broadleaved forest of Kheyroud, Iran. The linear and sinc models obtained by simplification of the random volume over ground model were used for deriving FHLin and FHSinc. Further investigation was conducted when S-1 and S-2 features including backscatters and multispectral information were added to FH predictions. Using the above-mentioned datasets and FH as an additional predictor, AGB was also predicted. K-nearest neighbor (k-NN), random forest (RF), and support vector regression (SVR) were employed for prediction. Lorey's mean height and AGB at sample plots were used in the accuracy assessment. Using the SVR method and synergy of FHSinc, S-1, and S-2 features, the FH prediction was improved (FHimp) with RMSE of 3.18 m and R2 = 0.59. The AGB prediction with RF and the combination of S-1 and S-2 features resulted in RMSE = 62.88 Mg·ha-1 (19.77%) that was improved to RMSE = 51.27 Mg·ha-1 (16.12%) when FHimp included. This study highlighted the capability of TanDEM-X InSAR coherence with certain geometry for FH prediction. Also, the importance of FH in AGB predictions can stimulate further attempts aiming at higher spatiotemporal accuracies.

This paper introduces a new compact Chair-shaped MIMO antenna with two radiating elements and a single layer of frequency-selective surface (FSS) for 5G Sub-6GHz communication systems. They use two techniques, Parasitic element, and (FSS), for isolation and gain enhancement, respectively. The $1\times2$ MIMO antenna using a coplanar waveguide (CPW) fed. Moreover, an FSS array structure consisting of (a 68-unit) Square-shaped structure with Circular Slot (SCS) shaped cells is employed using a new technique (Surround Technique) to enhance the gain and isolation between the elements of the MIMO antenna. The proposed MIMO antenna system is printed on a Rogers 4350B substrate with a thickness of 0.508 mm. The antenna’s performance is evaluated using S-parameters, radiation properties, and MIMO characteristics. The MIMO antenna system works in the Sub 6-GHz 5G band, which ranges from 3 to 6 GHz. Adding the FSS layer enhances the MIMO’s antenna gain to a peak measured gain of 7.96 dBi and it also improves the MIMO antenna’s isolation. The performance metrics of the proposed MIMO antenna were also investigated, including measures values of ECC = 0.004, DG = 9.99 dB, CCL = 0.2 bit/s/Hz, MEG = -3.13 dBi, and TARC = > 0dB exhibits advantageous antenna characteristics. The observed radiation characteristics of the suggested MIMO antenna system indicate its suitability for the upcoming 5G communication systems.

This work introduces a wideband metamaterial (MM) absorber designed to operate effectively across a wide reception angle and be polarization-insensitive within ISM band (2.4 GHz) applications. The proposed absorber unit cell comprises four copper sectors loaded with lumped resistors and a full copper ground plane hosted onto two FR4 substrates. Furthermore, an air layer suspended between the ground plane and a FR4 substrate is applied to achieve wideband absorption. In addition, the simulation results show that particular design factors, such as lumped resistors and unit cell geometry, can be optimized to improve the efficiency of the absorber. The simulations demonstrate that the proposed absorber achieves a wideband absorption, exceeding 90%, over a broad frequency range from 1.94 GHz to 2.98 GHz. The designed absorber was fabricated and tested, and the simulation and measurement results were agreed well.

In a modern industrial process, the accurate measurement of two-phase flow signals is of vital importance for various purposes including condition-based monitoring of pipelines, the avoidance of costly material and environmental damages, and danger to the operational personnel. In this work, a novel blind approach for denoising two-phase liquid and gas flow in a horizontal liquid-gas pipe is proposed. This approach employs the structured signal subspace (SSS) method on the multichannel signal acquired from the transducers connected to the wall of the pipe. The proposed approach utilizes only the output observations from the sensor, i.e, the recorded signal, and does not require any knowledge of the input signal or the channel. The multichannel signals recorded are first pre-processed using the principal component analysis (PCA) and then the blind SSS method is used to denoise the input signal before estimating it. The numerical results showed that the proposed algorithm outperforms the state-of-the-art algorithms (SOTA) which includes the eigen value decoposition (EGD)-based method and the independent component analysis (ICA)-based method, while the proposed PCA-SSS method achieved a performance of $-22.7dB$ in the presence of Gaussian noise, the EGD and ICA acheived $-16.39dB$ and $-18.09dB$ , respectively, showing the superiority of the proposed method. Similar analysis were performed in the presence of a non-Gaussian noise and colored noise and the proposed algorithm also outperformed the other methods. Hence, the PCA-SSS method can be used for a better characterization of a slug flow regime by exploiting the Toeplitz structure embedded in the signal vector acquired from the array of sensors via the communication model for denoising the two-phase flow, and does not rely on the knowledge of the input signal vector.

In industrial electromechanical manufacturing, ensuring the accuracy of electrical partial discharge measurements is crucial for the reliability and efficiency of electrical power systems. The cornerstone of such precision is a well-designed partial discharge calibrator that strictly aligns with the IEC 60270 standard parameters, including apparent charge, rise time, and pulse repetition frequency. This research delineates the systematic development and subsequent performance evaluations of an innovative industrial partial discharge calibrator. The calibrator uniquely integrates advanced functionality while maintaining cost efficiency, offering an alternative to the current partial discharge calibrator market. Its design allows for a broad spectrum of pulse frequency adjustments, making it adaptable to various apparent charge bandwidths and highlighting its adaptability for high-voltage applications. Building upon the foundational understanding of partial discharge as characterized by standards such as IEC 60270, the study underscores the calibrator’s capacity to elevate the standards of partial discharge measurements within high-voltage equipment.

There is a general concern that the increasing penetration of electric vehicles (EVs) will result in higher aging failure probability of equipment and reduced network reliability. The electricity costs may also increase, due to the exacerbation of peak load led by uncontrolled EV charging. This paper proposes a linear optimization model for the assessment of the benefits of EV smart charging on both network reliability improvement and electricity cost reduction. The objective of the proposed model is the cost minimization, including the loss of load, repair costs due to aging failures, and EV charging expenses. The proposed model incorporates a piecewise linear model representation for the failure probability distributions and utilizes a machine learning approach to represent the EV charging load. Considering two different test systems (a 5-bus network and the IEEE 33-bus network), this paper compares aging failure probabilities, service unavailability, expected energy not supplied, and total costs in various scenarios with and without the implementation of EV smart charging.

A zero-voltage zero-current-switching (ZVZCS) full-bridge DC-DC converter, based on a dual transformer with two output filter capacitors, is proposed in this paper. The proposed topology is suitable for application in renewable energy at medium voltage, integrated with a DC collection system. To realize voltage and power regulation for the converter, pulse width modulation (PWM) is employed with a fixed duty cycle for the main switches. Through careful design of the turns ratio of the main transformer, most of the power under full load range is delivered by the main full-bridge circuit, which also achieves zero-voltage switching (ZVS). This significantly reduces the switching losses. A small portion of the total power is delivered by the auxiliary circuit that realizes ZVZCS. Therefore, the turns ratio of the auxiliary transformer can be optimized such that the efficiency of the proposed converter can be improved. The optimzation is discussed in detail, along with the design parameters of the proposed converter. A prototype of 200V-2kV/3kW has been developed to verify the simulation results and validate the working of the proposed scheme.

Preserving sensor nodes’ energy in underwater sensor networks (UWSNs) stands as a crucial priority. UWSNs find key applications in ocean monitoring, offshore oil and gas exploration, and underwater robotic operations. Our study introduces a novel approach (CTSP) utilizing clustering alongside the Traveling Salesman (TS) Protocol to optimize data routing in UWSNs while minimizing energy usage. By adjusting the pathways of data transmission among sensor nodes, this method aims to curtail the network’s overall power consumption. CTSP primarily relies on two fundamental components: TS and clustering. Leveraging the TS Protocol allows the determination of the most efficient route between any pair of sensor nodes within the network. The approach ensures that each sensor node transmits data solely to its nearest neighbor, thereby reducing the energy required for transmission. Utilizing the positions of sensor nodes as input, a clustering algorithm forms larger groups. Enhanced communication within clusters and reduced long-range communication between clusters contribute to energy conservation. Simulation results demonstrate that the proposed method significantly diminishes power consumption compared to traditional routing methods like the LEACH algorithm. Precisely, the CTSP method exhibits the potential to reduce energy usage by up to 50%, presenting a feasible option for energy-efficient data routing in underwater settings.

The escalating incidence of thyroid cancer over the past decade underscores the imperative for effective classification and early detection of thyroid nodules. An automated system for this purpose could significantly aid physicians, particularly in expediting diagnostic processes. However, attaining this objective has proven challenging, primarily attributed to the constrained dataset size for medical images and the laborious process of feature extraction. This research addresses these challenges by thoroughly exploring the importance of extracting meaningful features for tumor detection and introducing a quantum-based convolutional neural network. The proposed approach employs a quantum filter transformation for intricate feature extraction, coupled with a classical neural network for the classification of thyroid nodules in ultrasound images. The classification process involves two distinct categorizations: distinguishing between nodules that are benign or malignant, and identifying the specific suspicious class to which the nodule is attributed. The amalgamation of both classifiers yields a comprehensive characterization of thyroid nodules, showcasing notable accuracy. For tumor classification, the model achieves an accuracy of 97.63%, precision of 97.72%, and recall of 97.64% on a test dataset containing 127 images. Similarly, for suspicious level classification, the model attains an accuracy of 93.87%, precision of 94.58%, and recall of 93.88% on a test dataset containing 98 images. These results surpass the performance of existing models, marking a significant advancement in the field of thyroid nodule classification. The proposed model represents a promising and innovative methodology that could offer valuable support mechanisms in clinical settings, facilitating the rapid classification and diagnosis of thyroid cancer.

Environmental monitoring for the detection and regulation of toxic gases is of paramount importance. This study proposes the modeling and fabrication of MEMS devices based on the standard Polysilicon Multi-Users MEMS Process (PolyMUMPs) for acetone gas detection. Titanium dioxide ( $TiO_{2}$ ) nanoparticles has been incorporated onto the top sensor’s surface, amplifying its detection sensitivity. The study delves into Electrothermal mechanisms that has integrated with the embedded microheater to induce thermal forces. This enhancement significantly improves the acetone gas detection capabilities of the device with high sensitivity and low detection limit. Results include the displacement amplitude of the uncoated PolyMUMPs sensor, with exploration of the coated device’s behavior under different heating voltages. Electrothermal actuation is employed, and the output voltage is measured using an MS3110 universal capacitive readout circuit. Comparisons of uncoated and coated devices has been investigated, demonstrating the influence of the sensing material that improve the detection performance. This investigation delves into the impact of varying coating thickness on the PolyMUMPs sensor’s performance. The mass sensitivity of the device is found to be 3.8574 mHz/pg, while the detection limit indicates that the device can detect up to 44 part per billion (ppb) at resonance frequency of 7.627 kHz.

This work tackles the scheduling challenge of microgrids for smart homes, aiming to optimize energy management with both renewable and non-renewable sources. A power control center orchestrates the microgrid, coordinating distributed energy resources (DERs) for peak demand fulfillment and excess energy utilization. We propose a proportional-integral control system for efficient demand response, achieving reduced post-scheduling costs and a peak-to-average ratio. Comparative analysis reveals Ant Colony Optimization outperforms Binary Particle Swarm Optimization in cost and peak-to-average ratio reduction. Simulations explore two scenarios: Case 1 integrates with the main grid for reliability, while Case 2 utilizes solely renewable energy sources. Although Case 2 exhibits superior performance, Case 1’s dependence on the main grid offers greater real-world feasibility. Therefore, Case 1 with optimized DER scheduling emerges as the recommended solution for enhancing microgrid efficiency and ensuring reliable power supply in smart homes.

Vision Transformers, known for their innovative architectural design and modeling capabilities, have gained significant attention in computer vision. This paper presents a dual-path approach that leverages the strengths of the Multi-Axis Vision Transformer (MaxViT) and the Improved Multiscale Vision Transformer (MViTv2). It starts by encoding speech signals into Constant-Q Transform (CQT) spectrograms and Mel Spectrograms with Short-Time Fourier Transform (Mel-STFT). The CQT spectrogram is then fed into the MaxViT model, while the Mel-STFT is input to the MViTv2 model to extract informative features from the spectrograms. These features are integrated and passed into a Multilayer Perceptron (MLP) model for final classification. This hybrid model is named the “MaxViT and MViTv2 Fusion Network with Multilayer Perceptron (MaxMViT-MLP).” The MaxMViT-MLP model achieves remarkable results with an accuracy of 95.28% on the Emo-DB, 89.12% on the RAVDESS dataset, and 68.39% on the IEMOCAP dataset, substantiating the advantages of integrating multiple audio feature representations and Vision Transformers in speech emotion recognition.

The main goals of fifth generation (5G) systems are to significantly increase the network capacity and to support new 5G service requirements. Ultra network densification with small cells is among the key pillars for 5G evolution. The inter-small-cell 5G backhaul network involves massive data traffic. Hence, it is important to have a centralized, efficient multi-hop routing protocol for backhaul networks to manage and speed up the routing decisions among small cells, while considering the 5G service requirements. This paper proposes a parallel multi-hop routing protocol to speed up routing decisions in 5G backhaul networks. To this end, we study the efficiency of utilizing the parallel platforms of cloud computing and high-performance computing (HPC) to manage and speed up the parallel routing protocol for different communication network sizes and set recommendations for utilizing cloud resources to adopt the parallel protocol. Our numerical results indicate that the HPC parallel implementation outperforms the cloud computing implementation, in terms of routing decision speed-up and scalability to large network sizes. In particular, for a large network size with 2048 nodes, our HPC implementation achieves a routing speed-up of 37x. However, the best routing speed-up achieved using our cloud computing implementation is 15.5x, and is recorded using one virtual machine (VM) for a network size of 1024 nodes. In summary, there is a trade-off between a better performance for HPC vs. flexible resources of cloud computing. Thus, choosing best fit platform for 5G routing protocols depends on the deployment scenarios at 5G core or edge network.

Leveraging machine learning to facilitate the optimization process is an emerging field that holds the promise to bypass the fundamental computational bottleneck caused by classic iterative solvers in critical applications requiring near-real-time optimization. The majority of existing approaches focus on learning data-driven optimizers that lead to fewer iterations in solving an optimization. In this paper, we take a different approach and propose to replace the iterative solvers altogether with a trainable parametric set function, that outputs the optimal arguments/parameters of an optimization problem in a single feed forward. We denote our method as Learning to Optimize the Optimization Process ( ${\mathcal{LOOP}}$ ). We show the feasibility of learning such parametric (set) functions to solve various classic optimization problems including linear/nonlinear regression, principal component analysis, transport-based coreset, and quadratic programming in supply management applications. In addition, we propose two alternative approaches for learning such parametric functions, with and without a solver in the ${\mathcal{LOOP}}$ . Finally, through various numerical experiments, we show that the trained solvers could be orders of magnitude faster than the classic iterative solvers while providing near optimal solutions.

The traditional two-factor authentication (2FA) methods primarily rely on the user manually entering a code or token during the authentication process. This can be burdensome and time-consuming, particularly for users who must be authenticated frequently. To tackle this challenge, we present a novel 2FA approach replacing the user’s input with decisions made by Machine Learning (ML) that continuously verifies the user’s identity with zero effort. Our system exploits unique environmental features associated with the user, such as beacon frame characteristics and Received Signal Strength Indicator (RSSI) values from Wi-Fi Access Points (APs). These features are gathered and analyzed in real-time by our ML algorithm to ascertain the user’s identity. For enhanced security, our system mandates that the user’s two devices (i.e., a login device and a mobile device) be situated within a predetermined proximity before granting access. This precaution ensures that unauthorized users cannot access sensitive information or systems, even with the correct login credentials. Through experimentation, we have demonstrated our system’s effectiveness in determining the location of the user’s devices based on beacon frame characteristics and RSSI values, achieving an accuracy of 92.4%. Additionally, we conducted comprehensive security analysis experiments to evaluate the proposed 2FA system’s resilience against various cyberattacks. Our findings indicate that the system exhibits robustness and reliability in the face of these threats. The scalability, flexibility, and adaptability of our system render it a promising option for organizations and users seeking a secure and convenient authentication system.

This article investigates the extraction of complex refractive indices from the amplitude and phase of the transmitted electric field. In the first step, an incident plane wave has been assumed and the amplitude and phase of the transmitted plane wave is calculated analytically. In this calculation, different values of the complex refractive index have been assumed for the non-magnetic material under test. In fact, the real part and imaginary part of the refractive index are assumed in the range of [1–10] and [0–1], respectively. Furthermore, a general study is made by an assumption of the material thickness to simulation wavelength ratio in the range of [0.01–20]. Due to examining the measurement noise, noisy data are produced for different values of signal-to-noise ratio in the range of [25–40] dB. Due to the difficulties of estimating the refractive index confronted in the theoretical or iterative methods, a Long short-term memory (LSTM) network is proposed and used for the estimation of complex refractive index based on the amplitude and phase of the transmitted electric field. It is shown that the estimation accuracy of about 97% can be achieved in the trained network. Furthermore, the estimation accuracy as a function of thickness-to-wavelength ratio, signal-to-noise ratio, and the values of real and imaginary parts of the refractive index are studied in detail and shown that higher estimation accuracy can be achieved. The simulated results have been confirmed by the measurement for the thickness-to-wavelength ratio below 0.1 and a good agreement has been found. Therefore, the proposed method can replace analytical or repetitive methods as an optimal and more accurate method.

Modern vehicles rely heavily on interconnected electronic control units (ECUs) through in-vehicle networks to perform crucial functions such as braking and monitoring engine RPMs. However, the increased number of ECUs and their connectivity to the in-vehicle network poses a security risk due to the lack of encryption and authentication protocols such as the controller area network (CAN). To address this problem, machine learning (ML) based intrusion detection systems (IDSs) have been proposed. However, existing IDSs suffer from low detection accuracy, limited real-time response, and high resource requirements. This study proposes an accurate and low-complexity IDS for in-vehicle networks based on feature fusion and ensemble learning called the Feature Fusion and Stacking-based IDS (FFS-IDS). FFS-IDS fuses multiple features extracted from raw network traffic and then classifies traffic instances into intrusive and non-intrusive categories using a stacking ensemble learning of basic machine learning classifiers. Specifically, a decision tree is employed as a base classifier, and random forest is used as a meta-learner. This work implements and validates the FFS-IDS using real-time car hacking data sets and achieves better performance than individual decision tree classifiers and popular ensemble learning methods such as Random Forest, LightGBM, AdaBoost, and ExtraTree algorithms. The results demonstrate that FFS-IDS can detect Denial of Service (DoS), Gear spoofing, and RPM spoofing attacks with up to 99% accuracy and Fuzzy attacks with up to 97.5% accuracy using benchmark datasets. Overall, this study shows the effectiveness and practicality of FFS-IDS in detecting intrusions in in-vehicle networks, which is essential for ensuring the cybersecurity and safety of modern vehicles. Future work in this area could involve exploring additional feature extraction techniques and fine-tuning hyperparameters to improve the performance of IDSs further.