Search

The study presents an innovative diagnostic framework that synergises Convolutional Neural Networks (CNNs) with a Multi-feature Kernel Supervised within-class-similar Discriminative Dictionary Learning (MKSCDDL). This integrative methodology is designed to facilitate the precise classification of individuals into categories of Alzheimer's Disease, Mild Cognitive Impairment (MCI), and Cognitively Normal (CN) statuses while also discerning the nuanced phases within the MCI spectrum. Our approach is distinguished by its robustness and interpretability, offering clinicians an exceptionally transparent tool for diagnosis and therapeutic strategy formulation. We use scandent decision trees to deal with the unpredictability and complexity of neuroimaging data. Considering that different people's brain scans are different, this enables the model to make more detailed individualised assessments and explains how the algorithm illuminates the specific neuroanatomical regions that are indicative of cognitive impairment. This explanation is beneficial for clinicians because it gives them concrete ideas for early intervention and targeted care. The empirical review of our model shows that it makes diagnoses with a level of accuracy that is unmatched, with a classification efficacy of 98.27%. This shows that the model is good at finding important parts of the brain that may be damaged by cognitive diseases. [ABSTRACT FROM AUTHOR]

Big Data processing architectures are now widely recognized as one of the most significant innovations in Computing in the last decade. Their enormous potential in collecting and processing huge volumes of data scattered throughout the Internet is opening the door to a new generation of fully distributed applications that, by leveraging the large amount of resources available on the network will be able to cope with very complex problems achieving performances never seen before. However, the Internet is known to have severe scalability limitations in moving very large quantities of data, and such limitations introduce the challenge of making efficient use of the computing and storage resources available on the network, in order to enable data-intensive applications to be executed effectively in such a complex distributed environment. This implies resource scheduling decisions which drive the execution of task towards the data by taking network load and capacity into consideration to maximize data access performance and reduce queueing and processing delays as possible. Accordingly, this work presents a data-centric meta-scheduling scheme for fully distributed Big Data processing architectures based on clustering techniques whose goal is aggregating tasks around storage repositories and driven by a new concept of 'gravitational' attraction between the tasks and their data of interest. This scheme will benefit from heuristic criteria based on network awareness and advance resource reservation in order to suppress long delays in data transfer operations and result into an optimized use of data storage and runtime resources at the expense of a limited (polynomial) computational complexity. [ABSTRACT FROM AUTHOR]

Abstract: With the rapid growth and the increasing complexity of network infrastructures and the evolution of attacks, identifying and preventing network abuses is getting more and more strategic to ensure an adequate degree of protection from both external and internal menaces. In this scenario many techniques are emerging for inspecting network traffic and discriminating between anomalous and normal behaviors to detect undesired or suspicious activities. Unfortunately, the concept of normal or abnormal network behavior depends on several factors and its recognition requires the availability of a model aiming at characterizing current behavior, based on a statistical idealization of past events. There are two main challenges when generating the training data needed for effective modeling. First, network traffic is very complex and unpredictable, and second, the model is subject to changes over time, since anomalies are continuously evolving. As attack techniques and patterns change, previously gained information about how to tell them apart from normal traffic may be no longer valid. Thus, a desirable characteristic of an effective model for network anomaly detection is its ability to adapt to change and to generalize its behavior to multiple different network environments. In other words, a self-learning system is needed. This suggests the adoption of machine learning techniques to implement semi-supervised anomaly detection systems where the classifier is trained with “normal” traffic data only, so that knowledge about anomalous behaviors can be constructed and evolve in a dynamic way. For this purpose we explored the effectiveness of a detection approach based on machine learning, using the Discriminative Restricted Boltzmann Machine to combine the expressive power of generative models with good classification accuracy capabilities to infer part of its knowledge from incomplete training data. [Copyright &y& Elsevier]

In this paper, a new traffic engineering-capable routing and wavelength assignment scheme is proposed to efficiently handle LSP and lightpath setup requests with different QoS requirements on modern multi-layer (fully optical core and time-division multiplexed edge) transport networks. The objectives of the proposed algorithm are to minimize the rejection probability by maximizing the network load balancing and efficiently handling the grooming of several LSPs on the same lightpath while respecting the constraints of the optical node architecture and considering both traffic engineering and QoS requirements. The proposed solution consists of a two-stage RWA algorithm: each time a new request arrives, an on-line dynamic grooming scheme finds a set of feasible lightpaths which fulfill the QoS and traffic engineering requirements; then, the best feasible lightpath is selected, aiming to keep the network unbalancing and blocking probability as low as possible in the medium and long term, according to a novel global path affinity minimization concept. Extensive simulation experiments have been performed in which our on-line dynamic RWA algorithm demonstrated significant performances. Thanks to its optimal network resource usage and to its reasonable computational space and time complexity, the algorithm can be very attractive for the next-generation optical wavelength-switched networks. [ABSTRACT FROM AUTHOR]

Advances in optical technologies have enabled the deployment of wavelength division-multiplexed (WDM) transmission systems capable of providing huge amounts of bandwidth across long distances. In this scenario, dynamic routing for direct provisioning of optical paths at the WDM layer becomes a challenging problem. Any distributed algorithm for routing dynamic traffic demands on optical transport infrastructures should be simple, flexible, efficient and scalable. The contribution of this paper is a novel integrated routing and grooming scheme for setting-up bandwidth guaranteed paths on hybrid wavelength and label switched networks. Our proposal exploits and refines the minimum interference routing idea according to an improved and re-optimized resource and traffic-aware approach, where critical links are detected and weighted according to a low complexity all-pairs minimum cut strategy that substantially reduce the overall number of calculations and hence the computational cost. The valuable results achieved in the comparison against other well-known reference techniques clearly demonstrate that our algorithm is very time-efficient while performing better in terms of blocking probability. [ABSTRACT FROM AUTHOR]

With the growing demand and commercial availability of cloud services, the need for comparison of their functionality against different prices and performance has arisen. A relevant and fair comparison is still challenging due to diverse deployment options and dissimilar features of different services. This paper addresses a hybrid multi-criteria decision-making model involving the selection of cloud services among the available alternatives. The proposed methodology assigns various ranks to cloud services based on the quantified quality-of-service parameters using a novel extended Grey Technique for Order of Preference by Similarity to Ideal Solution integrated with analytical hierarchical process. Further, we analyse the proposed cloud service selection method in terms of sensitivity analysis, adequacy under change in alternatives, adequacy to support group decision-making, and handling of uncertainty. This analysis helps both researchers and practitioners for analysing more fruitful approaches for cloud service selection. [ABSTRACT FROM AUTHOR]

Several cloud services with comparable functionality are now available to customers at different prices and performance levels. Often, there may be trade-offs among different functional and non-functional requirements fulfilled by different cloud providers. Hence, it is difficult to evaluate the relative performances of the cloud services and their ranking based on various quality of service attributes. In this paper, we propose a modified data envelopment analysis and a modified super-efficiency data envelopment analysis for evaluating the cloud services and their efficiencies considering user preferences. We compare these methods of cloud service selection based on sensitivity analysis, adequacy to changes in DMUs, adequacy to support decision making and modeling of uncertainty. The comparison helps customers to choose a cloud service that is most suitable to their requirements and also creates a healthy competition among the cloud service providers. [ABSTRACT FROM AUTHOR]

Despite the wide deployment of advanced wireless coverage infrastructures, finding the best way for achieving secure mobile communication in every-day's life activities is still an open question. Indeed, a large number of mobile terminals, supporting multiple networking technologies, may be used to manage data from everywhere and at anytime. However, the effort required for achieving security, given the complexity of cryptographic algorithms, heavily affects the power consumption of terminals. Such energy demand, together with the one required to manage communication activities, makes energy-efficient secure communication among hardware-constrained handheld devices a challenging topic. In this work, we introduce an analytic energy model for secure communication among multi-mode terminals. This model describes the energy consumption of mobile terminals operating within a dynamic network scenario, considering both their interconnection and secure data exchange issues, in order to develop adaptive strategies for energy-efficient secure communications. Finally, the model has been validated through simulation. [ABSTRACT FROM AUTHOR]

This work analyzes a new and very subtle kind of security threat that can affect large-scale cloud-based IT service infrastructures, by exploiting the computational resources of their component data center to waste as much energy as possible. The consequence of these threats ranges from increased costs in the energy bill, to penalization for exceeding the agreed quantity of greenhouse gases (GHG) emissions, up to complete denial of service caused by electrical outages due to power budget exhaustion. We analyzed different types of such attacks with their potential impacts on the energy consumption, modeled their behavior and quantified how current energy-proportional technologies may provide attackers with great opportunities for raising the target facility emissions and costs. These efforts resulted in a simple model with some parametric reference values that can be used to estimate the impact of such attacks also in presence of very large infrastructures containing thousands or millions of servers. [ABSTRACT FROM AUTHOR]

Abstract: Energy consumption is now one of the most important issues for network carriers, since the majority of the energy needed for their operation is consumed in the wireless access and optical transport networks. The continuous growth in the wireless customers and traffic volumes and the consequent energy demand on modern carriers’ broadband infrastructures require reconsidering their energy efficiency, by starting from the formulation of new, more complete and representative network models that should become the foundations for modern energy-aware control plane architectures. Accordingly, this work presents a novel comprehensive energy model for next-generation wireless access-over-optical-transport networks characterized by hybrid power systems (i.e., multiple dynamically available power sources). The objective is to identify the energy-related information that need to be handled at the control plane layer to support energy-aware networking practices. Such information can be made available to suitable energy-aware routing and wavelength assignment algorithms that may exploit them to optimize the overall network energy-consumption and reducing the associated carbon footprint. The proposed model may be taken as a reference for the implementation of new energy-aware control plane protocols (routing and signaling) that make use of power-related considerations to achieve energy-efficiency and energy-awareness in wavelength-routed network infrastructures. [Copyright &y& Elsevier]

Two-stage anaerobic digestion (AD) is a promising method of converting organic waste into clean and sustainable energy. This work aimed to study the efficiency of a two-stage AD system with the production of hydrogen at the first acidogenic stage and methane in an electromethanogenic reactor. Two types of biocathodes based on carbon cloth were used in the electromethanogenic reactor: with and without retention of biomass on the electrode surface. A high applied voltage continuously supplied to the pair of electrodes was automatically maintained at 2.5 V. Hydrogen yield (0.11 NL/g volatile solids (VS) init or 2 mol H 2 /mol hexose) and hydrogen content in biogas (52%) were relatively high, even despite the extremely low pH (∼4.0). Improved retention of biomass on the surface of the carbon cloth-based biocathode led to greater stability of the AD process (no water electrolysis was observed) and a significant increase in the efficiency of electromethanogenesis, including the methane yield (by 40.5% up to 0.39 NL/g VS init), volumetric methane production rate (by 38.8% up to 1.16 NL/L/d) and current density (by 233% up to 0.56 A/m2). • T= 57.5 °C and pH ∼4 completely suppressed methanogenesis in the acidogenic reactor. • 2 mol H2/mol glucose during continuous dark fermentation of OF-MSW is achieved. • Better biomass attachment to the biocathode increased the current density by 233%. • Increase in CH4 yield by 40.5% and volumetric CH4 production rate by 38.8%. • A physical barrier is an effective tool to retain biomass on the biocathode surface. [ABSTRACT FROM AUTHOR]

Class imbalance problems have attracted the research community, but a few works have focused on feature selection with imbalanced datasets. To handle class imbalance problems, we developed a novel fitness function for feature selection using the chaotic salp swarm optimization algorithm, an efficient meta-heuristic optimization algorithm that has been successfully used in a wide range of optimization problems. This paper proposes an AdaBoost algorithm with chaotic salp swarm optimization. The most discriminating features are selected using salp swarm optimization, and AdaBoost classifiers are thereafter trained on the features selected. Experiments show the ability of the proposed technique to find the optimal features with performance maximization of AdaBoost. [ABSTRACT FROM AUTHOR]

Abstract: Current Internet e-mail facilities are built onto the foundation of standard rules and protocols, which usually allow a considerable amount of “freedom” to their designers. Each of these standards has been defined based on a number of vendor specific implementations, in order to provide common inter-working procedures for cross-vendor communication. Thus, a lot of optional and redundant information is being exchanged during e-mail sessions, which is available to implement versatile covert channel mechanisms. This work exploits this possibility by presenting a simple but effective steganographic scheme that can be used to deploy robust secret communication through spam e-mails. This scheme can offer unidirectional asynchronous one-to-one or one-to-many covert channel facilities that are able to bypass the most sophisticated firewalls and traffic analyzers. Its implementation neither affects the involved transport protocols nor causes any perceivable performance degradation or data loss to the end-users. The proposed scheme allows one to manage possible filtering/loss of the e-mails being the vehicle of the secret information. A novel retransmission method based on the Raptor codes has been adopted. The use of Raptor codes is key to correctly and efficiently manage the difficulty or impossibility to retransmit e-mails in the case of a unidirectional secret communication starting from one sender and directed to many recipients. In order to evaluate the performance characteristics of the proposed scheme, an empirical estimation of the covert channel bandwidth has been performed. [Copyright &y& Elsevier]

The design of the suspension model of road cars is extremely important for both driver comfort and safety. Road traffic injuries are currently the eighth leading cause of death worldwide and are predicted to become the seventh leading cause of death by 2030, according to the World Health Organization (WHO). Among other factors, road infrastructure is one of the most important factors to consider when designing the suspension model, especially as autonomous driving becomes more popular. In addition, there is a growing need to test and study the impact of road roughness and pavement type on autonomous vehicles (AVs). This work provides the following contributions: a platform for suspension system optimization–flexible, highly parameterizable and rich in applications. Our solution implements, tests and optimises the 2, 3 and 4 degrees of freedom (DOF) quarter-car models (QCM) with linear seat suspension and the 5 DOF half-car model (HCM), respectively, based on random road profiles generated according to ISO 8608 standards. In addition, many multi-objective optimization (MOO) algorithms have been implemented and applied in isolation to see their comparative performance (in terms of Pareto fronts, hypervolume, coverage, spread, and epsilon), but most importantly we have also introduced a meta-optimization technique by super-positioning Non-dominated Sorting Genetic Algorithm II and Strength Pareto Evolutionary Algorithm 2, where each of the algorithms, running concurrently, produces a percentage of the total number of individuals in a generation, percentage that is calculated according to the solution quality of the individuals in the previous generation. The experimental results were obtained by simulating the optimal configurations generated by the MOO heuristic algorithms on different categories of cars (small/medium). Finally, the Taguchi method and regression analysis were used to evaluate the effects of specific parameters on system responses such as sprung mass acceleration and displacement. The Taguchi simulation was performed to experimentally validate the results obtained by metaheuristic optimization methods. [ABSTRACT FROM AUTHOR]

Wireless sensor networks (WSNs) are composed of a large number of tiny sensor nodes that are self-organized through wireless communication. It aims to perceive, collect, and process information from network coverage areas. The open nature of WSNs makes them easily exposed to a variety of attacks and brings many security challenges. Furthermore, because of the limited resources, some nodes may refuse to forward packets by dropping them to save their resources such as battery, cache, and bandwidth. To resist the attacks from these selfish nodes and to punish them, we propose a fuzzy-based dynamic trust model in this paper. The model uses fuzzy sets combining with grey theory to evaluate every node's trust credibility based on direct trust and indirect trust relationship. Only those with higher trust values can be chosen to forward packets. Those untrustworthy nodes with lower trust values will be detected and excluded from the trust list. Thus, our proposal also produces an incentive to compel the selfish nodes to wellbehave again to participate in the WSN again. Additionally, we introduce the time slice scheme to guarantee a reliable node possess enough time to enjoy its services, which can solve the problem that a suddenly interrupted link causes a significant decrease of the trust value. Simulation results show that our dynamic trust model can not only demonstrate the effectiveness in detecting selfish nodes, but also possess better performance even if the bursty traffic exists. [ABSTRACT FROM AUTHOR]

Distributed cryptographic file systems enable file sharing among their users and need the adoption of a key management scheme for the distribution of the cryptographic keys to authorized users according to their specific degree of trust. In this paper we describe the architecture of a basic secure file sharing facility relying on a multi-party threshold-based key-sharing scheme that can be overlaid on top of the existing stackable networked file systems, and discuss its application to the implementation of distributed cryptographic file systems. It provides flexible access control policies supporting multiple combination of roles and trust profiles. A proof of concept prototype implementation within the Linux operating system framework demonstrated its effectiveness in terms of performance and security robustness. [ABSTRACT FROM AUTHOR]