Search

The accuracy of a predictive system is critical for predictive maintenance and to support the right decisions at the right times. Statistical models, such as ARIMA and SARIMA, are unable to describe the stochastic nature of the data. Neural networks, such as long short-term memory (LSTM) and the gated recurrent unit (GRU), are good predictors for univariate and multivariate data. The present paper describes a case study where the performances of long short-term memory and gated recurrent units are compared, based on different hyperparameters. In general, gated recurrent units exhibit better performance, based on a case study on pulp paper presses. The final result demonstrates that, to maximize the equipment availability, gated recurrent units, as demonstrated in the paper, are the best options.

This paper investigates and proposes a solution for Protocol Independent Switch Architecture (PISA) to process application layer data, enabling the inspection of application content. PISA is a novel approach in networking where the switch does not run any embedded binary code but rather an interpreted code written in a domain-specific language. The main motivation behind this approach is that telecommunication operators do not want to be locked in by a vendor for any type of networking equipment, develop their own networking code in a hardware environment that is not governed by a single equipment manufacturer. This approach also eases the modeling of equipment in a simulation environment as all of the components of a hardware switch run the same compatible code in a software modeled switch. The novel techniques in this paper exploit the main functions of a programmable switch and combine the streaming data processor to create the desired effect from a telecommunication operator perspective to lower the costs and govern the network in a comprehensive manner. The results indicate that the proposed solution using PISA switches enables application visibility in an outstanding performance. This ability helps the operators to remove a fundamental gap between flexibility and scalability by making the best use of limited compute resources in application identification and the response to them. The experimental study indicates that, without any optimization, the proposed solution increases the performance of application identification systems 5.5 to 47.0 times. This study promises that DPI, NGFW (Next-Generation Firewall), and such application layer systems which have quite high costs per unit traffic volume and could not scale to a Tbps level, can be combined with PISA to overcome the cost and scalability issues.

Nowadays, there are different kinds of public knowledge bases for cyber security vulnerability and threat intelligence which can be used for IoT security threat analysis. However, the heterogeneity of these knowledge bases and the complexity of the IoT environments make network security situation awareness and threat assessment difficult. In this paper, we integrate vulnerabilities, weaknesses, affected platforms, tactics, attack techniques, and attack patterns into a coherent set of links. In addition, we propose an IoT security ontology model, namely, the IoT Security Threat Ontology (IoTSTO), to describe the elements of IoT security threats and design inference rules for threat analysis. This IoTSTO expands the current knowledge domain of cyber security ontology modeling. In the IoTSTO model, the proposed multi-source knowledge reasoning method can perform the following tasks: assess the threats of the IoT environment, automatically infer mitigations, and separate IoT nodes that are subject to specific threats. The method above provides support to security managers in their deployment of security solutions. This paper completes the association of current public knowledge bases for IoT security and solves the semantic heterogeneity of multi-source knowledge. In this paper, we reveal the scope of public knowledge bases and their interrelationships through the multi-source knowledge reasoning method for IoT security. In conclusion, the paper provides a unified, extensible, and reusable method for IoT security analysis and decision making.

This paper considers the use of a post metadata-based approach to identifying intentionally deceptive online content. It presents the use of an inherently explainable artificial intelligence technique, which utilizes machine learning to train an expert system, for this purpose. It considers the role of three factors (textual context, speaker background, and emotion) in fake news detection analysis and evaluates the efficacy of using key factors, but not the inherently subjective processing of post text itself, to identify deceptive online content. This paper presents initial work on a potential deceptive content detection tool and also, through the networks that it presents for this purpose, considers the interrelationships of factors that can be used to determine whether a post is deceptive content or not and their comparative importance.

With the expansion of the Internet of Things (IoT), security incidents about exploiting vulnerabilities in IoT devices have become prominent. However, due to the characteristics of IoT devices such as low power and low performance, it is difficult to apply existing security solutions to IoT devices. As a result, IoT devices have easily become targets for cyber attackers, and malware attacks on IoT devices are increasing every year. The most representative is the Mirai malware that caused distributed denial of service (DDoS) attacks by creating a massive IoT botnet. Moreover, Mirai malware has been released on the Internet, resulting in increasing variants and new malicious codes. One of the ways to mitigate distributed denial of service attacks is to render the creation of massive IoT botnets difficult by preventing the spread of malicious code. For IoT infrastructure security, security solutions are being studied to analyze network packets going in and out of IoT infrastructure to detect threats, and to prevent the spread of threats within IoT infrastructure by dynamically controlling network access to maliciously used IoT devices, network equipment, and IoT services. However, there is a great risk to apply unverified security solutions to real-world environments. In this paper, we propose a malware simulation tool that scans vulnerable IoT devices assigned a private IP address, and spreads malicious code within IoT infrastructure by injecting malicious code download command into vulnerable devices. The malware simulation tool proposed in this paper can be used to verify the functionality of network threat detection and prevention solutions.

Small and medium-sized enterprises represent the majority of enterprises globally and yet have some difficulties in understanding the impact that cybersecurity threats could have on their businesses and the damage they could do to their assets. This study aims to measure the effectiveness of security practices at small-sized enterprises in Saudi Arabia in the event of a cybersecurity attack. Our paper is among the first research papers to measure the effectiveness of cybersecurity practices and the threat posed by cybersecurity breaches among small enterprises in the event of cybersecurity attacks. A total of 282 respondents participated, all of them representing small-sized enterprises in Saudi Arabia. The study applies multiple regression tests to analyze the effectiveness of 12 cybersecurity practices in three aspects: financial damage, loss of sensitive data, and restoration time, at small enterprises. The findings indicate that having an inspection team and a recovery plan may limit the financial damage caused by cybersecurity attacks on small enterprises. The results also show that cybersecurity awareness, knowledge of cybersecurity damage, and professionals’ salaries were related to the loss of sensitive data. Furthermore, the results indicate that contact with cybersecurity authorities and having an inspection team have statistically significant effects on restoration time.

Face detection, which is an effortless task for humans, is complex to perform on machines. The recent veer proliferation of computational resources is paving the way for frantic advancement of face detection technology. Many astutely developed algorithms have been proposed to detect faces. However, there is little attention paid in making a comprehensive survey of the available algorithms. This paper aims at providing fourfold discussions on face detection algorithms. First, we explore a wide variety of the available face detection algorithms in five steps, including history, working procedure, advantages, limitations, and use in other fields alongside face detection. Secondly, we include a comparative evaluation among different algorithms in each single method. Thirdly, we provide detailed comparisons among the algorithms epitomized to have an all-inclusive outlook. Lastly, we conclude this study with several promising research directions to pursue. Earlier survey papers on face detection algorithms are limited to just technical details and popularly used algorithms. In our study, however, we cover detailed technical explanations of face detection algorithms and various recent sub-branches of the neural network. We present detailed comparisons among the algorithms in all-inclusive and under sub-branches. We provide the strengths and limitations of these algorithms and a novel literature survey that includes their use besides face detection.

The rise of deep learning technology has broadly promoted the practical application of artificial intelligence in production and daily life. In computer vision, many human-centered applications, such as video surveillance, human-computer interaction, digital entertainment, etc., rely heavily on accurate and efficient human pose estimation techniques. Inspired by the remarkable achievements in learning-based 2D human pose estimation, numerous research studies are devoted to the topic of 3D human pose estimation via deep learning methods. Against this backdrop, this paper provides an extensive literature survey of recent literature about deep learning methods for 3D human pose estimation to display the development process of these research studies, track the latest research trends, and analyze the characteristics of devised types of methods. The literature is reviewed, along with the general pipeline of 3D human pose estimation, which consists of human body modeling, learning-based pose estimation, and regularization for refinement. Different from existing reviews of the same topic, this paper focus on deep learning-based methods. The learning-based pose estimation is discussed from two categories: single-person and multi-person. Each one is further categorized by data type to the image-based methods and the video-based methods. Moreover, due to the significance of data for learning-based methods, this paper surveys the 3D human pose estimation methods according to the taxonomy of supervision form. At last, this paper also enlists the current and widely used datasets and compares performances of reviewed methods. Based on this literature survey, it can be concluded that each branch of 3D human pose estimation starts with fully-supervised methods, and there is still much room for multi-person pose estimation based on other supervision methods from both image and video. Besides the significant development of 3D human pose estimation via deep learning, the inherent ambiguity and occlusion problems remain challenging issues that need to be better addressed.

Machine learning-based systems are gaining interest in the field of medicine, mostly in medical imaging and diagnosis. In this paper, we address the problem of automatic cerebral microbleeds (CMB) detection in magnetic resonance images. It is challenging due to difficulty in distinguishing a true CMB from its mimics, however, if successfully solved, it would streamline the radiologists work. To deal with this complex three-dimensional problem, we propose a machine learning approach based on a 2D Faster RCNN network. We aimed to achieve a reliable system, i.e., with balanced sensitivity and precision. Therefore, we have researched and analysed, among others, impact of the way the training data are provided to the system, their pre-processing, the choice of model and its structure, and also the ways of regularisation. Furthermore, we also carefully analysed the network predictions and proposed an algorithm for its post-processing. The proposed approach enabled for obtaining high precision (89.74%), sensitivity (92.62%), and F1 score (90.84%). The paper presents the main challenges connected with automatic cerebral microbleeds detection, its deep analysis and developed system. The conducted research may significantly contribute to automatic medical diagnosis.

In recent years, interest in home energy management systems (HEMS) has grown significantly, as well as the development of Voice Assistants that substantially increase home comfort. This paper presents a novel merging of HEMS with the Assistant paradigm. The combination of both concepts has allowed the creation of a high-performance and easy-to-manage expert system (ES). It has been developed in a framework that includes, on the one hand, the efficient energy management functionality boosted with an Internet of Things (IoT) platform, where artificial intelligence (AI) and big data treatment are blended, and on the other hand, an assistant that interacts both with the user and with the HEMS itself. The creation of this ES has made it possible to optimize consumption levels, improve security, efficiency, comfort, and user experience, as well as home security (presence simulation or security against intruders), automate processes, optimize resources, and provide relevant information to the user facilitating decision making, all based on a multi-objective optimization (MOP) problem model. This paper presents both the scheme and the results obtained, the synergies generated, and the conclusions that can be drawn after 24 months of operation.

Many methods such as biomechanics and coaching have been proposed to help people learn a certain movement. There have been proposals for methods to discover characteristics of movement based on information obtained from videos and sensors. Especially in sports, it is expected that these methods can provide hints to improve movement skills. However, conventional methods focus on individual movements, and do not consider cases where external factors influence the movement, such as combat sports. In this paper, we propose a novel method called the Extraction for Successful Movement method (XSM method). Applying the method, this paper focuses on throwing techniques in judo to discover key factors that induce successful throwing from the postures right before initiating the throwing techniques. We define candidate factors by observing the video scenes where the throwing techniques are successfully performed. The method demonstrates the significance of the key factors according to the predominance of factors by χ2 test and residual analysis. Applying the XSM method to the dataset obtained from the videos of the Judo World Championships, we demonstrate the validity of the method with discussing the key factors related to the successful throwing techniques.

Due to the advancement of science and technology, modern cars are highly technical, more activity occurs inside the car and driving is faster; however, statistics show that the number of road fatalities have increased in recent years because of drivers’ unsafe behaviors. Therefore, to make the traffic environment safe it is important to keep the driver alert and awake both in human and autonomous driving cars. A driver’s cognitive load is considered a good indication of alertness, but determining cognitive load is challenging and the acceptance of wire sensor solutions are not preferred in real-world driving scenarios. The recent development of a non-contact approach through image processing and decreasing hardware prices enables new solutions and there are several interesting features related to the driver’s eyes that are currently explored in research. This paper presents a vision-based method to extract useful parameters from a driver’s eye movement signals and manual feature extraction based on domain knowledge, as well as automatic feature extraction using deep learning architectures. Five machine learning models and three deep learning architectures are developed to classify a driver’s cognitive load. The results show that the highest classification accuracy achieved is 92% by the support vector machine model with linear kernel function and 91% by the convolutional neural networks model. This non-contact technology can be a potential contributor in advanced driver assistive systems.

One of the biggest challenge in the field of deep learning is the parameter selection and optimization process. In recent years different algorithms have been proposed including bio-inspired solutions to solve this problem, however, there are many challenges including local minima, saddle points, and vanishing gradients. In this paper, we introduce the Whale Optimisation Algorithm (WOA) based on the swarm foraging behavior of humpback whales to optimise neural network hyperparameters. We wish to stress that to the best of our knowledge this is the first attempt that uses Whale Optimisation Algorithm for the optimisation task of hyperparameters. After a detailed description of the WOA algorithm we formulate and explain the application in deep learning, present the implementation, and compare the proposed algorithm with other well-known algorithms including widely used Grid and Random Search methods. Additionally, we have implemented a third dimension feature analysis to the original WOA algorithm to utilize 3D search space (3D-WOA). Simulations show that the proposed algorithm can be successfully used for hyperparameters optimization, achieving accuracy of 89.85% and 80.60% for Fashion MNIST and Reuters datasets, respectively.

Solar radiation prediction is an important process in ensuring optimal exploitation of solar energy power. Numerous models have been applied to this problem, such as numerical weather prediction models and artificial intelligence models. However, well-designed hybridization approaches that combine numerical models with artificial intelligence models to yield a more powerful model can provide a significant improvement in prediction accuracy. In this paper, novel hybrid machine learning approaches that exploit auxiliary numerical data are proposed. The proposed hybrid methods invoke different machine learning paradigms, including feature selection, classification, and regression. Additionally, numerical weather prediction (NWP) models are used in the proposed hybrid models. Feature selection is used for feature space dimension reduction to reduce the large number of recorded parameters that affect estimation and prediction processes. The rough set theory is applied for attribute reduction and the dependency degree is used as a fitness function. The effect of the attribute reduction process is investigated using thirty different classification and prediction models in addition to the proposed hybrid model. Then, different machine learning models are constructed based on classification and regression techniques to predict solar radiation. Moreover, other hybrid prediction models are formulated to use the output of the numerical model of Weather Research and Forecasting (WRF) as learning elements in order to improve the prediction accuracy. The proposed methodologies are evaluated using a data set that is collected from different regions in Saudi Arabia. The feature-reduction has achieved higher classification rates up to 8.5% for the best classifiers and up to 15% for other classifiers, for the different data collection regions. Additionally, in the regression, it achieved improvements of average root mean square error up to 5.6% and in mean absolute error values up to 8.3%. The hybrid models could reduce the root mean square errors by 70.2% and 4.3% than the numerical and machine learning models, respectively, when these models are applied to some dataset. For some reduced feature data, the hybrid models could reduce the root mean square errors by 47.3% and 14.4% than the numerical and machine learning models, respectively.

With the advent of the new media mobile Internet era, the network public opinion in colleges and universities, as an extension of social network public opinion, is also facing a crisis in the prevention, control, and governance system. In this paper, the Fiddler was used to collect the comments and other relevant data of the COVID-19 topic articles on the WeChat Official Accounts of China’s top ten universities in 2020. The BILSTM_LSTM sentiment analysis model was used to analyze the sentiment tendency of the comments, and the LDA topic model was used to mine the topics of the comments with different emotional attributes at different stages of COVID-19. Based on sentiment analysis and text mining, entities and relationships in the theme graph of public opinion events in colleges and universities were identified, and the Neo4j graph database was established to construct the sentimental knowledge graph of the pandemic theme of university public accounts. People’s attitudes in university public opinion are easily influenced by a variety of factors, and the degree of emotional disposition changes over time, with the stage the pandemic is in, and with different commentators; official account opinion topics change with the development of the time stage of the pandemic, and students’ positive and negative comment topics show a diverse trend. By incorporating topic mining into the sentimental knowledge graph, the graph can realize functions such as the emotion retrieval of comments on university public numbers, a source search of security threats in university social networks, and monitoring of comments on public opinion under the theme of the pandemic, which provides new ideas for further exploring the research and governance system of university network public opinion and is conducive to preventing and resolving campus public opinion crises.

As the amount of data collected and analyzed by machine learning technology increases, data that can identify individuals is also being collected in large quantities. In particular, as deep learning technology—which requires a large amount of analysis data—is activated in various service fields, the possibility of exposing sensitive information of users increases, and the user privacy problem is growing more than ever. As a solution to this user’s data privacy problem, homomorphic encryption technology, which is an encryption technology that supports arithmetic operations using encrypted data, has been applied to various field including finance and health care in recent years. If so, is it possible to use the deep learning service while preserving the data privacy of users by using the data to which homomorphic encryption is applied? In this paper, we propose three attack methods to infringe user’s data privacy by exploiting possible security vulnerabilities in the process of using homomorphic encryption-based deep learning services for the first time. To specify and verify the feasibility of exploiting possible security vulnerabilities, we propose three attacks: (1) an adversarial attack exploiting communication link between client and trusted party; (2) a reconstruction attack using the paired input and output data; and (3) a membership inference attack by malicious insider. In addition, we describe real-world exploit scenarios for financial and medical services. From the experimental evaluation results, we show that the adversarial example and reconstruction attacks are a practical threat to homomorphic encryption-based deep learning models. The adversarial attack decreased average classification accuracy from 0.927 to 0.043, and the reconstruction attack showed average reclassification accuracy of 0.888, respectively.

The use of innovative and sophisticated malware definitions poses a serious threat to computer-based information systems. Such malware is adaptive to the existing security solutions and often works without detection. Once malware completes its malicious activity, it self-destructs and leaves no obvious signature for detection and forensic purposes. The detection of such sophisticated malware is very challenging and a non-trivial task because of the malware’s new patterns of exploiting vulnerabilities. Any security solutions require an equal level of sophistication to counter such attacks. In this paper, a novel reinforcement model based on Monte-Carlo simulation called eRBCM is explored to develop a security solution that can detect new and sophisticated network malware definitions. The new model is trained on several kinds of malware and can generalize the malware detection functionality. The model is evaluated using a benchmark set of malware. The results prove that eRBCM can identify a variety of malware with immense accuracy.

Fault diagnosis in multistage manufacturing processes (MMPs) is a challenging task where most of the research presented in the literature considers a predefined inspection scheme to identify the sources of variation and make the process diagnosable. In this paper, a sequential inspection procedure to detect the process fault based on a sequential testing algorithm and a minimum monitoring system is proposed. After the monitoring system detects that the process is out of statistical control, the features to be inspected (end of line or in process measurements) are defined sequentially according to the expected information gain of each potential inspection measurement. A case study is analyzed to prove the benefits of this approach with respect to a predefined inspection scheme and a randomized sequential inspection considering both the use and non-use of fault probabilities from historical maintenance data.

Distributed Acoustic Sensing (DAS) is a promising new technology for pipeline monitoring and protection. However, a big challenge is distinguishing between relevant events, like intrusion by an excavator near the pipeline, and interference, like land machines. This paper investigates whether it is possible to achieve adequate detection accuracy with classic machine learning algorithms using simulations and real system implementation. Then, we compare classical machine learning with a deep learning approach and analyze the advantages and disadvantages of both approaches. Although acceptable performance can be achieved with both approaches, preliminary results show that deep learning is the more promising approach, eliminating the need for laborious feature extraction and offering a six times lower event detection delay and twelve times lower execution time. However, we achieved the best results by combining deep learning with the knowledge-based and classical machine learning approaches. At the end of this manuscript, we propose general guidelines for efficient system design combining knowledge-based, classical machine learning, and deep learning approaches.

In this paper, we present a handwritten character recognition (HCR) system that aims to recognize first-order logic handwritten formulas and create editable text files of the recognized formulas. Dense feedforward neural networks (NNs) are utilized, and their performance is examined under various training conditions and methods. More specifically, after three training algorithms (backpropagation, resilient propagation and stochastic gradient descent) had been tested, we created and trained an NN with the stochastic gradient descent algorithm, optimized by the Adam update rule, which was proved to be the best, using a trainset of 16,750 handwritten image samples of 28 × 28 each and a testset of 7947 samples. The final accuracy achieved is 90.13%. The general methodology followed consists of two stages: the image processing and the NN design and training. Finally, an application has been created that implements the methodology and automatically recognizes handwritten logic formulas. An interesting feature of the application is that it allows for creating new, user-oriented training sets and parameter settings, and thus new NN models.

In the context of smart agriculture, high-value data sensing in the entire crop lifecycle is fundamental for realizing crop cultivation control. However, the existing data sensing methods are deficient regarding the sensing data value, poor data correlation, and high data collection cost. The main problem for data sensing over the entire crop lifecycle is how to sense high-value data according to crop growth stage at a low cost. To solve this problem, a data sensing framework was developed by combining edge computing with the Internet of Things, and a novel data sensing strategy for the entire crop lifecycle is proposed in this paper. The proposed strategy includes four phases. In the first phase, the crop growth stage is divided by Gath-Geva (GG) fuzzy clustering, and the key growth parameters corresponding to the growth stage are extracted. In the second phase, based on the current crop growth information, a prediction method of the current crop growth stage is constructed by using a Tkagi-Sugneo (T-S) fuzzy neural network. In the third phase, based on Deng’s grey relational analysis method, the environmental sensing parameters of the corresponding crop growth stage are optimized. In the fourth phase, an adaptive sensing method of sensing nodes with effective sensing area constraints is established. Finally, based on the actual crop growth history data, the whole crop life cycle dataset is established to test the performance and prediction accuracy of the proposed method for crop growth stage division. Based on the historical data, the simulation data sensing environment is established. Then, the proposed algorithm is tested and compared with the traditional algorithms. The comparison results show that the proposed strategy can divide and predict a crop growth cycle with high accuracy. The proposed strategy can significantly reduce the sensing and data collection times and energy consumption and significantly improve the value of sensing data.

The paper places emphasis on primary energy resources, their covariation, and their correlation with socioeconomic factors and aims to provide a systematic analysis of their development over time. The analysis uses evidence from European Union (EU) country-level data and is based on visual analytics techniques. Different results from the same territories show that energy consumption does not always reflect or is due to climatological or meteorological conditions. Extensive use of visualization is adopted as a means of contributing to the understanding of energy use, some involved problems and concepts, and energy consumption trends over time. We present an approach that addresses the informatics challenges based on the integration of visualization software, data integration, and cluster analysis. Our cross-sectional energy review advocates that EU energy leaders are moving towards a low-carbon economy. The correlations of energy variables with economic and pollution effects are stronger in greater levels of energy use, which means that energy use has an obvious impact on economic growth and the environment. Visual and automated methods employed for the analysis, reveal the direction, the strength, and the nature of the dependence structure, in clusters covering the range of energy use in EU 28 countries.

In view of the demand of location awareness in a special complex environment, for an unmanned aerial vehicle (UAV) airborne multi base-station semi-passive positioning system, the hybrid positioning solutions and optimized site layout in the positioning system can effectively improve the positioning accuracy for a specific region. In this paper, the geometric dilution of precision (GDOP) formula of a time difference of arrival (TDOA) and angles of arrival (AOA) hybrid location algorithm is deduced. Mayfly optimization algorithm (MOA) which is a new swarm intelligence optimization algorithm is introduced, and a method to find the optimal station of the UAV airborne multiple base station’s semi-passive positioning system using MOA is proposed. The simulation and analysis of the optimization of the different number of base stations, compared with other station layout methods, such as particle swarm optimization (PSO), genetic algorithm (GA), and artificial bee colony (ABC) algorithm. MOA is less likely to fall into local optimum, and the error of regional target positioning is reduced. By simulating the deployment of four base stations and five base stations in various situations, MOA can achieve a better deployment effect. The dynamic station configuration capability of the multi-station semi-passive positioning system has been improved with the UAV.

This paper is an initial endeavor to bridge the gap between powerful Deep Reinforcement Learning methodologies and the problem of exploration/coverage of unknown terrains. Within this scope, MarsExplorer, an openai-gym compatible environment tailored to exploration/coverage of unknown areas, is presented. MarsExplorer translates the original robotics problem into a Reinforcement Learning setup that various off-the-shelf algorithms can tackle. Any learned policy can be straightforwardly applied to a robotic platform without an elaborate simulation model of the robot's dynamics to apply a different learning/adaptation phase. One of its core features is the controllable multi-dimensional procedural generation of terrains, which is the key for producing policies with strong generalization capabilities. Four different state-of-the-art RL algorithms (A3C, PPO, Rainbow, and SAC) are trained on the MarsExplorer environment, and a proper evaluation of their results compared to the average human-level performance is reported. In the follow-up experimental analysis, the effect of the multi-dimensional difficulty setting on the learning capabilities of the best-performing algorithm (PPO) is analyzed. A milestone result is the generation of an exploration policy that follows the Hilbert curve without providing this information to the environment or rewarding directly or indirectly Hilbert-curve-like trajectories. The experimental analysis is concluded by evaluating PPO learned policy algorithm side-by-side with frontier-based exploration strategies. A study on the performance curves revealed that PPO-based policy was capable of performing adaptive-to-the-unknown-terrain sweeping without leaving expensive-to-revisit areas uncovered, underlying the capability of RL-based methodologies to tackle exploration tasks efficiently. The source code can be found at: https://github.com/dimikout3/MarsExplorer.

A long interactive TCP connection chain has been widely used by attackers to launch their attacks and thus avoid detection. The longer a connection chain, the higher the probability the chain is exploited by attackers. Round-trip Time (RTT) can represent the length of a connection chain. In order to obtain the RTTs from the sniffed Send and Echo packets in a connection chain, matching the Sends and Echoes is required. In this paper, we first model a network traffic as the collection of RTTs and present the rationale of using the RTTs of a connection chain to represent the length of the chain. Second, we propose applying MMD data mining algorithm to match TCP Send and Echo packets collected from a connection. We found that the MMD data mining packet-matching algorithm outperforms all the existing packet-matching algorithms in terms of packet-matching rate including sequence number-based algorithm, Yang’s approach, Step-function, Packet-matching conservative algorithm and packet-matching greedy algorithm. The experimental results from our local area networks showed that the packet-matching accuracy of the MMD algorithm is 100%. The average packet-matching rate of the MMD algorithm obtained from the experiments conducted under the Internet context can reach around 94%. The MMD data mining packet-matching algorithm can fix the issue of low packet-matching rate faced by all the existing packet-matching algorithms including the state-of-the-art algorithm. It is applicable to network-based stepping-stone intrusion detection.

This paper presents a robust optimization model to find out the day-ahead energy and reserve to be scheduled by an electric vehicle (EV) aggregator. Energy can be purchased from, and injected to, the distribution network, while upward and downward reserves can be also provided by the EV aggregator. Although it is an economically driven model, the focus of this work relies on the actual availability of the scheduled reserves in a future real-time. To this end, two main features stand out: on one hand, the uncertainty regarding the EV driven pattern is modeled through a robust approach and, on the other hand, a set of non-anticipativity constraints are included to prevent from unavailable future states. The proposed model is posed as a mixed-integer robust linear problem in which binary variables are used to consider the charging, discharging or idle status of the EV aggregator. Results over a 24-h case study show the capability of the proposed model.

In this paper, a novel neural network-based robust control method is presented for a vehicle-oriented problem, in which the main goal is to ensure stable motion of the vehicle under critical circumstances. The proposed method can be divided into two main steps. In the first step, the model matching algorithm is proposed, which can adjust the nonlinear dynamics of the controlled system to a nominal, linear model. The aim of model matching is to eliminate the effects of the nonlinearities and uncertainties of the system to increase the performances of the closed-loop system. The model matching process results in an additional control input, which is computed by a neural network during the operation of the control system. Furthermore, in the second step, a robust H∞ is designed, which has double purposes: to handle the fitting error of the neural network and ensure the accurate tracking of the reference signal. The operation and efficiency of the proposed control algorithm are investigated through a complex test scenario, which is performed in the high-fidelity vehicle dynamics simulation software, CarMaker.

Short-Term Load Forecasting is critical for reliable power system operation, and the search for enhanced methodologies has been a constant field of investigation, particularly in an increasingly competitive environment where the market operator and its participants need to better inform their decisions. Hence, it is important to continue advancing in terms of forecasting accuracy and consistency. This paper presents a new deep learning-based ensemble methodology for 24 h ahead load forecasting, where an automatic framework is proposed to select the best Box-Jenkins models (ARIMA Forecasters), from a wide-range of combinations. The method is distinct in its parameters but more importantly in considering different batches of historical (training) data, thus benefiting from prediction models focused on recent and longer load trends. Afterwards, these accurate predictions, mainly the linear components of the load time-series, are fed to the ensemble Deep Forward Neural Network. This flexible type of network architecture not only functions as a combiner but also receives additional historical and auxiliary data to further its generalization capabilities. Numerical testing using New England market data validated the proposed ensemble approach with diverse base forecasters, achieving promising results in comparison with other state-of-the-art methods.

Cloud computing is an emerging paradigm that offers flexible and seamless services for users based on their needs, including user budget savings. However, the involvement of a vast number of cloud users has made the scheduling of users’ tasks (i.e., cloudlets) a challenging issue in selecting suitable data centres, servers (hosts), and virtual machines (VMs). Cloudlet scheduling is an NP-complete problem that can be solved using various meta-heuristic algorithms, which are quite popular due to their effectiveness. Massive user tasks and rapid growth in cloud resources have become increasingly complex challenges, therefore, an efficient algorithm is necessary for allocating cloudlets efficiently to attain better execution times, resource utilisation, and waiting times. This paper proposes a cloudlet scheduling, locust inspired algorithm to reduce the average makespan and waiting time and to boost VM and server utilisation. The CloudSim toolkit was used to evaluate our algorithm’s efficiency, and the obtained results revealed that our algorithm outperforms other state-of-the-art nature-inspired algorithms, improving the average makespan, waiting time, and resource utilisation.

Internet of Things (IoT) networks have been integrated into industrial infrastructure schemes, positioning themselves as devices that communicate highly classified information for the most critical companies of world nations. Currently, and in order to look for alternatives to mitigate this risk, solutions based on Blockchain algorithms and Machine Learning techniques have been implemented separately with the aim of mitigating potential threats in IIoT networks. In this paper, we sought to integrate the previous solutions to create an integral protection mechanism for IoT device networks, which would allow the identification of threats, activate secure information transfer mechanisms, and it would be adapted to the computational capabilities of industrial IoT. The proposed solution achieved the proposed objectives and is presented as a viable mechanism for detecting and containing intruders in an IoT network. In some cases, it overcomes traditional detection mechanisms such as an IDS.

The growing availability of mobile devices has lead to an arising development of smart cities services that share a huge amount of (personal) information and data. Without accurate and verified management, they could become severe back-doors for security and privacy. In this paper, we propose a smart city infrastructure able to integrate a distributed privacy-preserving identity management solution based on attribute-based credentials (p-ABC), a user-centric Consent Manager, and a GDPR-based Access Control mechanism so as to guarantee the enforcement of the GDPR’s provisions. Thus, the infrastructure supports the definition of specific purpose, collection of data, regulation of access to personal data, and users’ consents, while ensuring selective and minimal disclosure of personal information as well as user’s unlinkability across service and identity providers. The proposal has been implemented, integrated, and evaluated in a fully-fledged environment consisting of MiMurcia, the Smart City project for the city of Murcia, CaPe, an industrial consent management system, and GENERAL_D, an academic GDPR-based access control system, showing the feasibility.

Building an effective Machine Learning (ML) model for a data set is a difficult task involving various steps. One of the most important steps is to compare a substantial amount of generated ML models to find the optimal one for deployment. It is challenging to compare such models with a dynamic number of features. Comparison is more than only finding differences of ML model performance, as users are also interested in the relations between features and model performance such as feature importance for ML explanations. This paper proposes RadialNet Chart, a novel visualisation approach, to compare ML models trained with a different number of features of a given data set while revealing implicit dependent relations. In RadialNet Chart, ML models and features are represented by lines and arcs, respectively. These lines are generated effectively using a recursive function. The dependence of ML models with a dynamic number of features is encoded into the structure of visualisation, where ML models and their dependent features are directly revealed from related line connections. ML model performance information is encoded with colour and line width in RadialNet Chart. Taken together with the structure of visualisation, feature importance can be directly discerned in RadialNet Chart for ML explanations. Compared with other commonly used visualisation approaches, RadialNet Chart can help to simplify the ML model comparison process with different benefits such as the following: more efficient in terms of helping users to focus their attention to find visual elements of interest and easier to compare ML performance to find optimal ML model and discern important features visually and directly instead of through complex algorithmic calculations for ML explanations.

Data-driven methodologies are among the most effective tools for damage detection of complex existing buildings, such as heritage structures. Indeed, the historical evolution and actual behaviour of these assets are often unknown, no physical models are available, and the assessment must be performed only based on the tracking of a set of damage-sensitive features. Selecting the most representative state indicators to monitor and sampling them with an adequate number of records are therefore essential tasks to guarantee the successful performance of the damage detection strategy. Despite their relevance, these aspects have been frequently taken for granted and little attention has been paid to them by the scientific community working in the field of Structural Health Monitoring. The present paper aims to fill this gap by proposing a multistep strategy to drive the selection of meaningful pairs of correlated features in order to support the damage detection as a one-class classification problem. Numerical methods to reduce the number of necessary acquisitions and estimate the performance of approximation techniques are also provided. The analyses carried out to test and validate the proposed strategy exploit a dense dataset collected during the long-term monitoring of an outstanding heritage structure, i.e., the Church of ‘Santa Maria de Belém’ in Lisbon., This work was partially financed by FCT/MCTES through national funds (PIDDAC) under the R&D Unit Institute for Sustainability and Innovation in Structural Engineering (ISISE), under reference UIDB/04029/2020. The research was also financed by national funds through FCT— Foundation for Science and Technology, under grant agreement SFRH/BD/115188/2016 attributed to the first author, and by the National Operational Programme on Research and Innovation (Attraction and International Mobility) PON-AIM 2014-2020 Line 2, through the European Social Fund and the National Rotation Fund.

Signcryption schemes leveraging chaotic constructions have garnered significant research interest in recent years. These schemes have proffered practical solutions towards addressing the vast security vulnerabilities in Electronic Cash Systems (ECS). The schemes can seamlessly perform message confidentiality and authentication simultaneously. Still, their applications in emerging electronic cash platforms require a higher degree of complexity in design and robustness, especially as billions of online transactions are conducted globally. Consequently, several security issues arise from using open wireless channels for online business transactions. In order to guarantee the security of user information over these safety-limited channels, sophisticated security schemes are solely desired. However, the existing signcryption schemes cannot provide the required confidentiality and authentication for user information on these online platforms. Therefore, the need for certificateless group signcryption schemes (CGSS) becomes imperative. This paper presents an efficient electronic cash system based on CGSS using conformable chaotic maps (CCM). In our design, any group signcrypter would encrypt information/data with the group manager (GM) and send it to the verifier, who confirms the authenticity of the signcrypted information/data using the public criteria of the group. Additionally, the traceability, unforgeability, unlinkability, and robust security of the proposed CGSS-CCM ECS scheme have been built leveraging computationally difficult problems. Performance evaluation of the proposed CGSS-CCM ECS scheme shows that it is secure from the Indistinguishably Chosen Ciphertext Attack. Finally, the security analysis of the proposed technique shows high efficiency in security-vulnerable applications. Overall, the scheme gave superior security features compared to the existing methods in the preliminaries.

The latest cloud resource management research has revealed that virtual machine (VM) consolidation allows for effectively managing the physical resources of cloud data centers. However, a tremendous waste of power and physical resources has been pointed as one of the research challenges related to the development of new methods for VM management in a cloud data center in order to deliver a wide range of IT services to clients effectively. This paper investigates a problem of power-aware VM consolidation under dynamic workloads, uncertainty, and a changing number of VMs. For this purpose, the authors propose a dynamic VM management method based on a beam search that takes into account four types of resources (CPU, memory, network throughput, and storage throughput) and six quality metrics. Optimal beam search algorithm parameters for the defined problem are determined using a new power-aware integral estimation method. The SLA violation minimization allows significant improvement of SLA quality metrics, accompanied by the decreased number of VM migrations and slight deterioration in the power consumption. The proposed method is evaluated using common widespread hardware configurations and Bitbrains workload traces. The experiments show that the proposed approach can ensure the efficient use of cloud resources in terms of SLA violation and the number of VM migrations.

The automotive industry has been transformed through technological progress during the past decade. Vehicles are equipped with multiple computing devices that offer safety, driving assistance, or multimedia services. Despite these advancements, when an incident occurs, such as a car crash, the involved parties often do not take advantage of the technological capabilities of modern vehicles and attempt to assign liability for the incident to a specific vehicle based upon witness statements. In this paper, we propose a secure, decentralized, blockchain-based platform that can be employed to store encrypted position and velocity values for vehicles in a smart city environment. Such data can be decrypted when the need arises, either through the vehicle driver’s consent or through the consensus of different authorities. The proposed platform also offers an automated way to resolve disputes between involved parties. A simulation has been conducted upon a mobility traffic dataset for a typical day in the city of Cologne to assess the applicability of the proposed methodology to real-world scenarios and the infrastructure requirements that such an application would have.

Cooperative adaptive cruise control (CACC) has important significance for the development of the connected and automated vehicle (CAV) industry. The traditional proportional integral derivative (PID) platoon controller adjustment is not only time-consuming and laborious, but also unable to adapt to different working conditions. This paper proposes a learning control method for a vehicle platooning system using a deep deterministic policy gradient (DDPG)-based PID. The main contribution of this study is automating the PID weight tuning process by formulating this objective as a deep reinforcement learning (DRL) problem. The longitudinal control of the vehicle platooning is divided into upper and lower control structures. The upper-level controller based on the DDPG algorithm can adjust the current PID controller parameters. Through offline training and learning in a SUMO simulation software environment, the PID controller can adapt to different road and vehicular platooning acceleration and deceleration conditions. The lower-level controller controls the gas/brake pedal to accurately track the desired acceleration and speed. Based on the hardware-in-the-loop (HIL) simulation platform, the results show that in terms of the maximum speed error, for the DDPG-based PID controller this is 0.02–0.08 m/s less than for the conventional PID controller, with a maximum reduction of 5.48%. In addition, the maximum distance error of the DDPG-based PID controller is 0.77 m, which is 14.44% less than that of the conventional PID controller.

The automatic segmentation of the aorta can be extremely useful in clinical practice, allowing the diagnosis of numerous pathologies to be sped up, such as aneurysms and dissections, and allowing rapid reconstructive surgery, essential in saving patients’ lives. In recent years, the success of Deep Learning (DL)-based decision support systems has increased their popularity in the medical field. However, their effective application is often limited by the scarcity of training data. In fact, collecting large annotated datasets is usually difficult and expensive, especially in the biomedical domain. In this paper, an automatic method for aortic segmentation, based on 2D convolutional neural networks (CNNs), using 3D CT (computed axial tomography) scans as input is presented. For this purpose, a set of 153 CT images was collected and a semi-automated approach was used to obtain their 3D annotations at the voxel level. Although less accurate, the use of a semi-supervised labeling technique instead of a full supervision proved necessary to obtain enough data in a reasonable amount of time. The 3D volume was analyzed using three 2D segmentation networks, one for each of the three CT views (axial, coronal and sagittal). Two different network architectures, U-Net and LinkNet, were used and compared. The main advantages of the proposed method lie in its ability to work with a reduced number of data even with noisy targets. In addition, analyzing 3D scans based on 2D slices allows for them to be processed even with limited computing power. The results obtained are promising and show that the neural networks employed can provide accurate segmentation of the aorta.

This paper presents research on the effects of the multiple injection strategies on the combustion and emission characteristics of a two-stroke heavy-duty marine engine at full load. The ANSYS FLUENT simulation software was used to conduct three-dimensional simulations of the combustion process and emission formations inside the engine cylinder in both single- and double-injection modes to analyze the in-cylinder pressure, temperature, and emission characteristics. The simulation results were then compared and showed good agreement with the measured values reported in the engine’s sea-trial technical reports. The simulation results showed reductions in the in-cylinder pressure and temperature peaks by 6.42% and 12.76%, while NO and soot emissions were reduced up to 24.16% and 68%, respectively, in the double-injection mode in comparison with the single-injection mode. However, the double-injection strategy increased the CO2 emission (7.58%) and ISFOC (23.55%) compared to the single-injection. These are negative effects of the double-injection strategy on the engine that the operators need to take into consideration. The results were in line with the literature reviews and would be good material for operators who want to reduce the engine exhaust gas emission in order to meet the stricter IMO emission regulations.

The recent explosive growth in the number of smart technologies relying on data collected from sensors and processed with machine learning classifiers made the training data imbalance problem more visible than ever before. Class-imbalanced sets used to train models of various events of interest are among the main reasons for a smart technology to work incorrectly or even to completely fail. This paper presents an attempt to resolve the imbalance problem in sensor sequential (time-series) data through training data augmentation. An Unrolled Generative Adversarial Networks (Unrolled GAN)-powered framework is developed and successfully used to balance the training data of smartphone accelerometer and gyroscope sensors in different contexts of road surface monitoring. Experiments with other sensor data from an open data collection are also conducted. It is demonstrated that the proposed approach allows for improving the classification performance in the case of heavily imbalanced data (the F1 score increased from 0.69 to 0.72, p<0.01, in the presented case study). However, the effect is negligible in the case of slightly imbalanced or inadequate training sets. The latter determines the limitations of this study that would be resolved in future work aimed at incorporating mechanisms for assessing the training data quality into the proposed framework and improving its computational efficiency.

Sensor networks (SN) are increasingly used for the observation and monitoring of spatiotemporal phenomena and their dynamics such as pollution, noise and forest fires. In multisensory systems, a sensor node may be equipped with different sensing units to observe and detect several spatiotemporal phenomena at the same time. Simultaneous detection of different phenomena can be used to infer their spatial interactions over space and time. For this purpose, decentralized spatial computing approaches have shown their potential for effective reasoning on spatial phenomena within a sensor network. However, in most cases, spatial extents of continuous dynamic phenomena are uncertain, and their relations and interactions cannot be inferred by the existing approaches at the sensor node level. To address this limitation, in this paper, we propose and develop a decentralized fuzzy rule-based spatial reasoning approach to depict the spatial relations that hold between two evolving spatial phenomena with fuzzy boundaries. The proposed method benefits from a more adapted fuzzy-crisp representation of dynamic phenomena observed by SN where each vague phenomenon is composed of five distinguished zones including the kernel, conjecture and exterior zone and their boundaries. For each detected phenomenon, a sensor node will report one of these zones based on its location. Aggregation of the information reported from the sensor nodes allows reasoning on spatial relations between the observed phenomena and their evolution. Such spatial information provides users with more valuable near real-time information on the state of different phenomena that can be used for informed decision-making.

DEVS is a powerful formal language to describe discrete event systems in modeling and simulation areas and useful for component-based design. One of the advantages of component-based design is reusability. To reuse or share DEVS models developed by many other modelers, a system to systematically store and retrieve many DEVS models should be supported. However, to the best of our knowledge, there does not exist such a system. In this paper, we propose GO-DEVS (Graph/Ontology-represented DEVS storage and retrieval system) to store and retrieve DEVS models using graph and ontology representation. For effective model sharing, an ontology is introduced when a DEVS model is developed. To search for DEVS models in an effective and efficient way, we propose two types of queries, IO query and structure query, and provide a method to store and query DEVS models on an RDBMS. Finally, we experimentally show GO-DEVS can process the queries efficiently.

Forecasting uncertainties limit the development of photovoltaic (PV) power generation. New forecasting technologies are urgently needed to improve the accuracy of power generation forecasting. In this paper, a novel ultra-short-term PV power forecasting method is proposed based on a deep belief network (DBN)-based Takagi-Sugeno (T-S) fuzzy model. Firstly, the correlation analysis is used to filter redundant information. Furthermore, a T-S fuzzy model, which integrates fuzzy c-means (FCM) for the fuzzy division of input variables and DBN for fuzzy subsets forecasting, is developed. Finally, the proposed method is compared to a benchmark DBN method and the T-S fuzzy model in case studies. The numerical results show the feasibility and flexibility of the proposed ultra-short-term PV power forecasting approach.

In recent years the amount of malware spreading through the internet and infecting computers and other communication devices has tremendously increased. To date, countless techniques and methodologies have been proposed to detect and neutralize these malicious agents. However, as new and automated malware generation techniques emerge, a lot of malware continues to be produced, which can bypass some state-of-the-art malware detection methods. Therefore, there is a need for the classification and detection of these adversarial agents that can compromise the security of people, organizations, and countless other forms of digital assets. In this paper, we propose a spatial attention and convolutional neural network (SACNN) based on deep learning framework for image-based classification of 25 well-known malware families with and without class balancing. Performance was evaluated on the Malimg benchmark dataset using precision, recall, specificity, precision, and F1 score on which our proposed model with class balancing reached 97.42%, 97.95%, 97.33%, 97.11%, and 97.32%. We also conducted experiments on SACNN with class balancing on benign class, also produced above 97%. The results indicate that our proposed model can be used for image-based malware detection with high performance, despite being simpler as compared to other available solutions.

Network attacking using malware has become very popular on the Internet and in many other networks, namely Vehicular Ad-hoc Network (VANET) networks. It is required to have the model describing the malware spreading based on factors, which directly affect this process to limit its influences. In this paper, we propose a mathematical model called SEIR-S (Susceptible–Exposed–Infectious–Recovered–Susceptible) based on the characteristics of the VANET network and the well-known disease-spreading model SIR (Susceptible–Infectious–Recovered). We take into account possible behaviors of malware and provide the corresponding states to vehicles: Susceptible (S), Exposed (E), Infectious (I), Recovered (R). We evaluate the basic reproduction number R0 of the model and perform a stability analysis of the proposed model. The results show that, when R0 < 1, the malware spreading will gradually decrease, and, when R0 > 1, that spreading cannot be extinguished. We also point out the condition that we can control the endemic in the VANET network. In addition, the correctness of the proposed model is verified using both numerical analysis and agent-based simulation on NetLogo.

Edge Computing enables to perform measurement and cognitive decisions outside a central server by performing data storage, manipulation, and processing on the Internet of Things (IoT) node. Also, Artificial Intelligence (AI) and Machine Learning applications have become a rudimentary procedure in virtually every industrial or preliminary system. Consequently, the Raspberry Pi is adopted, which is a low-cost computing platform that is profitably applied in the field of IoT. As for the software part, among the plethora of Machine Learning (ML) paradigms reported in the literature, we identified Rulex, as a good ML platform, suitable to be implemented on the Raspberry Pi. In this paper, we present the porting of the Rulex ML platform on the board to perform ML forecasts in an IoT setup. Specifically, we explain the porting Rulex’s libraries on Windows 32 Bits, Ubuntu 64 Bits, and Raspbian 32 Bits. Therefore, with the aim of carrying out an in-depth verification of the application possibilities, we propose to perform forecasts on five unrelated datasets from five different applications, having varying sizes in terms of the number of records, skewness, and dimensionality. These include a small Urban Classification dataset, three larger datasets concerning Human Activity detection, a Biomedical dataset related to mental state, and a Vehicle Activity Recognition dataset. The overall accuracies for the forecasts performed are: 84.13%, 99.29% (for SVM), 95.47% (for SVM), and 95.27% (For KNN) respectively. Finally, an image-based gender classification dataset is employed to perform image classification on the Edge. Moreover, a novel image pre-processing Algorithm was developed that converts images into Time-series by relying on statistical contour-based detection techniques. Even though the dataset contains inconsistent and random images, in terms of subjects and settings, Rulex achieves an overall accuracy of 96.47% while competing with the literature which is dominated by forward-facing and mugshot images. Additionally, power consumption for the Raspberry Pi in a Client/Server setup was compared with an HP laptop, where the board takes more time, but consumes less energy for the same ML task.

Deep neural networks have achieved state-of-the-art performance in image classification. Due to this success, deep learning is now also being applied to other data modalities such as multispectral images, lidar and radar data. However, successfully training a deep neural network requires a large reddataset. Therefore, transitioning to a new sensor modality (e.g., from regular camera images to multispectral camera images) might result in a drop in performance, due to the limited availability of data in the new modality. This might hinder the adoption rate and time to market for new sensor technologies. In this paper, we present an approach to leverage the knowledge of a teacher network, that was trained using the original data modality, to improve the performance of a student network on a new data modality: a technique known in literature as knowledge distillation. By applying knowledge distillation to the problem of sensor transition, we can greatly speed up this process. We validate this approach using a multimodal version of the MNIST dataset. Especially when little data is available in the new modality (i.e., 10 images), training with additional teacher supervision results in increased performance, with the student network scoring a test set accuracy of 0.77, compared to an accuracy of 0.37 for the baseline. We also explore two extensions to the default method of knowledge distillation, which we evaluate on a multimodal version of the CIFAR-10 dataset: an annealing scheme for the hyperparameter α and selective knowledge distillation. Of these two, the first yields the best results. Choosing the optimal annealing scheme results in an increase in test set accuracy of 6%. Finally, we apply our method to the real-world use case of skin lesion classification.

The generation of the mix-based expansion of modern power grids has urged the utilization of digital infrastructures. The introduction of Substation Automation Systems (SAS), advanced networks and communication technologies have drastically increased the complexity of the power system, which could prone the entire power network to hackers. The exploitation of the cyber security vulnerabilities by an attacker may result in devastating consequences and can leave millions of people in severe power outage. To resolve this issue, this paper presents a network model developed in OPNET that has been subjected to various Denial of Service (DoS) attacks to demonstrate cyber security aspect of an international electrotechnical commission (IEC) 61850 based digital substations. The attack scenarios have exhibited significant increases in the system delay and the prevention of messages, i.e., Generic Object-Oriented Substation Events (GOOSE) and Sampled Measured Values (SMV), from being transmitted within an acceptable time frame. In addition to that, it may cause malfunction of the devices such as unresponsiveness of Intelligent Electronic Devices (IEDs), which could eventually lead to catastrophic scenarios, especially under different fault conditions. The simulation results of this work focus on the DoS attack made on SAS. A detailed set of rigorous case studies have been conducted to demonstrate the effects of these attacks.

Rice quality assessment is essential for meeting high-quality standards and consumer demands. However, challenges remain in developing cost-effective and rapid techniques to assess commercial rice grain quality traits. This paper presents the application of computer vision (CV) and machine learning (ML) to classify commercial rice samples based on dimensionless morphometric parameters and color parameters extracted using CV algorithms from digital images obtained from a smartphone camera. The artificial neural network (ANN) model was developed using nine morpho-colorimetric parameters to classify rice samples into 15 commercial rice types. Furthermore, the ANN models were deployed and evaluated on a different imaging system to simulate their practical applications under different conditions. Results showed that the best classification accuracy was obtained using the Bayesian Regularization (BR) algorithm of the ANN with ten hidden neurons at 91.6% (MSE = &lt, 0.01) and 88.5% (MSE = 0.01) for the training and testing stages, respectively, with an overall accuracy of 90.7% (Model 2). Deployment also showed high accuracy (93.9%) in the classification of the rice samples. The adoption by the industry of rapid, reliable, and accurate methods, such as those presented here, may allow the incorporation of different morpho-colorimetric traits in rice with consumer perception studies.

Automatic ECG signal characterization is of critical importance in patient monitoring and diagnosis. This process is computationally intensive, and low-power, online (real-time) solutions to this problem are of great interest. In this paper, we present a novel, dedicated hardware implementation of the ECG signal processing chain based on Hermite functions, aiming for real-time processing. Starting from 12-bit ADC samples of the ECG signal, the hardware implements filtering, peak and QRS detection, and least-squares Hermite polynomial fit on heartbeats. This hardware module can be used to compress ECG data or to perform beat classification. The hardware implementation has been validated on a Field Programmable Gate Array (FPGA). The implementation is generated using an algorithm-to-hardware compiler tool-chain and the resulting hardware is characterized using a low-cost off-the-shelf FPGA card. The single-beat best-fit computation latency when using six Hermite basis polynomials is under 1 s with a throughput of 3 beats/s and with an average power dissipation around 28 mW, demonstrating true real-time applicability.