ALBERT

Description: 〈h3〉Abstract〈/h3〉 〈p〉In recent years, notable number of research studies have been conducted on the analysis of diffusion process in complex networks. One fundamental problem in this domain is to find the most influential spreader nodes. For achieving a successful spreading process, nodes having high spreading ability should be selected as spreaders. Many centrality measures have been proposed for determining and ranking the significance of nodes and detecting the best spreaders. The majority of proposed centrality measures require network global information which leads to high time complexity. Moreover, with the advent of large-scale complex networks, there is a critical need for improving accurate measures through using nodes’ local information. On the other hand, most of the formerly proposed centrality measures have attempted to select core nodes as spreaders but global bridge nodes have the highest spreading ability since they are located among several giant communities of the network. In this study, a new local and parameter-free centrality measure is proposed which is aimed at finding global bridge nodes in the network. Hence, two new local metrics, namely 〈em〉edge ratio and neighborhood diversity〈/em〉, are firstly defined which are used in the proposed method. Considering edge ratio of neighbors ensures that the selected node be not in the periphery location of the network. Furthermore, a node with high neighborhood diversity is likely a connector between some modules (dense parts) of the network. Therefore, a node with a high edge ratio and more diverse neighborhood has high spreading ability. The major merits of the proposed measure are near-linear time complexity, using local information and being parameter-free. For evaluating the proposed method, we conducted experiments on real-world networks. The results of comparing the proposed centrality measure with other measures in terms of epidemic models (SIR and SI), Kendall’s tau correlation coefficient and Rank-Frequency measures indicated that the proposed method outperforms the other compared centrality measures.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Current developments in information and electronic technologies have pushed a tremendous amount of applications to meet the demands of personal computing services. Various kinds of smart devices have been launched and applied in our daily lives to provide services for individuals; however, the existing computing frameworks including local silo-based and cloud-based architectures, are not quite fit for personal computing services. Meanwhile, personal computing applications exhibit special features, they are latency-sensitive, energy efficient, highly reliable, mobile, etc, which further indicates that a new computing architecture is urgently needed to support such services. Thanks to the emerging edge computing paradigm, we were inspired to apply the distributed cooperative computing idea at the data source, which perfectly solves issues occurring among existing computing paradigms while meeting the requirements of personal computing services. Therefore, we explore personal computing services utilizing the edge computing paradigm, discuss the overall edge-based system architecture for personal computing services, and design the conceptual framework for an edge-based personal computing system. We analyze the functionalities in detail. To validate the feasibility of the proposed architecture, a fall detection application is simulated in our preliminary evaluation as an example service in which three Support Vector Machine based fall detection algorithms with different kernel functions are implemented. Experimental results show edge computing architecture can improve the performance of the system in terms of total latency, with about 22.75% reduction on average in the case of applying 4G at the second hop even when the data and computing stream of the application is small.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Communications among IoT devices and backend servers are among the most critical aspects of IoT system, as it not only governs hardware requirement and system performance but also impacts security and privacy of the system. Multi-access edge computing (MEC) is an enabler for service development in 5G. Furthermore, MEC will reduce IoT system communication cost such as latency and bandwidth thanks to its ubiquitously nearby processing and storage capability. Also, MEC will provide opportunities to enhance IoT system’s network privacy, which cannot be solved by IoT system itself easily. In this paper, we present MEC enhanced mobility support system for IoT system: with MEC it can be implemented with low operation cost and overhead and can protect IoT system’s user from being exploited due to the exposed network location of data uploading and the control signals.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Detecting and tracking events from logging data is a critical element for security and system administrators and thus attracts more and more research efforts. However, there exists a major limitation in current processes of Event Logging analysis, related to the verbosity and language-dependence of messages produced by many logging systems. In this paper, a novel methodology was proposed to tackle this limitation by analysing event messages through a Natural Language Processing task in order to annotate them with semantic metadata. These metadata are further used to enable semantic searches or domain ontology population that help administrator to filter only relevant event and to correlate them for a prompt and efficient response and incident analysis.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Network virtualization is widely considered as a mainstay for overcoming the Internet’s ossification problem, and virtual network embedding (VNE) is a critical issue. Over recent years, growing energy costs and increased ecological awareness have stimulated the interest in reducing energy consumption by Internet service providers (ISP). Dependability is also an important requirement, as it involves metrics such as reliability and availability, which directly impact quality of service (QoS). Prior works on virtual network embedding have focused mainly on maximizing revenue for Internet service providers (ISPs), and they did not consider energy consumption and dependability metrics jointly in the mapping. This paper presents an energy-efficient mapping of dependable virtual networks. The approach considers a problem formulation that concomitantly takes into account energy consumption and availability constraints for virtual network embedding problem, and an algorithm based on Greedy Randomized Adaptive Search Procedure (GRASP) metaheuristic is adopted. The algorithm utilizes a sensitivity analysis based on availability importance to achieve the QoS required by each virtual network, and models based on reliability block diagrams (RBD) and stochastic Petri nets (SPN) are utilized to estimate availability. Results demonstrate the feasibility of the proposed approach, and they show the trade-off between availability, energy consumption, cost and revenue.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Thinning algorithms have been widely applied in many applications such as computer graphics, computer vision and medical imaging. The performances of thinning algorithms are hampered by the increase of image’s size and image’s content. By employing graphics processing units (GPUs), the calculation of thinning algorithms can be accelerated. In this paper, we present GPU parallel versions of well-known binary thinning algorithms namely the ZS algorithm, the GH algorithm, the AW algorithm and the hybrid algorithm belonging to different classes for an efficient implementation using CUDA as a parallel programming model. These thinning algorithms are then evaluated and compared according to quality and runtime measures. The obtained results show that the GPU multithreaded implementations of the selected algorithms present high computational speed outperforming the central processing unit sequential implementations achieving the best average speedup when executing the AW algorithm.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉This study elucidates the dynamic behaviour of the two competing mutually exclusive epidemic (meme) spreading model with the alert of memes over multiplex social networks. Each meme spreads over a distinct contact networks 〈span〉 〈span〉\((CN_1,CN_2)\)〈/span〉 〈/span〉 of an undirected multiplex social network. The behavioural responses of agents (alerts) to the spread of competing memes is disseminated through information dissemination network (IDN). Here, IDN has the same nodes but different links with respect to the respective 〈span〉 〈span〉\(CN_i(i=1,2)\)〈/span〉 〈/span〉. The analytical treatment of this model is analysed through the mean field approximation of the epidemic process. Also, it has been shown through numerical illustrations that a node in the alert state is less probable to become infected than a node in the susceptible state. Moreover, co-existence of both the memes, the survival threshold, the absolute dominance threshold of the two competitive memes and the alert threshold for minimizing the severity of meme spread are analytically explored and numerically illustrated.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Recently, the Internet of things (IoT) has received a lot of attention from both industry and academia. A reliable and secure IoT connection and communication is essential for the proper working of the IoT network as a whole. One of the ways to achieve robust security in an IoT network is to enable and build trusted communication among the things (nodes). In this area, the existing IoT literature faces many critical issues, such as the lack of intelligent cluster-based trust approaches for IoT networks and the detection of attacks on the IoT trust system from malicious nodes, such as bad service providers. The existing literature either does not address these issues or only addresses them partially. Our proposed solution can firstly detect on-off attacks using the proposed fuzzy-logic based approach, and it can detect contradictory behaviour attacks and other malicious nodes. Secondly, we develop a fuzzy logic-based approach to detect malicious nodes involved in bad service provisioning. Finally, to maintain the security of the IoT network, we develop a secure messaging system that enables secure communication between nodes. This messaging system uses hexadecimal values with a structure similar to serial communication. We carried out extensive experimentation under varying network sizes to validate the working of our proposed solution and also to test the efficiency of the proposed methods in relation to various types of malicious behavior. The experiment results demonstrate the effectiveness of our approach under various conditions.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The imbalance energy consumption and high data traffic at intermediate nodes degrade the network performance. In this paper, we propose: energy grade and balance load distribution corona, EG without corona and DA without corona based schemes to distribute data traffic across the network nodes for efficient energy consumption. The dynamic adjustment of transmission range in first scheme helps in reducing data load. Additionally, the transmission range is purely based on distance, energy and data load of the receiver node for achieving maximum network lifetime. Second scheme divides a data packet into three fractions; small, medium and large for transmitting from various paths to evenly distribute the data load on the network nodes. In third scheme, depth adjustment of void node is performed to resume network operations, whereas, the load distribution and transmission range mechanisms are the same. The extensive simulations are carried out to show the effectiveness of proposed schemes in terms of PDR, energy consumption, and load distribution against the baseline scheme.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Over the past few years, there has been a rapid increase of data originating from evolving networks such as social networks, sensor networks and others. A major challenge that arises when handling such networks and their respective graphs is the ability to issue a historical query on their data, that is, a query that is concerned with the state of the graph at previous time instances. While there has been a number of works that index the historical data in a time-centric manner (i.e. according to the time instance an update event occurs), in this work, we focus on the less-explored vertex-centric storage approach (i.e. according to the entity in which an update event occurs). We demonstrate that the design choices for a vertex-centric model are not trivial, by proposing two different modelling and storage models that leverage NoSQL technology and investigating their tradeoffs. More specifically, we experimentally evaluate the two models and show that under certain cases, their relative performance can differ by several times. Finally, we provide evidence that simple baseline and non-NoSQL solutions are slower by up to an order of magnitude.〈/p〉