The Journal of Supercomputing

Published by Springer Nature
Online ISSN: 1573-0484
Learn more about this page
Recent publications
Article
Driven by big data, neural networks evolve more complex and the computing capacity of a single machine is often difficult to meet the demand. Distributed deep learning technology has shown great performance superiority for handling this problem. However, a serious issue in this field is the existence of stragglers, which significantly restricts the performance of the whole system. It is an enormous challenge to fully exploit the computing capacity of the system based on parameter server architecture, especially in a heterogeneous environment. Motivated by this, we designed a method named EP4DDL to minimize the impact of the straggler problem by load balance technique. In a statistical view, the approach introduces a novel metric named performance variance to give a comprehensive inspection of stragglers and employs flexible parallelism techniques for each node. We verify the algorithm on standard benchmarks and demonstrate that it can reduce training time to 57.46%, 24.8%, and 11.5%, respectively, without accuracy loss compared with the FlexRR, Con-SGD, and Falcon.
 
Article
Human resource management is the cornerstone of enterprise success. In the process of enterprise management and control, the design of human resource management mode is a very important part of its management. Data mining technology provides a valuable and meaningful knowledge of extracting and mining big data. And such a technology will be an advantageous tool for human resource experts in the face of difficult and unknown talent screening. In the past, talent screening relied on many factors, such as experience, knowledge, performance and judgment ability. The screening criteria are no longer sufficient, because in this knowledge economy and business environment, the factors that facilitate someone to sit in a certain position today may not apply the next day, but in talent management, the definition of output is ensured that the right people are in the right jobs. Based on the above factors, how to select talents in the field of talent management and predict the possible future development of talents has become a challenge and problem for every organization. In this research, this research proposed data mining method with the decision tree technology to analyze data and find out the key factors that affect the on-the-job time through mining data. The results show that extended the application of data mining to the field of human resource management. Through the decision tree technology, this research contributions are companies can improve their recruitment methods and pay more attention to job seekers in the location. The policy of employee retention and employee recruitment will be improved. The key factors that affect the in-service time are predicted, and some key information that may affect the in-service time is obtained. This will help the company make correct decisions and effectively reduce the cost of company operations. This will help the company make correct decisions and effectively reduce the cost of company operations.
 
Article
Gene expression data play a significant role in the development of effective cancer diagnosis and prognosis techniques. However, many redundant, noisy, and irrelevant genes (features) are present in the data, which negatively affect the predictive accuracy of diagnosis and increase the computational burden. To overcome these challenges, a new hybrid filter/wrapper gene selection method, called mRMR-BAOAC-SA, is put forward in this article. The suggested method uses Minimum Redundancy Maximum Relevance (mRMR) as a first-stage filter to pick top-ranked genes. Then, Simulated Annealing (SA) and a crossover operator are introduced into Binary Arithmetic Optimization Algorithm (BAOA) to propose a novel hybrid wrapper feature selection method that aims to discover the smallest set of informative genes for classification purposes. BAOAC-SA is an enhanced version of the BAOA in which SA and crossover are used to help the algorithm in escaping local optima and enhancing its global search capabilities. The proposed method was evaluated on 10 well-known microarray datasets, and its results were compared to other current state-of-the-art gene selection methods. The experimental results show that the proposed approach has a better performance compared to the existing methods in terms of classification accuracy and the minimum number of selected genes.
 
Article
Identifying near-duplicate data can be applied to any type of content and has been widely used for increasing search engines' efficiency, detecting plagiarism or spam, etc. As a near-duplicate detection (NDD) method, sectional MinHash (S-MinHash) estimates the similarity between the text content in high accuracy by considering the section of every document's attributes with similarity estimation. However, due to the addition in computational complexity, it still has some performance issues such as being slow. The proposed sectional Min–Max Hash method aims to reduce the hashing time while preserving and improving the accuracy of detecting near-duplicate documents. We achieved this goal by combining S-MinHash with Min–Max Hash method. The results show that our new method reduces the hashing time and provides more speed due to using half of the random hash functions that S-MinHash needs to build up the signature matrix. Furthermore, we conducted experiments to compare our sectional Min–Max Hash with the baseline methods on the evaluated dataset and confirmed that in terms of the running time and algorithm's precision, the proposed method yields better results than the S-MinHash and other NDD techniques. Also, by assuming that we have two sections, as the best-case performance for sectional algorithms on the evaluated dataset, the error rate reduced significantly in the proposed method, and the F-score reached up to 99%.
 
Article
As the growing of data volumes due to the successive development of new mobile devices and the creation of new applications, the emergence of multi-access edge computing can successfully improve quality of service based on reduced latency and lower system energy consumption. The introduction of software-defined networking technologies in multi-access edge computing environments supports access to more network devices and enhances the scalability and service management flexibility of mobile edge computing environments. The limited nature of computing resources in mobile edge computing environments makes resource management a critical issue. Therefore, to minimize the energy consumption and latency of task execution in mobile edge computing environment, and to ensure reasonable resource allocation during task execution, a resource management strategy based on multi-objective optimization in edge computing environment is proposed. In this strategy, the overall energy consumption weighting and minimization problem is solved by optimizing the management of communication and computing resources, and an improved NSGA-II algorithm is proposed to rationally allocate communication and computational resources for each task. To deal with load imbalance caused by large traffic fluctuations in multi-access edge computing environments based on software-defined networks, in this paper, a load-balancing-oriented switch migration strategy is proposed in which a switch migration algorithm based on an improved ant colony algorithm is proposed to optimally select the switch migration process so that the static deployment of the controller adapts to the changing needs of dynamic flows in the network. Experimental results demonstrate that the proposed resource management strategy minimizes the latency and energy consumption during task execution and increases resource utilization and average throughput of servers. The proposed switch migration strategy can effectively achieve load balancing and reduce the response time.
 
Article
The Elliptic curve cryptosystem is a public-key cryptosystem that receives more focus in recent years due to its higher security with smaller key size when compared to RSA. Smartcards and other applications have highlighted the importance of security in resource-constrained situations. To meet the increasing need for speed in today’s applications, hardware acceleration with cryptographic algorithms is required. In this paper, we present a novel parallel architecture for elliptic curve scalar multiplication based on a modified Lopez-Dahab–Montgomery(LDM) algorithm, to reduce the total time delay for computing scalar multiplication. It comprises three main steps: affine to projective conversion, point addition and doubling in the main loop followed by reconversion to affine coordinate. The modified parallel algorithm with new inversion in the reconversion yields lesser clock cycle and total time delay compared to existing techniques in the literature for the National Institute of Standards and Technology recommended trinomial GF(2233) . Our proposed architecture implemented on Virtex4 and Virtex7 FPGA technologies, respectively, achieved a lesser clock cycle of 956, which yields a lesser delay of 20.025 and 8.22 μs. Compared with the state-of-the-art of existing techniques, two multiplications are reduced in the reconstruction process and our processor yields 18.29% and 27.21% increase in area-time performance in Virtex 4 and Virtex 7 devices, respectively.
 
Article
The h-extra edge-connectivity is an important parameter for the reliability evaluation and fault tolerance analysis of the easily scalable interconnection networks of parallel and distributed systems. The h-extra edge-connectivity of the topological structure of an interconnection network G, denoted by λh(G), is the minimum cardinality of a set of link malfunctions whose deletion disconnects G and each remaining component has at least h processors. In this paper, for the integer n≥3, we find that the h-extra edge-connectivity of n-dimensional pentanary cube (obtained by the n-th Cartesian product of K5), denoted by λh(K5n), presents a concentration behavior on the value 4×5n-1 (resp. 6×5n-1) for some exponentially large enough h: ⌈2×5n-13⌉≤h≤5n-1 (resp. ⌈4×5n-13⌉≤h≤2×5n-1). That is, for about 40.00 percent of 1≤h≤⌊5n/2⌋, the exact values of the h-extra edge-connectivity of n-dimensional pentanary cube are either 4×5n-1 or 6×5n-1.
 
Article
Early diagnosis and therapy are the most essential strategies to prevent deaths from diseases, such as cancer, brain tumors, and heart diseases. In this regard, information mining and artificial intelligence approaches have been valuable tools for providing useful data for early diagnosis. However, high-dimensional data can be challenging to examine, practically difficult to visualize, and costly to measure and store. Transferring a high-dimensional portrayal of the data to a lower-dimensional one without losing important information is the focal issue of dimensionality reduction. Therefore, in this study, dimensionality reduction-based medical data classification is presented. The proposed methodology consists of three modules: pre-processing, dimension reduction using an adaptive artificial flora (AAF) algorithm, and classification. The important features are selected using the AAF algorithm to reduce the dimension of the input data. From the results, a dimension-reduced dataset is obtained. The reduced data are then fed as input to the hybrid classifier. A hybrid support vector neural network is proposed for classification. Finally, the effectiveness of the proposed method is analyzed in terms of different metrics, namely accuracy, sensitivity, and specificity. The proposed method is implemented in MATLAB.
 
Article
Fog-integrated cloud (FiC) contains a fair amount of heterogeneity, leading to uncertainty in the resource provisioning. An admission control manager (ACM) is proposed, using a modified fuzzy inference system (FiS), to place a request based on the request’s parameters, e.g., CPU, memory, storage, and few categorical parameters, e.g., job priority and time sensitivity. The ACM considers the extended three-layer architecture of FiC. FiC nodes are classified into three computing nodes: fog node, aggregated fog node, and cloud node using modified FiS model. For performance study, extensive simulation experiments have been carried out on real Google trace. Different batches on the number of relevant rules are created and compared on metrics of job execution time, memory overhead, accuracy, and hit ratio with the modified rules. The proposed work has also been compared with the state of the art. The results have been encouraging and exhibit the benefits of the proposed model apart from being it lightweight with reduced number of rules, especially suited for the FiC.
 
Article
In recent years, combinatorial optimization has been widely studied. The existing optimization solutions are prone to fall into local optimal solutions and have a lower probability of obtaining global optimal solutions. Quantum approximate optimization algorithm (QAOA) is an effective algorithm that can obtain the optimal solution with high probability. In this paper, the problem Hamiltonian is obtained by summing the problem function and the deformed constraints. Through theoretical formula derivation, the problem Hamiltonian is transformed into the Ising model. The performance of the experimental result under different optimizers and asynchronous lengths is verified on pyQPanda. The experimental results show that when using the problem Hamiltonian method set in this paper, the probability of obtaining the optimal solution is 99.59%. Compared with other methods, the proposed method can alleviate the risk of falling into local optimal solutions and obtain the global optimal solution with a higher probability.
 
Article
Due to the increase and complexity of computer systems, reducing the overhead of fault tolerance techniques has become important in recent years. One technique in fault tolerance is checkpointing, which saves a snapshot with the information that has been computed up to a specific moment, suspending the execution of the application, consuming I/O resources and network bandwidth. Characterizing the files that are generated when performing the checkpoint of a parallel application is useful to determine the resources consumed and their impact on the I/O system. It is also important to characterize the application that performs checkpoints, and one of these characteristics is whether the application does I/O. In this paper, we present a model of checkpoint behavior for parallel applications that performs I/O; this depends on the application and on other factors such as the number of processes, the mapping of processes and the type of I/O used. These characteristics will also influence scalability, the resources consumed and their impact on the IO system. Our model describes the behavior of the checkpoint size based on the characteristics of the system and the type (or model) of I/O used, such as the number I/O aggregator processes, the buffering size utilized by the two-phase I/O optimization technique and components of collective file I/O operations. The BT benchmark and FLASH I/O are analyzed under different configurations of aggregator processes and buffer size to explain our approach. The model can be useful when selecting what type of checkpoint configuration is more appropriate according to the applications’ characteristics and resources available. Thus, the user will be able to know how much storage space the checkpoint consumes and how much the application consumes, in order to establish policies that help improve the distribution of resources.
 
Article
Given a large data graph, trimming techniques can reduce the search space by removing vertices without outgoing edges. One application is to speed up the parallel decomposition of graphs into strongly connected components (SCC decomposition), which is a fundamental step for analyzing graphs. We observe that graph trimming is essentially a kind of arc-consistency problem, and AC-3, AC-4, and AC-6 are the most relevant arc-consistency algorithms for application to graph trimming. The existing parallel graph trimming methods require worst-case O(nm)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(nm)$$\end{document} time and worst-case O(n)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(n)$$\end{document} space for graphs with n vertices and m edges. We call these parallel AC-3-based as they are much like the AC-3 algorithm. In this work, we propose AC-4-based and AC-6-based trimming methods. That is, AC-4-based trimming has an improved worst-case time of O(n+m)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(n+m)$$\end{document} but requires worst-case space of O(n+m)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(n+m)$$\end{document}; compared with AC-4-based trimming, AC-6-based has the same worst-case time of O(n+m)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(n+m)$$\end{document} but an improved worst-case space of O(n)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathcal O(n)$$\end{document}. We parallelize the AC-4-based and AC-6-based algorithms to be suitable for shared-memory multi-core machines. The algorithms are designed to minimize synchronization overhead. For these algorithms, we also prove the correctness and analyze time complexities with the work-depth model. In experiments, we compare these three parallel trimming algorithms over a variety of real and synthetic graphs on a multi-core machine, where each core corresponds to a worker. Specifically, for the maximum number of traversed edges per worker by using 16 workers, AC-3-based traverses up to 58.3 and 36.5 times more edges than AC-6-based trimming and AC-4-based trimming, respectively. That is, AC-6-based trimming traverses much fewer edges than other methods, which is meaningful especially for implicit graphs. In particular, for the practical running time, AC-6-based trimming achieves high speedups over graphs with a large portion of trimmable vertices.
 
Article
In this study, we have developed a set of virtual reality (VR) human–robot interaction technology acceptance model for learning direct current and alternating current, aiming to use VR technology to immerse students in the generation, existence, and flow of electricity. We hope that using VR to transform abstract physical concepts into tangible objects will help students learn and comprehend abstract electrical concepts. The VR technology acceptance model was developed using the Unity 3D game kit to be accessed using the HTC Vive VR headset. The scene models, characters, and objects were created using Autodesk 3DS Max and Autodesk Maya, and the 2D graphics were processed in Adobe Photoshop. The results were evaluated using four metrics for our technology acceptance model. The four metrics include the content, design, interface and media content, and practical requirements. The average score of the content is 4.73. The average score of the design is 4.12. The average score of the interface and media content is 4.34. The average score of the practical requirements is 3.72. All the items on the effectiveness questionnaire of the technology acceptance model had average scores in the range 4.25–4.75. Therefore, all teachers were strongly satisfied with the trial teaching activity. The average score of each statement ranged within 3.58–4.03 for the satisfaction with the teaching material contents. Hence, the students were somewhat satisfied with this teaching activity. The average score of each statement ranged from 3.43 to 4.96 for the satisfaction with the implementation of the technology acceptance model. This result shows that the respondents were generally satisfied with the learning outcomes associated with these materials. The average score per question in this questionnaire was 3.92, and most of the questions have an average score greater than 3.8 for the feedback pertaining to satisfaction with the teaching material contents. In summary, a deeply immersive and interactive game was created using tactile somatosensory devices and VR that aim to utilize and enhance the fun and benefits associated with learning from games.
 
Article
Epistasis can be defined as the statistical interaction of genes during the expression of a phenotype. It is believed that it plays a fundamental role in gene expression, as individual genetic variants have reported a very small increase in disease risk in previous Genome-Wide Association Studies. The most successful approach to epistasis detection is the exhaustive method, although its exponential time complexity requires a highly parallel implementation in order to be used. This work presents Fiuncho, a program that exploits all levels of parallelism present in x86_64 CPU clusters in order to mitigate the complexity of this approach. It supports epistasis interactions of any order, and when compared with other exhaustive methods, it is on average 358, 7 and 3 times faster than MDR, MPI3SNP and BitEpi, respectively.
 
Article
Execution of multiple applications on Multi-Processor System-on-Chips (MPSoCs) significantly boosts performance and energy efficiency. Although various researchers have suggested Network-on-Chip (NoC) architectures for MPSoCs, the problem still needs more investigations for the case of multi-application MPSoCs. In this paper, we propose a fully automated synthesis flow in five steps for the design of custom NoC fabrics for multi-application MPSoCs. The steps include: preprocessing, core to router allocation, voltage island merging, floorplanning, and router to router connection. The proposed flow finds design solutions that satisfy the performance, bandwidth, and power constraints of all input applications. If the user decides, the proposed synthesis adds network-level reconfiguration to improve the efficiency of the obtained design solutions. With the reconfiguration option, the proposed flow comes up with adaptive NoC architectures that satisfy each application’s communication requirements while power-gate idle resources, e.g., router ports and links. If reconfiguration option is not set by the user, the proposed flow considers the top communication requirements among the applications in finding design solutions. We have used the proposed synthesis flow to design custom NoCs for several combined graphs of real-world applications and synthetic graphs. Results show that the reconfiguration option can save up to 98% in the energy-delay product (EDP) of the ultimate designs.
 
Article
Complex system theory is increasingly applied to develop control protocols for distributed computational and networking resources. The paper deals with the important subproblem of finding complex connected structures having excellent navigability properties using limited computational resources. Recently, the two-dimensional hyperbolic space turned out to be an efficient geometry for generative models of complex networks. The networks generated using the hyperbolic metric space share their basic structural properties (like small diameter or scale-free degree distribution) with several real networks. In the paper, a new model is proposed for generating navigation trees for complex networks embedded in the two-dimensional hyperbolic plane. The generative model is not based on known hyperbolic network models: the trees are not inferred from the existing links of any network; they are generated from scratch instead and based purely on the hyperbolic coordinates of nodes. We show that these hyperbolic trees have scale-free degree distributions and are present to a large extent both in synthetic hyperbolic complex networks and real ones (Internet autonomous system topology, US flight network) embedded in the hyperbolic plane. As the main result, we show that routing on the generated hyperbolic trees is optimal in terms of total memory usage of forwarding tables.
 
Article
Network reconfiguration is an important means of improving network invulnerability. However, most existing network reconfiguration methods fail to consider node importance, edge importance, and hierarchical characteristics, and the local and global information of command and control (C2) networks are difficult to satisfy comprehensively. Therefore, this study designed a hierarchy-entropy-based method for reconfiguring C2 networks. By combining hierarchical and operational link entropy, the probability of inter-node edge reconfiguration based on hierarchy entropy is proposed. Additionally, methods for calculating the node level-up, cross-level, and swap degrees, and a portfolio reconfiguration strategy are proposed. Finally, to validate the proposed method, a case study was simulated, and the repair probability, adjustable parameters, and reconfiguration effects of the different reconfiguration methods and modes were determined. The comparison results demonstrate that the proposed algorithm improves the reconfiguration effect and reduces the reconfiguration cost.
 
Article
Task graphs provide a simple way to describe scientific workflows (sets of tasks with dependencies) that can be executed on both HPC clusters and in the cloud. An important aspect of executing such graphs is the used scheduling algorithm. Many scheduling heuristics have been proposed in existing works; nevertheless, they are often tested in oversimplified environments. We provide an extensible simulation environment designed for prototyping and benchmarking task schedulers, which contains implementations of various scheduling algorithms and is open-sourced, in order to be fully reproducible. We use this environment to perform a comprehensive analysis of workflow scheduling algorithms with a focus on quantifying the effect of scheduling challenges that have so far been mostly neglected, such as delays between scheduler invocations or partially unknown task durations. Our results indicate that network models used by many previous works might produce results that are off by an order of magnitude in comparison to a more realistic model. Additionally, we show that certain implementation details of scheduling algorithms which are often neglected can have a large effect on the scheduler’s performance, and they should thus be described in great detail to enable proper evaluation.
 
Article
The Goore Game (GG) is a model for collective decision-making under uncertainty, which can be used as a tool for stochastic optimization of a discrete variable function. The Goore Game has a fascinating property that can be resolved in an entirely distributed manner with no intercommunication between the players. In this paper, we introduce a new model called Cellular Goore Game (CGG). CGG is a network of Goore Games in which, at any time, every node (or node in a subset of the nodes) in the network plays the role of a referee that participates in a GG with its neighboring players (voters). Like GG, each player independently selects its optimal action between two available actions based on their gains and losses received from its adjacent referees. Players in CGG know nothing about how other players are playing or even how/why they are rewarded/penalized by the voters. CGG may be used for modeling systems that can be described as massive collections of simple objects interacting locally with each other. Through simulations, the behavior of CGG for different networks of players/voters is studied. This paper presents a novel CGG-based approach to efficiently solve the Quality-of-Service (QoS) control for clustered WSNs to show the potential of CGG. Also, a CGG-based QoS control algorithm for WSNs with multiple sinks is proposed that dynamically adjusts the number of active sensors during WSN operation. Several experiments have been conducted to evaluate the performance of these algorithms. The obtained results show that the proposed CGG-based algorithms are superior to the existing algorithms in terms of the QoS control performance metrics.
 
Article
Mobile Edge Computing (MEC) provides a new opportunity to reduce the latency of IoT applications significantly. It does so by offloading computation-intensive tasks in applications from IoT devices to mobile edges, which are located N-close proximity to the IoT devices. However, the prior researches focus on supporting computation offloading for a specific type of applications. Meanwhile, making multi-task and multi-server offloading decisions in highly complex and dynamic MEC environments remains intractable. To address this problem, this paper proposes a novel approach called MultiOff. First, we propose a generic program structure that supports on-demand computation offloading. Applications conforming to this structure can extract the flowcharts of program fragments via code analysis. Second, a novel cost-efficient offloading strategy based on a Multi-task Particle Swarm Optimization algorithm using the Genetic Algorithm operators (MPSO-GA) is proposed. MPSO-GA makes offloading decisions by analyzing program fragment flowcharts and context. Finally, each application can be offloaded at the granularity of services with the offloading scheme, minimizing the system cost while satisfying the deadline constraint for each application. We evaluate MultiOff on several real-world applications and the experimental results show that MultiOff can support computation offloading for different types of applications at the fine-grained granularity of services. Moreover, MPSO-GA can save about 2.11–17.51%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} system cost compared with other classical methods while meeting time constraints.
 
Article
In the era of the Internet of Things (IoT), the volume of data is increasing immensely causing rapid growth in network data communication and data congestion. Computational offloading thus becomes a crucial and imperative action in terms of delay-sensitive task completion and data processing for the resource constraint end-users. Nowadays fog computing, as a complement of cloud computing, has emerged a well-known concept in terms of enhancing data processing capability as well as energy conservation in low-powered networks. In this paper, we consider a heterogeneous fog-cloud network architecture where the data processing is performed on the local or remote computing device by adopting a binary offloading policy. Based on the proposed system model, we calculate the total delay and energy consumption of data processing throughout the network and formulate a mixed-integer optimization problem to jointly optimize the offloading decision and bandwidth allocation. In order to solve such an NP-hard problem, we have proposed a deep-learning-based binary offloading strategy that employs multiple parallel deep neural networks (DNNs) to make offloading decisions. Such offloading decisions are subsequently placed in a relay memory system to train and test all DNNs. Simulation results show a near-optimal performance of the proposed offloading strategy while remarkably maintaining the quality of service by decreasing overall delay and energy consumption.
 
Article
In the past decade, social media networks have received much attention among ordinary people, agencies, and research scholars. Twitter is one of the fastest-growing social media tools. By means of the Twitter application on smartphones, users are able to immediately report events happening around them on a real-time basis. The information disseminated by millions of active users every day generates a new version of a dynamic database that contains information about various topics. Twitter data can be utilized as a major traffic data source along with conventional sensors. In this aspect, this paper presents a novel firefly algorithm-based feature selection with a deep learning model for traffic flow analysis (FFAFS-DLTFA) using Twitter data. The goal of FFAFS-DLTFA is to determine the class labels for tweets as relevant to traffic events. The proposed FFAFS-DLTFA encompasses several processes, such as preprocessing, feature extraction, feature selection, and classification. Primarily, tweets are preprocessed in several ways, such as tokenization, removal of stop words, and stemming. At the same time, three types of embedding vectors, unigram, bigram, and POS features, are used. In addition, the firefly algorithm (FFA) is applied for the optimal selection of feature subsets. Finally, a deep neural network (DNN) model is applied for the identification of tweets into three classes, namely, positive, neutral, and negative. The performance validation of FFAFS-DLTFA takes place using the benchmark Kaggle repository, and the results are inspected under different aspects. The experimental values demonstrate the better performance of FFAFS-DLTFA on the other techniques with the maximum accuracy of 98.83%.
 
Components of Google AIY Voice Kit
Process of Google AIY Voice Kit [41]
Architecture of Google AIY Voice Kit
Algorithm of SR-PII System
Article
Currently, many smart speakers, even social robots, appear on the market to help people's lives become more convenient. Usually, people use smart speakers to check their daily schedule or control home appliances in their house. Many social robots also include smart speakers. They have the common property of being used in voice control machines. Regardless of where the smart speaker is installed and used, when people start a conversation with voice equipment, a security or privacy risk is exposed. Hence, we want to build a speech recognition (SR) that contains the privacy identification information (PII) system in this paper. We call this the SR-PII system. We used a Google Artificial-Intelligence-Yourself (AIY) Voice Kit released from Google to build a simple, smart dialog speaker and included our SR-PII system. In our experiments, we test SR accuracy and the reliability of privacy settings in three environments (quiet, noise, and playing music). We also examine the cloud response and speaker response times during our experiments. The results show that the speaker response is approximately 3.74 s in the cloud environment and approximately 9.04 s from the speaker. We also showed the response accuracy of the speaker, which successfully prevented personal information with the SR-PII system in three environments. The speaker has a response mean time of approximately 8.86 s with 93% mean accuracy in a quiet room, approximately 9.18 s with 89% mean accuracy in a noisy environment, and approximately 9.62 s with 90% mean accuracy in an environment that plays music. We conclude that the SR-PII system can secure private information and that the most important factor affecting the response speed of the speaker is the network connection status. We hope that people can, through our experiments, have some guidelines in building social robots and installing the SR-PII system to protect users’ personal identification information.
 
Article
The Jacobi iterative algorithm has the characteristic of low computational load, and multiple components of the solution can be solved independently. This paper applies these characteristics to the ternary optical computer, which can be used for parallel optimization because it has a large number of data bits and reconfigurable processor bits. Therefore, a new parallel design scheme is constructed to solve the problem of slow efficiency in solving large linear equations. And the elaborate experiment is used to verify. The experimental method is to simulate the calculation on the ternary optical computer experimental platform. Then, the resource consumption is numerically calculated and summarized to measure the feasibility of the parallel design. Eventually, the results show that the parallel design has obvious advantages in computing speed. The Jacobi iterative algorithm is optimized in parallel on ternary optical processor for the first time. There are two parallel highlights of the scheme. First, the n components are calculated in full parallel. Second, the modified signed-digit (MSD) multiplier based on the minimum module and one-step MSD adder are used to calculate each component to eliminate the impact of large amount of data on calculation time. The research provides a new method for fast solution of large linear equations.
 
Article
Due to the increase in electronic documents containing medical information, the search for specific information is often complex and time-consuming. This has prompted the development of new tools designed to address this issue. Automated visual question/answer (VQA) systems are becoming more challenging to develop. These are computer programs that take images and questions as input and then combine all inputs to generate text-based answers. Due to the enormous amount of question and the limited number of specialists, many issues stay unanswered. It’s possible to solve this problem by using automatic question classifiers that guide queries to experts based on their subject preferences. For these purposes, we propose a VQA approach based on a hybrid deep learning model. The model consists of three steps: (1) the classification of medical questions based on a BERT model; (2) image and text feature extraction using a deep learning model, more specifically the extraction of medical image features by a hybrid deep learning model; and (3) text feature extraction using a Bi-LSTM model. Finally, to predict the appropriate answer, our approach uses a KNN model. Additionally, this study examines the influence of the Adam, AdaGrad, Stochastic gradient descent and RMS Prop optimization techniques on the performance of the network. As a consequence of the studies, it was shown that Adam and SGD optimization algorithms consistently produced higher outcomes. Experiments using the ImageCLEF 2019 dataset revealed that the suggested method increases BLEU and WBSS values considerably.
 
Article
Detection of the selfish node in a delay tolerant network (DTN) can sharply reduce the loss incurred in a network. The algorithm's current pedigree mainly focuses on the rely on nodes, records, and delivery performance. The community structure and social aspects have been overlooked. Analysis of individual and social tie preferences results in an extensive detection time and increases communication overhead. In this article, a heterogeneous DTN topology with high-power stationary nodes and mobile nodes on Manhattan's accurate map is designed. With the increasing complexity of social ties and the diversified nature of topology structure, there need for a method that can effectively capture the essence within the speculated time. In this article, a novel deep autoencoder-based nonnegative matrix factorization (DANMF) is proposed for DTN topology. The topology of social ties projected onto low-dimensional space leads to effective cluster formation. DANMF automatically learns an appropriate nonlinear mapping function by utilizing the features of data. Also, the inherent structure of the deep autoencoder is nonlinear and has strong generalization. The membership matrices extracted from the DANMF are used to design the weighted cumulative social tie that eventually, along with the residual energy, is used to detect the network's selfish node. The testing of the designed model is carried out on the real dataset of MIT reality. The proficiency of the developed algorithm has been well tested and proved at every step. The methods employed for social tie extraction are NMF and DANMF. The methodology is rigorously experimented on various scenarios and has improved around 80% in the worst-case scenario of 40% nodes turning selfish. A comprehensive comparison is made with the other existing state-of-the-art methods which are also incentive-based approaches. The developed method has outperformed and has shown the supremacy of the current methods to capture the latent, hidden structure of the social tie.
 
System Model of the EP2LBS Scheme
a Running time of the LBS server b Running time of the user c Total running time of the EP2LBS scheme vs. existing LBS schemes. (†\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\dagger $$\end{document} is standard deviation of the execution time.)
Article
Mobile users frequently change their location and often desire to avail of location-based services (LBS). LBS server provides services to users at the service charge. The user queries the LBS server for services, and the LBS server replies queries’ answer with the associated fee. This exchange may breach the user’s privacy. Users’ query privacy and LBS server services’ privacy is a challenging issue. Many privacy-preserving LBS schemes have been proposed, such as trusted third party, homomorphic encryption, and private information retrieval. These existing schemes mostly suffer from poor efficiency and privacy issue. We propose an efficient privacy-preserving scheme for location-based services (EP2LBS) using a lattice-based oblivious transfer protocol. The proposed EP2LBS scheme’s security depends on the combination of decisional ring-learning with errors assumption and perfect secrecy assumption. This enables the EP2LBS scheme to preserve the user’s query privacy and LBS server’s services privacy. The theoretical and experimental results show that the EP2LBS scheme requires lower communication and computation costs at server and user as compared to the current-state-of-the-art schemes.
 
Article
The Internet of Medical Things (IoMT) is a bionetwork of allied medical devices, sensors, wearable biosensor devices, etc. It is gradually reforming the healthcare industry by leveraging its capabilities to improve personalized healthcare services by enabling seamless communication of medical data. IoMT facilitates prompt emergency responses and provides improved quality of medical services with minimum cost. With the advancement of modern technology, progressively ubiquitous medical devices raise critical security and data privacy concerns through resource constraints and open connectivity. Vulnerabilities in IoMT devices allow unauthorized access for potential entry into healthcare and sensitive personal data. In addition, the patient may experience severe physical damage with the attack on IoMT devices. To provide security to IoMT devices and privacy to patient data, we have proposed a novel IoMT framework with the hybridization of Bayesian optimization and extreme learning machine (ELM). The proposed model derives encouraging performance with enhanced accuracy in decision-making process compared to similar state-of-the-art methods.
 
Article
Data deduplication is a process that gets rid of excessive duplicates of data and minimizes the storage capacity to a large extent. This process mainly optimizes redundancies without compromising the data fidelity or integrity. However, the major challenge faced by most data deduplication systems is secure cloud storage. Cloud computing relies on the ability and security of all information. In the case of distributed storage, data protection and security are critical. This paper presents a Secure Cloud Framework for owners to effectively handle cloud-based information and provide high security for information (SCF). Weaknesses, Cross-Site Scripting (XSS), SQL perfusion, adverse processing, and wrapping are all examples of significant attacks in the cloud. This paper proposes an improved Secure File Deduplication Avoidance (SFDA) algorithm for block-level deduplication and security. The deduplication process allows cloud customers to adequately manage the distributed storage space by avoiding redundant information and saving transfer speed. A deep learning classifier is used to distinguish the familiar and unfamiliar data. A dynamic perfect hashing scheme is used in the SFDA approach to perform convergent encryption and offer secure storage. The Chaotic krill herd optimization (CKHO) algorithm is used for the optimal secret key generation process of the Advanced Encryption Standard (AES) algorithm. In this way, the unfamiliar data are encrypted one more time and stored in the cloud. The efficiency of the results is demonstrated via the experiments conducted in terms of computational cost, communication overhead, deduplication rate, and attack level. For file sizes of 8 MB, 16 MB, 32 MB, and 64 MB, the proposed methodology yields a deduplication rate of 53%, 62%, 54%, and 44%, respectively. The dynamic perfect hashing and the optimal key generation using the CKHO algorithm minimizes the data update time and the time taken to update a total of 1024 MB data is 341.5 ms. The improved SFDA algorithm's optimal key selection approach reduces the impact of an attack by up to 12% for a data size of 50 MB, whereas the existing system is mostly impacted by data size, and its attack level rises by up to 19 percent for the same data size.
 
Article
Power consumption is likely to remain a significant concern for exascale performance in the foreseeable future. In addition, graphics processing units (GPUs) have become an accepted architectural feature for exascale computing due to their scalable performance and power efficiency. In a recent study, we found that we can achieve a reasonable amount of power and energy savings based on the selection of algorithms. In this research, we suggest that we can save more power and energy by varying the block size in the kernel configuration . We show that we may attain more savings by selecting the optimum block size while executing the workload. We investigated two kernels on NVIDIA Tesla K40 GPU, a Bitonic Mergesort and Vector Addition kernels, to study the effect of varying block sizes on GPU power and energy consumption. The study should offer insights for upcoming exascale systems in terms of power and energy efficiency.
 
Article
In real-time rendering, a 3D scene is modelled with meshes of triangles that the GPU projects to the screen. They are discretized by sampling each triangle at regular space intervals to generate fragments which are then added texture and lighting effects by a shader program. Realistic scenes require detailed geometric models, complex shaders, high-resolution displays and high screen refreshing rates, which all come at a great compute time and energy cost. This cost is often dominated by the fragment shader, which runs for each sampled fragment. Conventional GPUs sample the triangles once per pixel; however, there are many screen regions containing low variation that produce identical fragments and could be sampled at lower than pixel-rate with no loss in quality. Additionally, as temporal frame coherence makes consecutive frames very similar, such variations are usually maintained from frame to frame. This work proposes Dynamic Sampling Rate (DSR), a novel hardware mechanism to reduce redundancy and improve the energy efficiency in graphics applications. DSR analyzes the spatial frequencies of the scene once it has been rendered. Then, it leverages the temporal coherence in consecutive frames to decide, for each region of the screen, the lowest sampling rate to employ in the next frame that maintains image quality. We evaluate the performance of a state-of-the-art mobile GPU architecture extended with DSR for a wide variety of applications. Experimental results show that DSR is able to remove most of the redundancy inherent in the color computations at fragment granularity, which brings average speedups of 1.68x and energy savings of 40%.
 
Article
Swarm-Intelligence (SI), the collective behavior of decentralized and self-organized system, is used to efficiently carry out practical missions in various environments. To guarantee the performance of swarm, it is highly important that each object operates as an individual system while the devices are organized as simple as possible. This paper proposes an efficient, scalable, and practical swarming system using gas detection device. Each object of the proposed system has multiple sensors and detects gas in real time. To let the objects move toward gas rich spot, we propose two approaches for system design, vector-sum based, and Reinforcement Learning (RL) based. We firstly introduce our deterministic vector-sum-based approach and address the RL-based approach to extend the applicability and flexibility of the system. Through system performance evaluation, we validated that each object with a simple device configuration performs its mission perfectly in various environments.
 
Article
In this work, we propose a multi-tier architectural model to separate functionality and security concerns for distributed cyber-physical systems. On the line of distributed computing, such systems require the identification of leaders for distribution of work, aggregation of results, etc. Further, we propose a fault-tolerant leader election algorithm that can independently elect the functionality and security leaders. The proposed election algorithm identifies a list of potential leader capable nodes to reduce the leader election overhead. It keeps identifying the highest potential node as the leader, whenever needed, including the situation when one has failed. We also explain the proposed architecture and its management method through a case study. Further, we perform several experiments to evaluate the system performance. The experimental results show that the proposed architectural model improves the system performance in terms of latency, average response time, and the number of real-time tasks completed within the deadline.
 
Article
Aspect-level sentiment classification has been widely used by researchers as a fine-grained sentiment classification task to predict the sentiment polarity of specific aspect words in a given sentence. Previous studies have shown relatively good experimental results using graph convolutional networks, so more and more approaches are beginning to exploit sentence structure information for this task. However, these methods do not link aspect word and context well. To address this problem, we propose a method that utilizes a hierarchical multi-head attention mechanism and a graph convolutional network (MHAGCN). It fully considers syntactic dependencies and combines semantic information to achieve interaction between aspect words and context. To fully validate the effectiveness of the method proposed in this paper, we conduct extensive experiments on three benchmark datasets, which, according to the experimental results, show that the method outperforms current methods.
 
Article
DNA sequencing is one of the important sub-disciplines of bioinformatics, which has various applications in medicine, history, demography, and archaeology. De novo sequencing is the most challenging problem in this field. De novo sequencing is used for recognizing a new genome and for sequencing unknown parts of the genome such as in cancer cells. For assembling the genome, first, small fragments of the genome (called reads) that are located randomly on the genome are sequenced by the sequencing machine. Then, they are sent to the processing machine to be aligned on the genome. To sequence the whole genome, the reads must cover it entirely. The minimum number of reads to cover the genome is given by the Lander–Waterman's coverage bound. In this paper, we generalize the later scheme to de novo sequencing and reduce the total number of required bases by Lander–Waterman's coverage bound. We investigate the performance of the scheme such as the longest generated contig length, the execution time of the algorithm, different read lengths, and probability of error in the genome assembly. The results show the computational complexity and execution time of the algorithm in parallel on human genome with length 50,000 bases. We also show that the proposed method can generate contigs with 90 percent genome length.
 
Article
  • Ram KumarRam Kumar
  • S. C. SharmaS. C. Sharma
Query expansion is an important approach utilized to improve the efficiency of data retrieval tasks. Numerous works are carried out by the researchers to generate fair constructive results; however, they do not provide acceptable results for all kinds of queries particularly phrase and individual queries. The utilization of identical data sources and weighting strategies for expanding such terms are the major cause of this issue which leads the model unable to capture the comprehensive relationship between the query terms. In order to tackle this issue, we developed a novel approach for query expansion technique to analyze the different data sources namely WordNet, Wikipedia, and Text REtrieval Conference. This paper presents an Improved Aquila Optimization-based COOT(IAOCOOT) algorithm for query expansion which retrieves the semantic aspects that match the query term. The semantic heterogeneity associated with document retrieval mainly impacts the relevance matching between the query and the document. The main cause of this issue is that the similarity among the words is not evaluated correctly. To overcome this problem, we are using a Modified Needleman Wunsch algorithm algorithm to deal with the problems of uncertainty, imprecision in the information retrieval process, and semantic ambiguity of indexed terms in both the local and global perspectives. The k most similar word is determined and returned from a candidate set through the top-k words selection technique and it is widely utilized in different tasks. The proposed IAOCOOT model is evaluated using different standard Information Retrieval performance metrics to compute the validity of the proposed work by comparing it with other state-of-art techniques.
 
Article
  • Heena WadhwaHeena Wadhwa
  • Rajni AronRajni Aron
The fog-assisted cloud computing gives better quality of service (QoS) to Internet of things (IoT) applications. However, the large quantity of data transmitted by the IoT devices results in the overhead of bandwidth and increased delay. Moreover, large amounts of data transmission generate resource management issues and decrease the system’s throughput. This paper proposes the optimized task scheduling and preemption (OSCAR) model to overcome the limitations and improve the QoS. The dataset used for the study is a real-time crowd-based dataset which provides task information. The processes involved in this paper are as follows: (i) Initially, the tasks from the IoT devices are clustered based on the priority and deadline by implementing expectation–maximization (EM) clustering to decrease the computational complexity and bandwidth overhead. (ii) The clustered tasks are then scheduled by implementing a modified heap-based optimizer based on the QoS and service level agreement (SLA) constraints. (iii) Distributed resource management is performed by allocating resources to the tasks based on multiple constraints. The categorical deep Q network is the deep reinforcement learning model is implemented for this purpose. The dynamic nature of tasks from the IoT devices is addressed by performing preemption of tasks using the ranking method, where the tasks with higher priority, with a short deadline replaces less priority task by moving it into the waiting queue. The proposed model is experimented with in the iFogsim simulation tool and evaluated in terms of average response time, loss ratio, resource utilization, average makespan time, queuing waiting time, percentage of tasks satisfying the deadline and throughput. The proposed OSCAR model outperforms the existing model in achieving the QoS and SLA with maximal throughput and reduced response time.
 
Article
Producing a large family of resource-constrained multi-processing systems on chips (MPSoC) is challenging, and the existing techniques are generally geared toward a single product. When they are leveraged for a variety of products, they are expensive and complex. Further in the industry, a considerable lack of analysis support at the architectural level induces a strong dependency on the experiences and preferences of the designer. This paper proposes a formal foundation and analysis of MPSoC product lines based on a featured transition system (FTS) to express the variety of products. First, features diagrams are selected to model MPSoC product lines, which facilitate capturing its semantics as FTS. To this end, the probabilistic model checker verifies the resulting FTS that is decorated with tasks characteristics and processors’ failure probability. The experimental results indicate that the formal approach offers quantitative results on the relevant product that optimizes resource usage when exploring the product family.
 
Article
Wireless communication among vehicular ad hoc network (VANET) entities is secured through cryptography, which is used for authentication as well as to ensure the overall security of messages in this environment. Authentication protocols play a significant role and are therefore required to be free of vulnerabilities that allow entity impersonation, unauthorized entry, and general misuse of the system. A resourceful adversary can inflict serious damage to VANET systems through such vulnerabilities. We consider several VANET authentication protocols in the literature and identify vulnerabilities. In addition to the commonly considered vulnerabilities in VANETs, we observe that the often-overlooked relay attack is possible in almost all VANET authentication protocols. Relay attacks have the potential to cause damage in VANETs through misrepresentation of vehicle identity, telematic data, traffic-related warnings, and information related to overall safety in such networks. We discuss possible countermeasures to address identified vulnerabilities. We then develop an authentication protocol that uses ambient conditions to secure against relay attacks and other considered vulnerabilities. We include security proof for the proposed protocol.
 
Article
Industries are going through the fourth industrial revolution (Industry 4.0), where technologies like the Industrial Internet of things, big data analytics, and machine learning (ML) are extensively utilized to improve the productivity and efficiency of manufacturing systems and processes. This work aims to further investigate the applicability and improve the effectiveness of ML prediction models for fault diagnosis in the smart manufacturing process. Hence, we propose several methodologies and ML models for fault diagnosis for smart manufacturing process applications. A case study has been conducted on a real dataset from a semiconductor manufacturing (SECOM) process. However, this dataset contains missing values, noisy features, and class imbalance problem. This imbalance problem makes it so difficult to accurately predict the minority class, due to the majority class size difference. In the literature, efforts have been made to alleviate the class imbalance problem using several synthetic data generation techniques (SDGT) on the UCI machine learning repository SECOM dataset. In this work, to handle the imbalance problem, we employed, compared, and evaluated the feasibility of three SDGT on this dataset. To handle issues related to the missing values and noisy features, we implemented two missing values imputation techniques and feature selection techniques, respectively. We then developed and compared the performance of ten predictive ML models against these proposed methodologies. The results obtained across several evaluation metrics of performance were significant. A comparative analysis shows the feasibility and validate the effectiveness of these SDGT and the proposed methodologies. Some among the proposed methodologies could produce an accuracy in the range of 99.5% to 100%. Furthermore, based on a comparative analysis with similar models from the literature, our proposed models outpaced those proposed in the literature.
 
Article
Software-Defined Networking (SDN) and Network Function Virtualization (NFV) are promising technologies for delivering software-based networks to the user community. The application of Machine Learning (ML) in SDN and NFV enables innovation and easiness towards network management. The shift towards the softwarization of networks broadens the many doors of innovation and challenges. As the number of devices connected to the Internet is increasing swiftly, the SDNFV traffic management mechanism will provide a better solution. Many ML techniques applied to SDN focus more on the classification problems like network attack patterns, routing techniques, QoE/QoS provisioning. The approach of the application of ML to SDNFV and SDN controller placement has a lot of scope to explore. This work aims to develop an ML approach for network traffic management by predicting the number of controllers likely to be placed in the network. The proposed prediction mechanism is a centralized one and deployed as Virtual Network Function (VNF) in the NFV environment. The number of controllers is estimated using the predicted traffic and placed in the optimal location using the K-Medoid algorithm. The proposed method is suitably analysed for performances metrics. The proposed approach effectively combines SDN, NFV and ML for the better achievement of network automation.
 
Article
CNNs have achieved remarkable image classifcation and object detection results over the past few years. Due to the locality of the convolution operation, although CNNs can extract rich features of the object itself, they can hardly obtain global context in images. It means the CNN-based network is not a good candidate for detecting objects by utilizing the information of the nearby objects, especially when the partially obscured object is hard to detect. ViTs can get a rich context and dramatically improve the prediction in complex scenes with multi-head self-attention. However, it sufers from long inference time and huge parameters, which leads ViTbased detection network that is hardly be deployed in the real-time detection system. In this paper, frstly, we design a novel plug-and-play attention module called mix attention (MA). MA combines channel, spatial and global contextual attention together. It enhances the feature representation of individuals and the correlation between multiple individuals. Secondly, we propose a backbone network based on mix attention called MANet. MANet-Base achieves the state-of-the-art performances on ImageNet and CIFAR. Last but not least, we propose a lightweight object detection network called CAT-YOLO, where we make a trade-of between precision and speed. It achieves the AP of 25.7% on COCO 2017 test-dev with only 9.17 million parameters, making it possible to deploy models containing ViT on hardware and ensure real-time detection. CAT-YOLO could better detect obscured objects than other state-of-the-art lightweight models.
 
Article
Existing algorithms have difficulty in solving the two tasks of localization and classification simultaneously when performing traffic sign detection on realistic images of complex traffic scenes. In order to solve the above problems, a new road traffic sign dataset is created, and based on the YOLOv4 algorithm, for the complexity of realistic traffic scene images and the large variation in the size of traffic signs in the images, the multi-scale feature extraction module, cascade feature fusion module and attention mechanism module are designed to improve the algorithm’s ability to locate and classify traffic signs simultaneously. Experimental results on the newly created dataset show that the improved algorithm achieves a mean average precision of 84.44%, which is higher than several major CNN-based object detection algorithms for the same type of task.
 
Article
As the complexity of the cyber-physical systems (CPSs) increase, system modeling and simulation tend to be performed on different platforms where collaborative modeling activities are performed on distributed clients, while the simulations of systems are carried out in specific simulation environments, such as high-performance computing (HPC). However, there is a great gap between system models usually designed in system modeling language (SysML) and simulation code, and the existing model transformation-based simulation methods and tools mainly focus on either discrete or continuous models, ignoring the fact that the simulation of hybrid models is quite important in designing complex systems. To this end, a model transformation approach is proposed to simulate hybrid SysML models under a discrete event system specification (DEVS) framework. In this approach, to depict hybrid models, simulation-related meta-models with discrete and continuous features are extracted from SysML views without additional extension. Following the meta object facility (MOF), DEVS meta-models are constructed based on the formal definition of DEVS models, including discrete, hybrid and coupled models. Moreover, a series of concrete mapping rules is defined to transform the discrete and continuous behaviors based on the existing state machine mechanism and constraints of SysML, separately. Such an approach may facilitate a SysML system engineer to use a DEVS-based simulator to validate system models without the necessity of understanding DEVS theory. Finally, the effectiveness of the proposed method is verified by a defense system case.
 
Zenbo 24 animated faces with emotion expression [7]. (This image is individually adapted from https://zenbo.asus.com/developer/documents/Overview/Design-Guideline/Zenbo-Introduction/Emotions)
Experiment design of questionnaires
An example of positive and negative questionnaires
Positive response in choosing colors in each emotion [7]
Negative response in choosing colors in each emotion [7]
Article
In this paper, we investigate the relationship between emotions and colors by showing robot animated emotion faces and colors to the participants through a series of surveys. We focused on representing a visualized emotion through a robot's facial expression and background colors. To complete the emotion design with animated faces and color background, we gave an experimental design for surveying the users' thoughts. We took an example of a robot animated face by using the ASUS Zenbo. We selected 11 colors as our color background and 24 facial expressions from Zenbo. To analyze our results from questionnaires, we used histograms to show the basic data situation and the multiple logistic regression analysis (MLRA) to see the marginal relationships. We separated our questionnaires into positive and negative questionnaires and divided the dataset into three cases to discuss the different relationships between color and emotion. Results showed that people preferred the blue color no matter whether the face was showing positive or negative emotion. The MLRA also showed the correct percentage is outstanding in case 2, either positive emotion or negative emotion. Participants thought Zenbo's robotic animated face was the same as they thought. Through our experimental design, we hope that people can consider more colors with emotion to design the human–robot interface that will be closer to the users' thoughts and make life more colorful with comfortable reactions with robots.
 
Article
Multistage Interconnection Networks (MINs) are an effective means of communication between multiple processors and memory modules in many parallel processing systems. Literature consists of numerous fault-tolerant MIN designs. However, due to the recent advances in the field of parallel processing, requiring large processing power, an increase in the demand to design and develop more reliable, cost-effective and fault-tolerant MINs is being observed. This work proposes two novel MIN designs, namely, Augmented-Shuffle Exchange Gamma Interconnection Network (A-SEGIN) and Enhanced A-SEGIN (EA-SEGIN). The proposed MINs utilize chaining of switches, and multiplexers & demultiplexers for providing a large number of alternative paths and thereby better fault tolerance. Different reliability measures, namely, 2-terminal, multi-source multi-destination, broadcast and network/global, respectively, of the proposed MINs have been evaluated with the help of all enumerated paths and well-known Sum-of-Disjoint Products approach. Further, overall performance, with respect to the number of paths, different reliability measures, hardware cost and cost per unit, of the proposed MINs has been compared with 19 other well-studied MIN layouts. The results suggest that the proposed MINs are very strong competitors of the preexisting MINs of their class owing to their better reliability and cost effectiveness.
 
Article
A quantum-dot cellular automaton is a new technology that solves all the disputes CMOS technology faces. Quantum-dot cellular automata-based computations run at ultra-high speeds with very high device density and low power consumption. Reversible logic design, featured in quantum-dot cellular automata, permits fully invertible computation. The arithmetic and logic units are the major components in all microprocessor-based systems that probably serve as the processing device's heart. This paper discusses an area-efficient quantum-dot cellular automata technology-based coplanar, reversible arithmetic and logic unit using the double Peres and Feynman gates. With a latency of 2.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2.5$$\end{document} clocks and a total area of 0.1 μm², the proposed arithmetic and logic unit performs 19 logic and arithmetic operations. QCA Designer and QD-E are used to simulate the proposed design and energy consumption, respectively. The proposed design's total energy dissipation, as measured by QCA Designer-E, is 5.45e−002 eV, and the average energy dissipation is 4.95e−003 eV. The proposed method has a considerable number of improvements in terms of latency, the number of operations, and area compared to earlier work.
 
Article
Based on the service orientation, a business service represents a coherent functionality that offers added value to the environment, regardless of how it is realized internally. The enterprise business service is a crucial section of enterprise architecture. Although many leading-edge enterprise architecture frameworks describe architecture in levels of abstraction, they still cannot provide an accurate syntactic and semantic description. If test cases are generated based on accurate descriptions of enterprise business services, the subsequent revisions and changes can be reduced. This research has one main contribution: it starts from the enterprise level, gains benefits from the enriched descriptions for enterprise business service, continues to generate appropriate syntactic and semantic models, and generates test cases from the formal model. In the suggested method, the goals of the enterprise will initially be extracted based on The Open Group Architecture Framework. Then, it will be subjected to syntactic modeling based on the ArchiMate language. Next, the semantics are added in terms of the Web Service Modeling Ontology framework and are manually formalized in B language by applying the defined transformation rules. Finally, the test coverage set will be examined on the formal model to generate test cases. The suggested method has been implemented in the marketing department of a petrochemical company. The results indicate the validity and efficiency of the method.
 
Article
Reliable and efficient delivery of diverse services with different requirements is among the main challenges of IoT systems. The challenges become particularly significant for IoT deployment in larger areas and high-performance services. The low-rate wireless personal area networks, as standard IoT systems, are well suited for a wide range of multi-purpose IoT services. However, their coverage distance and data rate constraints can limit the given IoT applications and restrict the creation of new ones. Accordingly, this work proposes a model that aims to correlate and expand the standard IoT systems from personal to wide areas, thus improving performance in terms of providing fast data processing and distant connectivity for IoT data access. The model develops two IoT systems for these purposes. The first system, 5GIoT, is based on 5G cellular, while the second, LTEIoT, is based on 4G long-term evolution (LTE). The precise assessment requires a reference system, for which the model further includes a standard IoT system. The model is implemented and results are obtained to determine the performance of the systems for diverse IoT use cases. The level of improvement provided by the 5GIoT and LTEIoT systems is determined by comparing them to each other as well as to the standard IoT system to evaluate their advantages and limitations in the IoT domain. The results show the relatively close performance of 5GIoT and LTEIoT systems while they both outperform the standard IoT by offering higher speed and distance coverage.
 
Top-cited authors
Jin Wang
  • Kyung Hee University
Mohammad Shojafar
  • University of Surrey
Ahamad Tajudin Khader
  • Universiti Sains Malaysia
Laith Abualigah
  • Amman Arab University
Enzo Baccarelli
  • Sapienza University of Rome