Search

Important progresses have been made in the field of artificial intelligence in recent years, and facial expression recognition (FER), which could greatly facilitate the development of human-computer interaction, has been becoming a significant research hotspot. In this paper, a novel nonnegative tensor factorization method is proposed based on low-rank subspace (NTF−LRS) for FER. Firstly, in order to find the high order correlations underlying multi-dimensional data, a data tensor model is constructed, which could represent different dimensional features ingeniously. And then, the low-rank subspace model is adopted to reconstruct the original tensor model, reduce the redundancy of the learned new tensor, and improve the discriminant abilities of inter-class information. Finally, the reconstructed tensor is decomposed to get factor matrices by nonnegative tensor factorization, where all factor matrices are used to extract subspace features. To verify the effectiveness of our proposal, two well-known facial expression datasets named as “JAFFE” and “CK+” are utilized for evaluation, and the experimental results show that the tensor-based method preserves the original structure of whole samples, which avoids the case of dimension curse because of vectorization. In addition, this method uses Laplacian graph to impose regularization on low-rank subspace model, which keeps the local relationship between sample neighbors.

Utility-oriented mining which integrates utility theory and data mining is a useful tool for understanding economic consumer behavior. Traditional algorithms for mining high-utility patterns (HUPs) applies a single/uniform minimum high-utility threshold (minutil) to obtain the set of HUPs, but in some real-life circumstances, some specific products may bring lower utilities compared with others, but their profit may offer some vital information. However, if minutil is set high, the patterns with low minutil are missed; if minutil is set low, the number of patterns becomes unmanageable. In this paper, an efficient one-phase utility-oriented pattern mining algorithm, called HIMU, is proposed for mining HUPs with varied item-specific minimum utility. A novel tree structure called a multiple item utility set-enumeration tree (MIU-tree), the global sorted and the conditional downward closure properties are introduced in HIMU. In addition, we extended the compact utility-list structure to keep the necessary information, and thus this one-phase HIMU model greatly reduces the computational costs and memory requirements. Moreover, two pruning strategies are then extended to enhance the performance. We conducted extensive experiments in several synthetic and real-world datasets; the results indicates that the designed one-phase HIMU algorithm can address the "rare item problem" and has better performance than the state-of-the-art algorithms in terms of runtime, memory usage, and scalability. Furthermore, the enhanced algorithms outperform the non-optimized HIMU approach., Under review in ACM Trans. on Data Science, 31 pages

Elephant flows (elephants) refer to the sequences of packets that contribute only 10% of the total volume but consume over 90% of the network bandwidth. They often cause network congestion and should be efficiently managed. Present cloud data centers often involve host- and switch-based approaches to detect and schedule elephants, but suffer (1) each host and switch in the network needs to be customized, and (2) dynamic models and advanced policies are difficult to be applied. Software Defined Cloud (SDC) addresses these issues by enabling controller-based approaches. With the aid of Machine Learning (ML) technologies, SDC can achieve learning-based models, flexible deployment, and early detection and schedule of elephants for the optimization of network performance and energy usage in a dynamic and intelligent manner. On this purpose, this article emphases the significance of models describing elephants, surveys the mechanisms that may apply to model, detect, and schedule elephants for SDC to optimize the network performance and energy usage. To the best of our knowledge, this work is the first effort that reviews the techniques in all these related subtopics simultaneously in the context of energy cloud.

Data center benefits cloud applications in providing high scalability and ensuring service availability. However, virtual machine (VM) placement in data center poses new challenges for service provisioning. For many cloud services such as storage and video streaming, present placement approaches are unable to support network-demanding services due to overwhelming communication traffic and time. Therefore VM placement concerning link capacity is vital to cloud data centers. In this paper, we define the network-aware VM placement optimization (NAVMPO) problem based on integer linear programming. The objective function of NAVMPO problem aims to minimize communication time for VMs of the same service type. Then we propose the service-oriented physical machine (PM) selection (SOPMS) algorithm and link-aware VM placement (LAVMP) algorithm. The SOPMS algorithm selects the most appropriate PM based on service-oriented architecture, and then the LAVMP algorithm deploys the most suitable VM to target PM regarding to the link capacity between them. Simulation results show that the proposed placement approach significantly decreases communication time compared to existing non-service-oriented and service-oriented VM placement algorithms, and also improves the average utility rate of PMs with lower power consumption.

Supporting the huge amount of network traffic from three-dimensional (3-D) streaming is an extremely important challenge, as 3-D streaming has the potential to cause network congestion and longer waiting times. These problems can also lead to 3-D video being blurred, and negatively affecting the user experience. This article uses the front camera of a mobile device to track a user's viewing angle, and then calculates the currently needed 3-D stream to find the most suitable peer for video source supply. This article also considers several impact factors to choose supply partners with different data flow via fuzzy theory. Consequently, the proposed method is able to determine the best way to download the desired 3-D stream without unnecessary congestion, thus maintaining the overall peer-to-peer network quality. Simulation results show that the proposed method can reduce the quantity of the video flow, while maintaining video quality, thus avoiding bottlenecks like bandwidth and network speed.

Energy efficiency is one of the most important concerns in cloud/edge computing systems. A major benefit of the Dynamic Voltage and Frequency Scaling (DVFS) technique is that a Virtual Machine (VM) can dynamically scale its computation frequency on an on-demand basis, which is helpful in reducing the energy cost of computation when dealing with stochastic workloads. In this paper, we study the joint workload allocation and computation resource configuration problem in distributed cloud/edge computing. We propose a new energy consumption model that considers the stochastic workloads for computation capacity reconfiguration-enabled VMs. We define Service Risk Probability (SRP) as the probability a VM fails to process the incoming workloads in the current time slot, and we study the energy-SRP tradeoff problem in single VM. Without specifying any distribution of the workloads, we prove that, theoretically there exists an optimal SRP that achieves minimal energy cost, and we derive the closed form of the condition to achieve this minimal energy point. We also derive the closed form for computing the optimal SRP when the workloads follow a Gaussian distribution. We then study the joint workload allocation and computation frequency configuration problem for multiple distributed VMs scenario, and we propose solutions to solve the problem for both Gaussian and unspecified distributions. Our performance evaluation results on both synthetic and real-world workload trace data demonstrate the effectiveness of the proposed model. The closeness between the simulation results and the analytical results prove that our proposed method can achieve lower energy consumption compared with fixed computation capacity configuration methods.

With the rapid growth of mobile devices and the emergence of 5G applications, the burden of cellular and the use of the licensed band have enormous challenges. In order to solve this problem, opportunity communication is regarded as a potential solution. It can use unlicensed bands to forward content to users under delay-tolerance constraints, as well as reduce cellular data traffic. Since opportunity communication is easily interrupted when User Equipment (UE) is moving, we adopt Artificial Intelligence (AI) to predict the location of the mobile UE. Then, the meta-heuristic algorithm is used to allocate multiple contents. In addition, deep learning-based methods almost need a lot of training time. Based on real-time requirements of the network, we propose AI-enabled opportunistic networks architecture, combined with Mobile Edge Computing (MEC) to implement edge AI applications. The simulation results show that the proposed multiple contents offloading mechanism can reduce cellular data traffic through UE location prediction and cache allocation.

With the maturity of technologies such as Internet of Things (IoT) and Cloud Computing, more and more medical institutions share patient's medical data in the IoT-connected healthcare system, whereas it also brings hidden dangers to data security and privacy protection. The application of Hierarchical Attribute-Based Encryption (HABE) in IoT-connected healthcare system helps to solve the problem of data sharing with a large number of users. Delegation is a function of HABE, which can realize the transfer of hierarchical user access rights, effectively reducing the workload of the trusted authority. However, when “delegation” is used to generate decryption keys for users who do not comply with the access structure, the “key-delegation abuse” problem arises, which can seriously damage the privacy of patient's data. Nevertheless, it has not attracted much attention in previous research. This paper proposes the problem of key-delegation abuse in Ciphertext-Policy Hierarchical Attribute-Based Encryption (CP-HABE). After the cryptanalysis of typical schemes, we conclude two reasons for key-delegation abuse in CP-HABE, which are randomizing the original key elements or keeping parts of them in a new decryption key. This paper proposes a new mechanism specifically by using directed graph and construct a CP-HABE scheme against key-delegation abuse (CP-HABE-AKDA). Aiming at the scenario of key leaking, we further present the traceable CP-HABE-AKDA solution for the IoT-connected healthcare system, which has the additional function of tracking and verifying the identity of key leaker.

The development of Internet of Vehicles (IoVs) need to be supported by high-speed network and a large number of Internet of Things (IoT) technologies. In order to meet the requirements of different IoV applications, effective transmission quality is one of the necessary conditions for the development of IoV. Therefore, this study applies the concept of Fog computing and cloud computing to IoV to effectively transmit a large number of multimedia data and sensing data. Based on the architecture of Software-Defined Wireless Network (SDWN), the optimization algorithm is proposed to reduce the transmission latency through optimizing access point selection and the load balancing of the Service Function (SF). The simulation results shows that the proposed mechanisms can distribute the traffic evenly and find the optimal access point.

Human behavior recognition (HBR), as a critical link for further intelligent and real-time smart city application design, has attracted much more attention in recent years. Although the related technologies have been developed rapidly and many solid achievements have been already obtained, there is still a lot of space to deeply enhance the related research including the recognition structures, algorithms, and so on, to meet the increasing requirements of Smart City construction. In this article, we first review the conventional HBR structure, and analyze the problems and challenges for future smart city applications. Then a parallel and multi-layer deep recognition architecture (PMDRA) is discussed, which could have more powerful and ubiquitous feature extraction ability because of the hierarchical utilization of the deep learning network. Meanwhile, the quantity adjustment mechanism for DRUs and DLNUs could help for designing the actual architecture according to the requirements of real scenarios.

Artificial intelligence is one of the important technologies for industrial applications, but it needs a lot of computing resources and sensing data to support. Therefore, big data transmission is a challenge for current network architectures. In order to have high-performance computing requirements, this paper proposes an emerging network architecture that combines edge computing and cloud computing to reduce the transmission of useless data and solve bottleneck problems. Moreover, we define the resource allocation problem about multiple remote radio heads and multiple baseband unit pools in the cloud radio access network for fifth generation. The long short-term memory is used to predict dynamic throughput and genetic algorithm based resource allocation algorithm is used to optimize resource allocation. The simulation results represented that the proposed mechanism can achieve high resource utilization and reduce power consumption.

Many well known organizations have predicted substantial growth for the future of 5G mobile communication. Because constructing a hardware budget is limited, access for a large number of mobile devices is limited for 5G requirements. The biological ecosystem is a very strong self-organizing system similar to the vision of the 3GPP proposed self-organizing network. This leads us to believe that BIN is a potential solution for 5G mobile communication. This article presents a complete mapping between the biological ecosystem and biologically-inspired system for 5G mobile communication. A case study is used to illustrate the proposed method.

Edge computing has emerged as a promising technique because of its advantages in providing low-latency computation offloading services for resource-limited mobile user devices and Internet of Things applications. Computationally intensive artificial intelligence (AI) tasks are well suited to be offloaded to the Cloudlet server, but there is a lack of energy-delay optimization models specifically designed for this edge AI scenario. In this paper, we propose a multiple algorithm service model (MASM) that provides heterogeneous algorithms with different computation complexities and required data sizes to fulfill the same task, and develop an optimization model that aims at reducing the energy and delay cost by optimizing the workload assignment weights and computing capacities of virtual machines, at the same time guaranteeing the quality of the results (QoRs). We propose a tide ebb algorithm to solve the MASM optimization model, and we prove its Parato optimality. Numerical results obtained demonstrate the effectiveness of our proposed method, and prove that the energy and delay costs can be significantly reduced by sacrificing the QoR of the offloaded AI tasks.

Since traffic in networks is growing rapidly, it is difficult for the existing network architecture to support the huge traffic requirement. This article proposes a novel intelligent architecture as a promising paradigm for B5G heterogeneous networks to optimize network resource usage and network performance. The main idea is to build a suitable network model through AI, and integrate edge computing and cloud computing to improve computing performance. In addition, this article gives appropriate recommendations of the deep learning method for different network issues. Since the deep learning method requires a large amount of computing resources, the network resource allocation needs to be paid attention to in this architecture. For complex environments of B5G heterogeneous networks, integrated packet forwarding is one potential technology to improve quality of service. Moreover, we discuss the challenges and open issues for B5G.

Small cell aims at improving the cell coverage and capacity of macrocell. Network operators investigate cell planning for improving system performance and for satisfying user requirements with minimal construction cost and least unserved users. Most of the existing literature merely considered a single objective function or investigated cell planning for small-scale and homogeneous networks. This study aims to optimise multiple objective functions of a large-scale and heterogeneous wireless network with multiple macrocells and multiple small cells. The authors formulate the multi-objective optimisation problem of cell planning for heterogeneous cellular networks. The three objects include construction cost, number of unserved users and network capacity. Then they propose the large-scale cell planning genetic algorithm (LSCPGA) to find a planning result with the lowest fitness value. Simulation results show that LSCPGA economises 9-20% construction cost, eliminates the number of unserved users from 4 to 10% and enhances 2-15% in network capacity compared to other planning algorithms.

To enable learning-based network management and optimization, the 5th Generation Mobile Communication Technology and Internet of Things systems usually involve software-defined networking (SDN) architecture and multiple SDN controllers to efficiently collect the big volume of runtime statistics, define network-wide policies, and enforce the policies over the whole network. To better plan the placement of controllers over SDN systems, this article proposes a generic controller placement problem (GCP) that considers the organization and placement of controllers as well as the switch attachment to optimize the delay between controllers and switches, the delay among controllers, and the load imbalance among controllers. To solve this problem without losing generality, a novel multi-objective genetic algorithm (MOGA) with a mutation based on a variant Particle Swarm Optimization (PSO) is proposed. This PSO chooses a global best position for a particle according to a pre-computed global best position set to lead the mutation of the particle. It successfully handles multiple conflicting objectives, fits the scenario of mutation, and can apply in many other flavors of MOGAs. Evaluations over 12 real Internet service provider networks show the effectiveness of our MOGA in reducing convergence time and improving the diversity and accuracy of the Pareto frontiers. The proposed approaches in formulating and solving the GCP in this article are general and can be applied in many other optimization problems with minor modifications.

The sustainable utilization of marine resources is a vital issue to enrich marine life and to prevent species extinction caused by overfishing. Nowadays, it is common that commercial and smaller vessels are equipped with an Automatic Identification System (AIS) and GPS for better vessel tracking to avoid vessel collision as well as mayday calls. Additionally, governments can monitor vessels’ sea activities through AIS messages, stopping them from overfishing or tracking if any vessel has caused marine pollution. However, because AIS devices cannot guarantee data security, they are susceptible to malicious attacks such as message modification or an illegitimate identity faking a distress signal that causes other vessels to change their course. Given the above, a comprehensive network security system of a sustainable marine environment should be proposed to ensure secure communication. In this paper, a stationary IoT-enabled (Internet of Things) vessel tracking system of a sustainable marine environment is proposed. The system combines network security, edge computing, and tracking management. It offers the following functions: (1) The IoT-based vessel tracking system tracks each aquafarmer’s farming zone and issues periodic warning to prevent vessel collision for pursuing a sustainable marine environment, (2) the system can serve as a relay station that evaluates whether a vessel’s AIS data is correct, (3) the system detects abnormal behavior and any irregular information to law enforcement, (4) the system’s network security mechanism adopts a group key approach to ensure secure communication between vessels, and (5) the proposed edge computing mechanism enables the tracking system to perform message authentication and analysis, and to reduce computational burden for the remote or cloud server. Experiment results indicate that our proposed system is feasible, secure, and sustainable for the marine environment, and the tendered network security mechanism can reduce the computational burden while still ensuring security.

In recent years, the use of unmanned aerial vehicles (UAVs) as aerial base stations to assist internet of things (IoT) has attracted wide interest from academia and industries due to UAV's high mobility, flexibility and low cost. To overcome the shortcomings of data backhaul in IoT, a UAV-assisted IoT framework is proposed in this paper. To maintain lifetimes of internet of things devices (IoTDs), the energy-efficiency of UAV-assisted IoT is maximized by jointly optimizing IoTDs' wake-up scheme and the fixed-wing UAV's trajectory. Since the optimization problem of energy-efficiency of UAV-assisted IoT is formulated as a mixed integer non-convex problem, an iterative algorithm is proposed by adopting successive convex approximation techniques. Simulation results show that the proposed algorithm achieves a higher energy-efficiency in UAV-assisted IoT compared with the traditional benchmark algorithm.

Accurate forecasting of short-term load forecasting is of great help to demand-side response and power dispatching. In order to improve the accuracy of short-term power load prediction, the original power load data signals are decomposed by using the Variational Mode Decomposition (VMD) method. The decomposed sub-signals and the original signals form a new data set, which is then trained by the neural network. The decomposed sub-signals reflect the detailed features inside the power load that are difficult to be learned by the neural network. Through VMD analysis, the neural network can learn richer information, which is more effective than the superposition prediction method. The neural network prediction model selects an architecture based on Attention-long short term memory (Attention-LSTM). The addition of attention mechanism enables important decomposed information to be fully learned. The effectiveness of this method is proved by experiment

As the foundation of Posture Analysis, recognizing human activity accurately in real time assists in using machines to intellectualize living condition and monitor health status. In this paper, we focus on recognition based on raw time series data, which are continuously sampled by wearable sensors, and a fine-grained evidence reasoning approach has been proposed to produce a timely and reliable result. First, the basic time unit of input data is selected by finding a tradeoff between accuracy and time cost. Then, the approach uses Long Short Term Memory to extract features and project raw multidimensional data into probability assignments, followed by trainable evidence combination and inference network that reduce uncertainly to improve the classification accuracy. Experiments validate the effectiveness of fine granularity and evidence reasoning while the final results indicate that the recognition accuracy of this approach can reach 96.4% with no additional complexity in training.

We propose a new efficient and effective task scheduling approach with stochastic time cost for computation offloading in mobile edge computing. We developed an optimization model that minimizes the maximum tolerable delay (MTD) by considering both the average delay and delay jitter. We also proposed an efficient conservative heterogeneous earliest-finish-time algorithm to solve the MTD-minimization problem. Numerical results obtained with our proposed approach demonstrate its effectiveness over previously proposed techniques.

High-utility sequential pattern (HUSP) mining is an emerging topic in the field of knowledge discovery in databases. It consists of discovering subsequences that have a high utility (importance) in sequences, which can be referred to as HUSPs. HUSPs can be applied to many real-life applications, such as market basket analysis, e-commerce recommendations, click-stream analysis, and route planning. Several algorithms have been proposed to efficiently mine utility-based useful sequential patterns. However, due to the combinatorial explosion of the search space for low utility threshold and large-scale data, the performances of these algorithms are unsatisfactory in terms of runtime and memory usage. Hence, this article proposes an efficient algorithm for the task of HUSP mining, called HUSP mining with UL-list (HUSP-ULL). It utilizes a lexicographic $q$ -sequence (LQS)-tree and a utility-linked (UL)-list structure to quickly discover HUSPs. Furthermore, two pruning strategies are introduced in HUSP-ULL to obtain tight upper bounds on the utility of the candidate sequences and reduce the search space by pruning unpromising candidates early. Substantial experiments on both real-life and synthetic datasets showed that HUSP-ULL can effectively and efficiently discover the complete set of HUSPs and that it outperforms the state-of-the-art algorithms.

With the fast development of industrial Internet of things (IIoT), a large amount of data is being generated continuously by different sources. Storing all the raw data in the IIoT devices locally is unwise considering that the end devices’ energy and storage spaces are strictly limited. In addition, the devices are unreliable and vulnerable to many threats because the networks may be deployed in remote and unattended areas. In this paper, we discuss the emerging challenges in the aspects of data processing, secure data storage, efficient data retrieval and dynamic data collection in IIoT. Then, we design a flexible and economical framework to solve the problems above by integrating the fog computing and cloud computing. Based on the time latency requirements, the collected data are processed and stored by the edge server or the cloud server. Specifically, all the raw data are first preprocessed by the edge server and then the time-sensitive data (e.g., control information) are used and stored locally. The non-time-sensitive data (e.g., monitored data) are transmitted to the cloud server to support data retrieval and mining in the future. A series of experiments and simulation are conducted to evaluate the performance of our scheme. The results illustrate that the proposed framework can greatly improve the efficiency and security of data storage and retrieval in IIoT.

Software-defined networking (SDN) is the key outcome of extensive research efforts over the past few decades toward transforming the Internet to a more programmable, configurable, and manageable infrastructure. At the same time, SDN will surely become a new target of cyber attackers. In this paper, we point out one of the critical vulnerabilities in SDNs, the capacity of controller, which is most likely to be attacked. Due to the logical centralized management, the breakdown of a controller may disrupt a whole SDN network, which can be easily occurred by Packet-In messages flooding attack (a network-level DDoS attack). To provide a robust environment in SDN, we propose an effective detection method, which has low overhead and high accuracy. We first classify the potential switches that are compromised using Bayesian Network, which is a supervised learning algorithm. Then, we deploy the anomaly detection on the vulnerable switches to detect the Packet-In messages flooding attack based on fuzzy c-means. Extensive simulations and testbed-based experiments show that the proposed solution can defeat the Packet-In messages flooding attack with low overhead and high accuracy.

Device-to-device content sharing is widely considered as an effective response to the prevalence of multimedia content sharing and local services. To ensure its advantages, how to optimally match potential providers to demanders of contents is of importance, especially considering the randomness of content location, as well as the coexistence of altruistic, selfish, and even malicious user behaviors. Accordingly, for such a matching process, it is advisable to achieve a win–win solution, rather than sacrificing performance of either potential providers or demanders. In this work, we employ the weighted directed graph theory to model the two-sided physical–social-aware preferences, which jointly exploit the physical and social networking characteristics of both sides. Then, we study the two-sided physical–social-aware matching problem, which essentially involves the joint issue of pairing scheduling and power control with respect to potential providers and demanders. To solve it in a tractable manner, the former is cast as a two-sided one-to-one matching game, in which the potential providers and demanders rank one another by using the proposed preferences, and the latter is transformed into the preference profile establishment of this matching game. Finally, a distributed algorithm based on the Dinkelbach iteration and deferred acceptance approaches is developed. The properties of the resulting stable outcomes are then studied, and the simulation results verify the availability and efficiency of the proposed algorithm.

Inspired by security applications in the industrial Internet of things, this paper focuses on the usage of impulsive neural network (NN) synchronization technique for intelligent image protection against illegal swiping and abuse. A class of nonlinear interconnected NNs with transmission delay and random impulse effect is first formulated and analyzed in this paper. In order to make network protocols more flexible, a randomized broadcast impulsive coupling scheme is integrated into the protocol design. Impulsive synchronization criteria are then derived for the chaotic NNs in the presence of nonlinear protocol and random broadcast impulse, with the impulse effect discussed. Illustrative examples are provided to verify the developed impulsive synchronization results and to show its potential application in image encryption and decryption.

Nowadays, with the rapid advance of network technology, the miniaturization of terminal equipment, and the trend of constant reduction of costs, the topic of IoT turns increasingly popular. As IoT becomes common in daily life, one can collect information at any time and in any environment, such as temperature, humidity, and PM2.5. Based on the analysis of such data, we can learn the degree of environmental comfort of a city, which is beneficial for the development of a smart city and other applications. However, the use of a large number of terminal equipment is likely to trigger huge demands of bandwidth, followed by the increase in the time of data transmission. In order to address this issue, this study proposed a service-oriented SDN-SFC load balance mechanism. It considered and classify the type and priority of service required by each terminal device. Then, it adopted the heuristic algorithm to plan the transmission paths among SFCs to reduce the load of each SF and improve the overall network performance. The simulation results indicate that the method proposed by this study can shorten the time of data transmission and achieve load balance.

Cloud computing provides the scalable computation capability based on a virtualization technique. The energy conservation for green computing is one of the vital issues while allocating resources. To improve energy efficiency, the dynamic power-saving resource allocation (DPRA) mechanism based on a particle swarm optimization algorithm is proposed. The DPRA mechanism not only considers the energy consumption of physical machine (PM) and virtual machine (VM) but also newly tackles the energy efficiency ratio of air conditioner. Moreover, the least squares regression method is utilized to forecast PM's resource utilization for allocating VM and eliminating VM migrations. In simulation, the proposed DPRA mechanism is compared with three familiar allocation schemes and one previous solution. Simulation results show that in the presence of VM number, DPRA outperforms traditional schemes and previous solution in terms of total energy consumption (includes PMs and air conditioners), total electric bill, VM migration, and the number of shutdown PMs, chosen as objective performance metrics.

In order to optimize the resource utilization of physical machines (PMs), the workload prediction of virtual machines (VMs) is vital but challenging. Most of existing literatures focus on either resource prediction or allocation individually, but both of them are highly correlated. In this paper, we propose a multiobjective genetic algorithm (GA) to dynamically forecast the resource utilization and energy consumption in cloud data center. We formulate a multiobjective optimization problem of resource allocation, which considers the CPU and memory utilization of VMs and PMs, and the energy consumption of data center. The proposed GA forecasts the resource requirement of next time slot according to the historical data in previous time slots. We further propose a VM placement algorithm to allocate VMs for next time slot based on the prediction results of GA. In our simulation-based analysis, the optimal solution for resource prediction under stable and unstable utilization tendency is found by the proposed GA. The prediction result is superior to the previous proposed Grey forecasting model. Results show that the proposed VM placement algorithm not only increases the average utilization level of CPU and memory but also decreases the energy consumption of cloud data center.

This paper provides a training system with an augmented reality interactive body movement for movement learning, such as gymnastics, martial arts, sports or dance learners. The technology of depth image sensor is used to detect, track and measure the body’s movements to collect the path of body’s movement in 3D space, and all images has been further modified to reveal the function of feedback immediately. The learner follows up the pre-recorded tutor’s movement to imitate tutor’s movement step by step. The training system would judge whether the learner’s movement correct or not, compares with tutor’s, and offer an analysis result in-situ. The learner could get a training as well as real expert guides without any constraints of space and time and with low cost in this system.

From web search and data mining, users’ click and purchase behaviors contain valuable information, thus numerous approaches have been proposed to identify embedded useful knowledge from them. In these real-life situations, each user may perform the same action/event multiple times, and multiple accessed events product different profit. Many utility-oriented data mining approaches thus have been extensively studied. Previous studies have the limitation that the overall utility of traditional pattern is limited since they rarely consider the inherent correlation. For example, from the purchase behavior, the low-utility patterns sometimes with a very high-utility pattern will be considered as a valuable pattern even if this behavior may be not highly correlated. A more intelligent framework that provides non-redundant and correlated behavior based on utility measure is thus desired. In this paper, we first present a novel method to extract non-redundant correlated purchase behaviors considering the utility and correlation factors. The high qualified patterns can be derived with high profit and strong correlation, which can lead to higher recall and reveal better precision. In the proposed projection-based approach, an efficient projection mechanism and a sorted downward closure property are developed to reduce the database size. Several pruning strategies are further developed to efficiently and effectively discover the desired patterns. An extensive experimental study showed that the novel non-redundant correlated high-utility pattern has more effectiveness than the previous knowledge representation. Moreover, the proposed algorithm is efficient in terms of execution time and memory usage.

The papers in included in this special issue examines new and novel communication methods for smart e-health systems. Recent developments in the healthcare domain have facilitated the integration of several technologies for smart, cost-effective, reliable, and pervasive health monitoring of chronic diseases. Current research efforts focus on developing real-time communication methods, mostly for body area networks (BANs) that are used to deliver patients´ information effectively. These efforts are limited to communication within a BAN; however, less attention has been paid to connect multiple BANs to remote servers in real time. In addition, there is a limited study on the integration of BANs with different technologies including mobile cloud computing—a technology that may assist in storing and processing the huge amount of BAN data at competitive costs. Machine to machine is also considered to be a valuable paradigm in delivering BAN data to a remote server/cloud for further analysis. This may assist in reducing risks and cost of remote health monitoring. Unlike conventional research in BANs where researchers focused on individual networks, there is a need to develop innovative communication methods with a focus on complete and smart e-health systems. This smart e-health system must integrate the aforementioned technologies with multiple BANs, and provide rich healthcare services to end users. It calls for research on versatile topics ranging from physical and medium access control protocols to BANs coexistence, traffic characterization, cloud resource allocation, and cloud monitoring and maintenance.

Estimating reliable class-conditional probability is the prerequisite to implement Bayesian classifiers, and how to estimate the probability density functions (PDFs) is also a fundamental problem for other probabilistic induction algorithms. The finite mixture model (FMM) is able to represent arbitrary complex PDFs by using a mixture of mutimodal distributions, but it assumes that the component mixtures follows a given distribution, which may not be satisfied for real world data. This paper presents a non-parametric kernel mixture model (KMM) based probability density estimation approach, in which the data sample of a class is assumed to be drawn by several unknown independent hidden subclasses. Unlike traditional FMM schemes, we simply use the k-means clustering algorithm to partition the data sample into several independent components, and the regional density diversities of components are combined using the Bayes theorem. On the basis of the proposed kernel mixture model, we present a three-step Bayesian classifier, which includes partitioning, structure learning, and PDF estimation. Experimental results show that KMM is able to improve the quality of estimated PDFs of conventional kernel density estimation (KDE) method, and also show that KMM-based Bayesian classifiers outperforms existing Gaussian, GMM, and KDE-based Bayesian classifiers.

Network is an abstract expression of subjects and the relationships among them in the real-world system. Research on community detection can help people understand complex systems and identify network functionality. In this paper, we present a novel approach to community detection that utilizes a nonnegative matrix factorization (NMF) model to divide overlapping community from networks. The study is based on the different physical meanings of the pair of matrices $W$ and $H$ to optimize the constraint condition. Many community detection algorithms based on NMF require the number of known communities as a prior condition, which limits the field of application of the algorithms. This paper handled the problem by feature matrix preprocessing and ranking optimization, so that the proposed algorithm can divide the network structure with unknown community number. Experiments demonstrated that the proposed algorithm can effectively divide the community structure, and identify network overlay communities and overlapping nodes.

By collecting and analyzing a vast quantity and different categories of information, traffic flow and road congestion can be predicted and avoided in intelligent transportation system. However, how to tackle with these big data is vital but challenging. Most of the existing literatures utilized batch method to process a bunch of road data that cannot achieve real-time traffic prediction. In this paper, we use the spouts and bolts in Apache Storm to implement a real-time traffic prediction model by analyzing enormous streaming data, such as road density, traffic events, and rainfall volume. The proposed SVM-based real-time highway traffic congestion prediction (SRHTCP) model collects the road data from the Taiwan Area National Freeway Bureau, the traffic events reported by road users from the Police Broadcasting Service in Taiwan, and the weather data from the Central Weather Bureau in Taiwan. We use fuzzy theory to evaluate the traffic level of road section in real time with considering road speed, road density, road traffic volume, and the rainfall of road sections. In addition, the SRHTCP model predicts the road speed of next time period by exploring streaming traffic and weather data. Results showed that the proposed SRHTCP model improves 25.6% prediction accuracy than the prediction method based on weighted exponential moving average method under the measurement of mean absolute relative error.

As vehicle applications, mobile devices and the Internet of Things are growing fast, and developing an efficient architecture to deal with the big data in the Internet of Vehicles (IoV) has been an important concern for the future smart city. To overcome the inherent defect of centralized data processing in cloud computing, fog computing has been proposed by offloading computation tasks to local fog servers (LFSs). By considering factors like latency, mobility, localization, and scalability, this article proposes a regional cooperative fog-computing-based intelligent vehicular network (CFC-IoV) architecture for dealing with big IoV data in the smart city. Possible services for IoV applications are discussed, including mobility control, multi-source data acquisition, distributed computation and storage, and multi-path data transmission. A hierarchical model with intra-fog and inter-fog resource management is presented, and energy efficiency and packet dropping rates of LFSs in CFC-IoV are optimized.

As satellite networks have played an indispensable role in many fields, how to integrate them with terrestrial networks (e.g., the Internet) has attracted significant attention in academia. However, it is challenging to efficiently build such an integrated network, since terrestrial networks are facing a number of serious problems, and since they do not provide good support for heterogeneous network convergence. In this article, we propose a flexible network architecture, HetNet, for efficient integration of heterogeneous satellite-terrestrial networks. Specifically, the HetNet synthesizes Locator/ID split and Information-Centric Networking to establish a general network architecture. In this way, it is able to achieve heterogeneous network convergence, routing scalability alleviation, mobility support, traffic engineering, and efficient content delivery. Moreover, the HetNet can further improve its network elasticity by using the techniques of Software-Defined Networking and Network Functions Virtualization. In addition, to evaluate the HetNet performance, we build a proof-of-concept prototype system and conduct extensive experiments. The results confirm the feasibility of the HetNet and its advantages.

Frequent itemset mining (FIM) is a fundamental set of techniques used to discover useful and meaningful relationships between items in transaction databases. In recent decades, extensions of FIM such as weighted frequent itemset mining (WFIM) and frequent itemset mining in uncertain databases (UFIM) have been proposed. WFIM considers that items may have different weight/importance. It can thus discover itemsets that are more useful and meaningful by ignoring irrelevant itemsets with lower weights. UFIM takes into account that data collected in a real-life environment may often be inaccurate, imprecise, or incomplete. Recently, these two ideas have been combined in the HEWI-Uapriori algorithm. This latter considers both item weights and transaction uncertainty to mine the high expected weighted itemsets (HEWIs) using a two-phase Apriori-based approach. Although the upper-bound proposed in HEWI-Uapriori can reduce the size of the search space, it still generates a large amount of candidates and uses a level-wise search. In this paper, a more efficient algorithm named HEWI-Utree is developed to efficiently mine HEWIs without performing multiple database scans and without generating candidates. This algorithm relies on three novel structures named element (E)-table, weighted-probability (WP)-table and WP-tree to maintain the information required for identifying and pruning unpromising itemsets early. Experimental results show that the proposed algorithm is generally much more efficient than traditional methods for WFIM and UFIM, as well as the state-of-the-art HEWI-Uapriori algorithm, in terms of runtime, memory consumption, and scalability.

Device-to-device (D2D) communication is one of the promising technologies in the 5th generation of wireless communication (5G). This paper investigates the achievable transmission capacity for multi-mode D2D communication. By utilizing stochastic geometry tools, we first provide the system model and derive the transmission capacity for two different sharing schemes of underlaying D2D mode. The first one is that each cellular spectrum can be reused by single D2D pair and the other one considers the multiple sharing among several D2D pairs with the same spectrum. In addition, we analyze the strategy for the optimal selection between both different sharing scenarios. Secondly, we analyze the transmission capacity for overlay D2D mode. Finally, with the optimal relay selection strategy, the maximization of transmission capacity for relay-assisted D2D communication is obtained under the constraint of minimum energy consumption. Furthermore, simulations are carried out and the results demonstrate the effectiveness and validity of our analysis.

With the progress of network technology in recent years, multimedia streaming applications have become increasingly popular. However, it is difficult to achieve quality of service and efficiency for multimedia streaming over vehicular networks because of the high mobility feature. Over the existing network architecture, it is difficult to immediately analyze the status of the entire network, and then establish the rules of allocation and management. However, the novel network architecture, software-defined networking, offers other options for making network management more efficient, especially for the 5G network environment. Hence, a buffer-aware streaming approach is proposed to allow users to play multimedia streaming over vehicular 5G networks, in the case of handover between different eNodeBs, to achieve minimum delay and have better quality of service. According to the user's mobility information, the status of the player buffer, and the current strength of the network signal, the proposed approach can provide the transmission strategy of multimedia streaming to the SDN controller. Finally, the experimental results proved that the proposed approach is able to not only adjust the priority of streaming content segments with the buffer and mobility status of user equipment to effectively retain overall streaming services quality, but also avoid the delay of streaming content transmission for 5G vehicular networks.

The latest High Efficiency Video Coding (HEVC) standard could achieve the highest coding efficiency compared with the existing video coding standards. To improve the coding efficiency of the intra frame, a quad-tree-based variable block size coding structure that is flexible to adapt to various texture characteristics of images and up to 35 intra-prediction modes for each prediction unit (PU) is adopted in HEVC. However, the computational complexity is increased dramatically because all the possible combinations of the mode candidates are calculated in order to find the optimal rate distortion cost using the Lagrange multiplier. To alleviate the encoder computational load, this paper proposes an adaptive mode decision algorithm based on texture complexity and direction for HEVC intra prediction. First, an adaptive coding unit selection algorithm according to each depth levels’ texture complexity is presented to filter out unnecessary coding block. Then, the original redundant mode candidates for each PU are reduced according to its texture direction. The simulation results show that the proposed algorithm could reduce around 56% encoding time on average while maintaining the encoding performance efficiently with only a 1.0% increase in BD-rate compared with the test model HM16 of HEVC.

This study constructs an approach to reproduce the real-time falls of humans, which uses a triaxial accelerometer and triaxial gyroscope to detect the occurrence of a fall, and an attitude algorithm to estimate the angles of each part of the human body, where Internet of healthcare things collects the information of each sensor, and a Bayesian Network deduces the next action. Inferential Bayesian probability could present more complete data of a fall to healthcare providers. Even if the data are damaged by the transmission network or equipment, the next action still could be deduced by Bayesian probability, and because the fall could be reproduced in a 3D Model on the client side, the fall occurrence is shown more intuitively, and could thus serve as reference for first aid.

The amount of data generated by individuals and enterprises is rapidly increasing. With the emerging cloud computing paradigm, the data and corresponding complex management tasks can be outsourced to the cloud for the management flexibility and cost savings. Unfortunately, as the data could be sensitive, the direct data outsourcing would have the problem of privacy leakage. The encryption can be used, before the data outsourcing, with the concern that the operations can still be accomplished by the cloud. We consider the multikeyword similarity search over outsourced cloud data. In particular, with the consideration of the text data only, multiple keywords are specified by the user. The cloud returns the files containing more than a threshold number of input keywords or similar keywords, where the similarity here is defined according to the edit distance metric. We propose three solutions, where blind signature provides the user access privacy, and a novel use of Bloom filter's bit pattern provides the speedup of search task at the cloud side. Our final design to achieve the search is secure against insider threats and efficient in terms of the search time at the cloud side. Performance evaluation and analysis are used to demonstrate the practicality of our proposed solutions.

As an important component of the intelligent transportation system, internet-of-vehicles technology has attracted considerable attention. The dissemination of event-driven emergency messages with high reliability and low delay is vital to ensure traffic safety and improve traffic efficiency in urban environment. In this study, the authors propose a reliable emergency message dissemination protocol taking into account the urban road characteristics and scalability requirements, which consists of a layout-aware ready-to-broadcast-emergency-message and clear-to-broadcast-emergency-message handshake mechanism and a redundant relay node adaptation mechanism. Finally, the simulation results confirm the feasibility and effectiveness of the proposed scheme.

With hard requirements of high performance for the next generation mobile communication systems, especially 5G networks, coverage has been the crucial problem which requires the deployment of more stations by the service providers. However, this deployment of new stations is not cost effective and requires network replanning. This issue can easily be overcome by the use of Unmanned Aerial Vehicles (UAVs) in the existing communication system. Thus, considering this as a problem, an intelligent solution is presented for the accurate and efficient placement of the UAVs with respect to the demand areas resulting in the increase in the capacity and coverage of the wireless networks. The proposed approach utilizes the priority-wise dominance and the entropy approaches for providing solutions to the two problems considered in this paper, namely, Macro Base Station (MBS) decision problem and the cooperative UAV allocation problem. Finally, network bargaining is defined over these solutions to accurately map the UAVs to the desired areas resulting in the significant improvement of the network parameters, namely, throughput, per User Equipment (UE) capacity, 5th percentile spectral efficiency, network delays and guaranteed signal to interference plus noise ratio by 6.3%, 16.6%, 55.9%, 48.2%, and 36.99%, respectively in comparison with the existing approaches.

Weighted Frequent Itemset Mining (WFIM) has been proposed as an extension of frequent itemset mining that considers not only the frequency of items but also their relative importance. However, using WFIM algorithms in real applications raises some problems. First, they do not consider how recent the patterns are. Second, traditional WFIM algorithms cannot handle uncertain data, although this type of data is common in real-life. To address these limitations, this paper introduces the concept of Recent High Expected Weighted Itemset (RHEWI), which considers the recency, weight and uncertainty of patterns. By considering these three factors, more up-to-date and relevant results are found. A projection-based algorithm named RHEWI-P is presented to mine RHEWIs using a novel upper-bound downward closure (UBDC) property. An improved version of this algorithm called RHEWI-PS is further proposed based on a novel sorted upper-bound downward closure (SUBDC) property for pruning unpromising candidate itemsets early. An experimental evaluation against the state-of-the-art HEWI-Uapriori algorithm was carried out on both real-world and synthetic datasets. Results show that the proposed algorithms are highly efficient and are acceptable for mining the desired patterns.

Currently, almost every family has at least one car; thus, vehicle density is increasing annually. However, road capacity is finite; consequently, traffic accident frequency may increase due to increasing vehicle density. Typically, car accidents result in traffic congestion because vehicles behind the accident are not aware of the event and continue to follow the front queue. To address this problem, some emergency services, such as emergency message broadcasting, have been proposed. However, not all drivers want to receive such messages because they intend to exit the route prior to the accident scene, which means that communication resources may be wasted. In this paper, we propose a prediction model to forecast vehicles behavior based on a Markov chain and identify which vehicles require the emergency message. In addition, the proposed model includes an efficient policy based on the shortest path for police cars and ambulances such that they can attend the accident scene quickly and relieve traffic congestion. Simulation results show that the proposed method reduces unnecessary message transmission and increases road utilization efficiently.

Frequent pattern mining (FPM) is an important topic in data mining for discovering the implicit but useful information. Many algorithms have been proposed for this task but most of them suffer from an important limitation, which relies on a single uniform minimum support threshold as the sole criterion to identify frequent patterns (FPs). Using a single threshold value to assess the usefulness of all items in a database is inadequate and unfair in real-life applications since each item is different and not all items should be treated as the same. Several algorithms have been developed for mining FPs with multiple minimum supports but most of them suffer from the time-consuming problem and require a large amount of memory. In this paper, we address this issue by introducing the novel approach named Frequent Pattern mining with Multiple minimum supports from the Enumeration-tree (FP-ME). In the developed Set-Enumeration-tree with Multiple minimum supports (ME-tree) structure, a new sorted downward closure (SDC) property of FPs and the least minimum support (LMS) concept with multiple minimum supports are used to effectively prune the search space. The proposed FP-ME algorithm can directly discover FPs from the ME-tree without candidate generation. Moreover, an improved algorithm, named FP-MEDiffSet, is also developed based on the DiffSet concept, to further increase mining performance. Substantial experiments on both real-life and synthetic datasets show that the proposed algorithms can not only avoid the rare item problem, but also efficiently and effectively discover the complete set of FPs in transactional databases while considering multiple minimum supports and outperform the state-of-the-art CFP-growth++ algorithm in terms of execution time, memory usage and scalability.

Coordination among multiple ad hoc systems can extend the applications of the networks operating without any requirement of centralized devices. One of the coordination systems is formed by the collaboration of aerial ad hoc network and ground ad hoc network that performs cognitive coordination to form an effective guidance system despite operating with different configurations and dynamics. For guidance system, unmanned aerial vehicles form a cognitive map and relay them effectively to the intended node which is a part of ground ad hoc unit. The data sharing is difficult between the networks operating under different circumstances and configurations. An efficient approach is required to provide low-complex relaying for faster communications in these coordinated networks. One of the solutions can be found from the fuzzy formation of the problem that can be applied to some optimization algorithms to efficiently select the cognitive corridors intended for enhanced transmission. In this paper, a fuzzy-based bee colony optimization algorithm is proposed which operates over cooperative trust value, cognitive relaying value and situational awareness to provide efficient cognitive relaying between the two ad hoc units, and uses a cloud server to provide coordination between the nodes. The effectiveness of the proposed scheme and its applicability is tested using simulations.

Software defined networking (SDN) paradigm currently does not integrate with any debugging or diagnosing mechanism. Services through Internet connection may not work properly due to network problems, such as loop and black hole problems. The study introduces sPing, a user-centred debugging mechanism for users debugging and troubleshooting network problems in SDN. The sPing diagnoses problems in data plane by collecting and recording specific packets based on programmable capability. It not only detects network loop and black hole but also discovers the hidden information of data link layer. The discovery of layer 2 information is unachievable to existing SDN-based debuggers. The authors experiment the sPing on three different brands of SDN switches for verifying its proof of concept, and simulate it in the Mininet emulator for evaluating its performance. Results showed that the proposed sPing mechanism assists SDN users in debugging network problems and in maintaining network functionalities.