Search

With the growth and expansion of cloud data centers, energy consumption has become an urgent issue for smart cities system. However, most of the current resource management approaches focus on the traditional cloud computing scheduling scenarios but fail to consider the feature of workloads from the Internet of Things (IoT) devices. In this paper, we analyze the characteristic of IoT requests and propose an improved Poisson task model with a novel mechanism to predict the arrivals of IoT requests. To achieve the trade-off between energy saving and service level agreement, we introduce an adaptive energy efficiency model to adjust the priority of the optimization objectives. Finally, an energy-efficient virtual machine scheduling algorithm is proposed to maximize the energy efficiency of the data center. The experimental results show that our strategy can achieve the best performance in comparison to other popular schemes.

The energy consumption issue of large-scale data centers is attracting more and more attention. Virtual machine consolidation can significantly reduce energy consumption by migrating virtual machines from one physical machine to another. However, excessive virtual machine consolidation can lead to dangerous Service Level Agreement (SLA) violations. Therefore, how to balance between effective energy consumption and SLA violations avoidance effectively is a paradox to be mediated. The virtual machine consolidation problem is NP-hard. The traditional heuristic algorithm is easy to fall into the local optimal and some meta-heuristic algorithms can help to avoid it. However, the existing meta-heuristic algorithms are with high complexity. Therefore, we propose a lower complexity multi-population ant colony system algorithm with the Extreme Learning Machine (ELM) prediction (ELM_MPACS). The algorithm firstly predicts the host state employing ELM and then the virtual machine on the overloaded host will be migrated to the normal host, while the virtual machine on the underloaded host will be consolidated to another underloaded host with higher utilization. Multiple populations concurrently construct migration plans and local search further optimizes the results obtained by each population to reduce SLA violations. We compare ELM_MPACS with the benchmark, heuristic and meta-heuristic algorithms. The experimental results have shown that compared with these algorithms, our algorithm reduces energy consumption, migration times and SLA violations effectively.

Multivariate Time Series Forecasting (MTSF) has recently emerged its growing importance in many industries. However, how to reduce the influence of the noise components existing in time series on prediction and extract features effectively are still two challenges in MTSF. This paper focuses on those two challenges and proposes a new prediction method based on decomposition-ensemble framework called adaptive sub-series clustering-stacked residual LSTMs-multi-level attention mechanism (ASC-SRLSTMs-MLAttn). The method consists of three stages: decomposition, prediction, and ensemble. In the decomposition stage, the target series is decomposed by Ensemble Empirical Mode Decomposition into multiple sub-series, which will be clustered and reconstructed by the ASC algorithm to reduce the complexity and the time consumption of prediction. In the prediction stage, the sub-series and correlation series will be fed into SRLSTMs-MLAttn for sub-series prediction. The model is based on the encoder-decoder architecture with stacked residual LSTMs as the encoder, which can effectively capture the dependencies among multi variables and the temporal features from multivariate time series. Besides, a multi-level attention mechanism (MLAttn), which makes full use of the encoding information of the encoder, has been introduced to further improve the prediction performance of the model. In the ensemble stage, the predicted values of each sub-series will be summed to obtain the final prediction of the target time series. We also demonstrate the superiority and effectiveness of our proposed method on four public datasets via the conducted comparison experiment and ablation study.

Scene text detection is to detect the position of a text in the natural scene, the quality of which will directly affect the subsequent text recognition. It plays an important role in fields such as image retrieval and autopilot. How to perform multi-scale and multi-oriented text detection in the scene still remains as a problem. This paper proposes an effective scene text detection method that combines the convolutional neural network (CNN) and recurrent neural network (RNN). In order to better adapt to texts in different scales, feature pyramid networks (FPN) have been applied in the CNN part to extract multi-scale features of the image. We then utilize bidirectional long-short-term memory (Bi-LSTM) to encode these features to make full use of the text sequence characteristics with the outputs as a series of text proposals. The generated proposals are finally linked into a text line through a well-designed text connector, which can be flexibly adapted to any oriented texts. The proposed method is evaluated on three public datasets: ICDAR2013, ICDAR2015, and USTB-SV1K. For ICDAR2013 and USTB-1K, we have reached 92.5% and 62.6% F-measure, respectively. Our method has reached 72.8% F-measure on the more challenging ICDAR2015 which demonstrates the effectiveness of our method.

Multivariate time series forecasting recently has received extensive attention with its wide application in finance, transportation, environment, and so on. However, few of the currently developed models have considered the impact of noise on prediction. Since multivariate time series contains multiple subsequences with strong nonlinear fluctuations, it is also difficult to obtain satisfactory prediction results. In this paper, aiming at improving prediction performance, we have proposed a novel ensemble three-phase model called adaptive noise reducer-stacked auto-encoder-validating-AdaBoost-based long short-term memory (ANR-SAE-VALSTM). We start with an introduction of a novel ANR for time series noise elimination. The SAEs are then used to extract features from the de-noised multivariate time series. Finally, we feed the de-noised features into the VALSTM to train an ensemble over-fitting prevention predictor. The proposed model is employed on the Beijing PM2.5 dataset and GEFCom2014 Electricity Price dataset. Compared with other popular models, the proposed model has achieved the best prediction performance in all prediction horizons. In addition, a careful ablation study is conducted to demonstrate the efficiency of our model design.

The pickup and delivery problem with time windows and last-in-first-out (LIFO) loading (PDPTWL) is a combinational optimization problem extended from the well-known vehicle routing problem (VRP), in which the type of customer point is no longer single and the loading order of the requests must meet the LIFO constraint. Due to its NP-hard nature, it is difficult for exact algorithms and heuristics with a linear structure to solve a large-scale problem in a reasonable time. In this paper, we propose a fast decomposition and reconstruction framework (D&R) to solve the PDPTWL with high quality in a relatively short time. An angle-based sweep method is used to decompose a complete solution into multiple sub-solutions, each of which is assigned to a tabu search for optimization. To speed up the whole process, the optimization procedure of sub-solutions is performed by different processors of multi-core CPU in parallel. Three neighborhood operators and three strategies to reduce the number of vehicles are designed to cope with the tabu search for further improvement. Moreover, the adaptive memory mechanism is added to provide a better start when the optimization procedure falls into the local optima. We compare our framework against the best known solutions on 119 instances with up to 300 requests, the results show that our framework is able to improve over 85% (107 out of 119) of the best known solutions. More specifically, the number of vehicles is optimized by about 60% (74 out of 119) and the driving distance by about 50% (59 out of 119). In addition on instances with the largest size of requests, the computational time of our framework can be 1/50 of the comparative results, confirming its efficiency.

As a medium of information transmission, text is widely existed in natural scenes, showing diversity in orientation, scale, font, color and shape. Accurate detection of scene text is a prerequisite for subsequent recognition. Though many previous methods have worked well on horizontal and multi-oriented text detection datasets, detecting arbitrary shape scene text still remains as a challenging problem. To solve the problem, this paper proposes an arbitrary shape scene text detection method. Based on Mask R-CNN, our method replaces the original ℓ1-smooth loss with the proposed IoU-related loss and adds a text scoring branch to align the confidence score with the text mask IoU to make the model highly relevant to IoU, achieving the goal of improving detection performance by improving IoU directly in a simple but effective way. The proposed method is evaluated on four public datasets: CTW-1500, Total-Text, ICDAR2015 and ICDAR2017-RCTW. For curved text detection datasets CTW-1500 and Total-Text, we have reached 79.2% and 81.1% H-mean respectively, showing that the proposed method has achieved competitive performance in arbitrary scene text detection.

Energy conservation is one of the most important challenges in wireless sensor networks (WSNs). Therefore, compared with the traditional networks, the WSNs not only need high-quality services with high throughput or low transmission delay, but also pay greater attention to energy utilization to extend network lifetime. The clustering routing algorithm is considered to be among the effective ways to collect and transmit data in WSNs. Cluster head (CH) plays a vital role in the cluster which is in charge of data aggregation and data transmission, so their energy consumption is higher than non-CH nodes. The traditional clustering algorithm tends to have the same size in each cluster. However, due to the randomness of the node distribution, the equal clustering mechanism obviously cannot reduce energy consumption. In order to solve this problem, this paper contributes a new unequal clustering algorithm, an energy-aware adaptive kernel density estimation algorithm (EAKDE), which aims to balance the energy dissipation among the CHs. EAKDE utilizes fuzzy logic to determine the priority of nodes competing for CH. In order to adapt the dynamic change of node conditions, adaptive kernel density estimation algorithm is utilized to assign the appropriate unequal cluster radius to sensor nodes. The simulation results demonstrate that, in different scenarios, EAKDE outperforms the other well-known algorithms in terms of network stability, network lifetime, and energy efficiency.

Service discovery is vital to event handling in Internet of Things applications which are based on the event-driven service-oriented architecture. However, in service discovery, the problem of service matching that establishes relationships between services and events has been seldom investigated through a semantic way. In this paper, to facilitate the efficiency of service discovery triggered by events, we propose a novel method of semantic service matching based on word embeddings. In this method, two types of semantic services about events (i.e., event-recognition services and event-handing services) are specified and matched through semantic similarity assessment that is conducted with word embeddings. Besides, to obtain highquality word embeddings, we present a hybrid approach for learning word embedding which treats words in distinct means according to word frequency. Experiments demonstrated on different data sets show that our method of semantic service matching is an effective way to facilitate event-driven service discovery, and the proposed training approach for word embeddings outperforms existing works and is able to improve the accuracy of event-driven service discovery.

The accuracy and efficiency of radio-frequency identification device tag localization is significant to a large number of existing and upcoming radio-frequency identification device applications. In this article, we have introduced a novel method to locate a radio-frequency identification device tag in three-dimensional space. The proposed method uses readers with multiple power levels to locate the passive radio-frequency identification device tags. First, we preprocess the data and estimate the distances between the readers and target tag. Then, we use sphere trilateration method to estimate the location of the radio-frequency identification device tags in three-dimensional space. Finally, we use planar fit method and the method of least squares to reduce the localization error. Both the simulation and testbed experiment results have shown that the proposed method can reach a higher positioning accuracy with less radio-frequency identification device readers and more robust to the environment than other existing methods.

The research of localization technology based on received signal strength and machine learning has recently attracted a lot of attentions, since with the help of enough labeled training data this technology is able to achieve high positioning accuracy. However, it is an expensive job to collect enough labeled training data in the broad outdoor space. In order to reduce the cost of building and maintaining training database, semi-supervised extreme learning machine is applied to solve the cellular network localization in this article. However, the performance of this algorithm is sensitive to the values of the hyper parameters. Without any systematic guidance, the optimal hyper parameters can only be selected by experienced workers through trial and error. To address this problem, we propose a novel algorithm by combining particle swarm optimization and semi-supervised extreme learning machine to automatically select the optimal hyper parameters of semi-supervised extreme learning machine in this article. The experiments demonstrate that applying particle swarm optimization in our optimization framework makes the hyper parameters of semi-supervised extreme learning machine algorithm self-adaptive in different conditions. Moreover, the proposed method is more stable than the general semi-supervised extreme learning machine and outperforms other compared methods.

The implementation of IP technology in wireless sensor networks has promoted the development of many smart scenarios. To enhance secure access in IP-enabled wireless sensor networks, access control to sensor nodes is a necessary process. However, access control currently faces two challenges, feasibility and preservation of user access privacy. In this paper, we propose eHAPAC, a novel privacy-preserving access control model for IP-enabled wireless sensor networks. The contributions of our paper include three parts. First, this paper integrates the Hidra access control protocol and APAC privacy-preserving model, addressing the issue of privacy-preserving access control in resource-constrained devices. Second, this paper proposes an enhanced Hidra protocol to implement the unlinkability of protocol message exchanges. Third, to solve the problem of third party credibility, this paper improves the group signature-based APAC model and utilizes blockchain technology to manage the storage and publication of public group signature keys. Security analysis and performance evaluation prove that our protocol is secure and effective.

As an emerging and promising computing paradigm in the Internet of things (IoT),edge computing can significantly reduce energy consumption and enhance computation capabilityfor resource-constrained IoT devices. Computation offloading has recently received considerableattention in edge computing. Many existing studies have investigated the computation offloadingproblem with independent computing tasks. However, due to the inter-task dependency in variousdevices that commonly happens in IoT systems, achieving energy-efficient computation offloadingdecisions remains a challengeable problem. In this paper, a cloud-assisted edge computing frameworkwith a three-tier network in an IoT environment is introduced. In this framework, we first formulatedan energy consumption minimization problem as a mixed integer programming problem consideringtwo constraints, the task-dependency requirement and the completion time deadline of the IoT service.To address this problem, we then proposed an Energy-efficient Collaborative Task ComputationOffloading (ECTCO) algorithm based on a semidefinite relaxation and stochastic mapping approachto obtain strategies of tasks computation offloading for IoT sensors. Simulation results demonstratedthat the cloud-assisted edge computing framework was feasible and the proposed ECTCO algorithmcould effectively reduce the energy cost of IoT sensors.

With the rapid development of indoor positioning technology, radio frequency identification (RFID) technology has become the preferred solution due to its advantages of non-line-of-sight, non-contact and rapid identification. However, the accuracy of existing RFID indoor positioning algorithms is easily affected by the tag density and algorithm efficiency, and their environmental robustness is not strong enough. In this paper, we have introduced an RFID positioning algorithm based on the Glowworm Swarm Optimization (GSO) fused with semi-supervised online sequential extreme learning machine (SOS-ELM), which is called the GSOS-ELM algorithm. The GSOS-ELM algorithm automatically adjusts the regularization weights of the SOS-ELM algorithm through the GSO algorithm, so that it can quickly obtain the optimal regularization weights under different initial conditions; at the same time, the semi-supervised characteristics of the GSOS-ELM algorithm can significantly reduce the number of labeled reference tags and reduce the cost of positioning systems. In addition, the online learning phase of the GSOS-ELM algorithm can continuously update the system to perceive changes in the environment and resist the environmental interference. We have carried out experiments to study the influence factors and validate the performance, both the simulation and testbed experiment results show that compared with other algorithms, our proposed GSOS-ELM localization system can achieve more accurate positioning results and has certain adaptability to the changes of the environment.

In heterogeneous vehicular networks, the communications among moving vehicles (MV) and between vehicles and the roadside infrastructures have posed a challenge for providing continuous mobility services. FMIPv6 looking toward the support of a timely and precise link layer (L2) trigger provides a fundamental mobility management approach. The time stamp of L2 trigger hence is crucial for guaranteeing the handover performance but is difficult to be obtained in the practical application scenarios. The unsatisfactory handover trigger includes the late L2 trigger and the premature L2 trigger. The former would cause a missed detection deteriorating the perceived quality of service, and the latter would cause a false alarm with serious performance loss and resources waste. To address this issue, we propose a solution to optimize the unsatisfactory handover trigger (O-UHT), including promptly determining the unsatisfactory handover trigger and elaborately developing the specific measures to reduce the performance loss caused by the above two cases. The simulations conducted on NS-2 platform have verified the efficiency.

To facilitate the efficient use of information generated from the Internet of Things (IoT) by sensors and automatically drive entities to meet the needs of human beings, this paper presents an information processing framework using semantic web technologies. The framework represents the entities as semantic web services and automatically creates service sequences as transactions by analyzing the information need in specific queries. Moreover, the framework ensures the effective management and control of information collected by mobile sensors during the transaction. The foundation of this framework lies in a set of ontologies we proposed: goal ontology, role ontology, constraint ontology, message ontology, status ontology, space-time ontology, and activity ontology. In addition to their respective definitions, we also present a detailed exploration of the relationships between them. Finally, we validate the effectiveness of the framework and the ontologies with a case study of the logistics system. The evaluation shows that the ontologies support the framework well and our approach increases the logistics system efficiency with lower operating costs.

Event recognition in smart spaces is an important and challenging task. Most existing approaches for event recognition purely employ either logical methods that do not handle uncertainty, or probabilistic methods that can hardly manage the representation of structured information. To overcome these limitations, especially in the situation where the uncertainty of sensing data is dynamically changing over the time, we propose a multi-level information fusion model for sensing data and contextual information, and also present a corresponding method to handle uncertainty for event recognition based on Markov logic networks (MLNs) which combine the expressivity of first order logic (FOL) and the uncertainty disposal of probabilistic graphical models (PGMs). Then we put forward an algorithm for updating formula weights in MLNs to deal with data dynamics. Experiments on two datasets from different scenarios are conducted to evaluate the proposed approach. The results show that our approach (i) provides an effective way to recognize events by using the fusion of uncertain data and contextual information based on MLNs and (ii) outperforms the original MLNs-based method in dealing with dynamic data.

Mobile sinks can achieve load-balancing and energy-consumption balancing across the wireless sensor networks (WSNs). However, the frequent change of the paths between source nodes and the sinks caused by sink mobility introduces significant overhead in terms of energy and packet delays. To enhance network performance of WSNs with mobile sinks (MWSNs), we present an efficient routing strategy, which is formulated as an optimization problem and employs the particle swarm optimization algorithm (PSO) to build the optimal routing paths. However, the conventional PSO is insufficient to solve discrete routing optimization problems. Therefore, a novel greedy discrete particle swarm optimization with memory (GMDPSO) is put forward to address this problem. In the GMDPSO, particle’s position and velocity of traditional PSO are redefined under discrete MWSNs scenario. Particle updating rule is also reconsidered based on the subnetwork topology of MWSNs. Besides, by improving the greedy forwarding routing, a greedy search strategy is designed to drive particles to find a better position quickly. Furthermore, searching history is memorized to accelerate convergence. Simulation results demonstrate that our new protocol significantly improves the robustness and adapts to rapid topological changes with multiple mobile sinks, while efficiently reducing the communication overhead and the energy consumption.