Forthcoming articles


International Journal of Sensor Networks


These articles have been peer-reviewed and accepted for publication in IJSNet, but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.


Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.


Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.


Articles marked with this Open Access icon are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.


Register for our alerting service, which notifies you by email when new issues of IJSNet are published online.


We also offer RSS feeds which provide timely updates of tables of contents, newly published articles and calls for papers.


International Journal of Sensor Networks (26 papers in press)


Regular Issues


  • Recent Advances in Wireless Sensor Networks with Environmental Energy Harvesting
    by Lei Shu, Wanjiun Liao, Jaime Lloret, Lei Wang 
    Keywords: .

  • A theoretical investigation on moving average filtering solution for fixed-path map matching of noisy position data   Order a copy of this article
    by Baris Baykant Alagoz 
    Abstract: Precisely estimation of moving object locations from position sensors promises useful implications for many fields of engineering. The mapping of a moving object on a predefined path is an important process for object tracking and remote control applications. Due to measurement noises of sensors and uncertainties, the measured object location may not precisely match to paths or roads in a map. This study presents a numerical method for a low computational-complexity solution of point to arc type mapping problems. This method has two main tasks: a noise reduction task by short-time moving average filtering of noisy two-dimensional position data, and a map matching task to estimate exact position of an object on a map. To evaluate performance of the investigated method, the algorithm is applied for bus route tracking simulations and results are discussed for several road scenarios and various levels of noise.
    Keywords: Map matching; noisy position data; short-time moving average filtering; point to arc mapping; object tracking.

  • A New Optimization Model and Algorithm for Virtual Optical Networks   Order a copy of this article
    by Shanshan Hao, Yuping Wang, Beicong Lv 
    Abstract: Virtual Optical Network is considered to be the next generation internet architecture and is of big value. Currently, it has many issues to be solved, e.g., node mapping, link mapping and spectrum assignment. More and more challenges will be faced when there are privileged nodes through which certain connection requests can not pass in the physical topology. To solve the above issues, this paper put forward a new optimization model in virtual optical network considering privileged nodes, and aims to minimize the maximum index of used frequency slots. A routing algorithm is designed for the model, an adaptive strategy is put forward to determine the crossover rate and mutation rate, whats more, a new crossover and mutation operator are proposed. Based on these, a novel genetic algorithm is proposed to solve the problem. To evaluate the efficiency of the model and the algorithm, numerous experiments are conducted on four widely used network topologies. Simulation results show the feasibility of the model and the effectiveness of the algorithm.
    Keywords: Network Scheduling; Optimization Model; Virtual Optical Network; Genetic Algorithm.

  • A Mixed-Coding Adaptive Differential Evolution for Optimizing the Architecture and Parameters of Feedforward Neural Networks   Order a copy of this article
    by Li Zhang, Hong Li 
    Abstract: This paper presents an adaptive differential evolution with mixed-coding strategy to evolve feedforward neural networks (FNNs). This algorithm with adaptive control parameters which can handle effectively binary variables and real variables, is used to optimize simultaneously FNN architecture and connection parameters (weights and biases) by a specific individual representation and evolutionary scheme. The performance of the algorithm has been evaluated on several benchmarks. The results demonstrate that the proposed algorithm can produce compact FNNs with good generalization ability.
    Keywords: Feedforward Neural Network; Evolutionary Neural Network; Differential Evolution; Generalization Ability.

  • Maximization of the number of $beta$-view covered targets in visual sensor networks   Order a copy of this article
    by Ling Guo, Deying Li, Yongcai Wang, Zhao Zhang, Guangmo Tong, Weili Wu, Dingzhu Du 
    Abstract: In some applications using visual sensor networks, the facing directions of targets are bounded. Existing full-view coverage (all the facing directions of a target constitutes a disk) is not necessary. Therefore, a novel model called $beta$-view coverage, where only necessary facing directions of a target are effectively viewed, is proposed but uses much fewer cameras than those used by full-view coverage model.rnrnBased on this new model, a new problem called $beta$-view covered Target Maximization (BVCTM) problem is proposed to maximize the number of $beta$-view covered targets given some fixed and freely rotatable camera sensors. The NP-hardness of this problem is proven. We transform BVCTM problem into an Integer Linear Programming problem equivalently. Thereafter, a $displaystyle{(1-e^{-1})}$-factor approximate algorithm and a camera-utility based greedy algorithm are given to it. Finally, we conduct many experiments and investigate the influence of many parameters on these two algorithms.
    Keywords: Visual Sensor Networks; Target Coverage; $beta$ View Coverage.

  • Quality of Barrier Cover with Wireless Sensors   Order a copy of this article
    by Weili Wu, Zhao Zhang, Chuangen Gao, Hai Du, Hua Wang, Ding-Zhu Du 
    Abstract: A set of wireless sensors is called a barrier cover if they can monitor the boundary of an area so that an intruder cannot enter the area without being found by any sensor. The quality of a barrier cover is the shortest length of path along which an intruder can enter the area from outside. This shortest length is called the width of the barrier cover. Given a set of sensors, how to find the minimum subset of sensors which form a barrier cover with required width? How to find the maximum number of disjoint subsets of sensors, each form a barrier cover with required width? In this paper, we study these two problems. We introduce a concept of $w$-simple barrier cover, which is a barrier cover with width $w$, consisting of a sequence of sensors as shown in Fig.2. For $w$-simple barrier covers, we design polynomial-time algorithms for solving above two problems. In general, we show the NP-hardness of above two problems and design efficient heuristics to solve them; the performance of these heuristics are evaluated by computational experiments.
    Keywords: wireless sensor network; barrier cover; computational complexity; algorithm.

  • Partition-aware centrality measures for connectivity restoration in mobile sensor networks   Order a copy of this article
    by Izzet Fatih Senturk 
    Abstract: Mobile Sensor Networks (MSNs) often operate unattended in environments where human intervention is limited. To sustain network operations, network connectivity must be maintained at all times. However, the network can be partitioned due to random node failures. To tolerate such failures in a reactive manner, network topology can be restructured through node mobility. Minimizing the mobility cost requires addressing two different challenges. First, identifying nodes to be relocated. Second, determining target locations for movement. We address the first problem by presenting three different partition-aware centrality measures based on closeness centrality, geometric centrality, and harmonic centrality. To determine the movement target, we consider the former locations of the upstream nodes so that simultaneous node failures can be tolerated with limited data collection scope. The approaches that we present in this paper not only ensure recovery but also minimize the recovery cost so that the network lifetime is extended.
    Keywords: Mobile sensor networks; topology management; closeness centrality; geometric centrality; harmonic centrality; connectivity restoration; fault tolerance; mobility.

  • An intelligent monitoring system for a pig breeding environment based on a wireless sensor network   Order a copy of this article
    by Chong Chen, Xingqiao Liu 
    Abstract: The piggery environment is an important factor affecting health level, growth and development of pigs, which plays a decisive role in safe pork production. To facilitate environmental monitoring for large-scale pig farms, an environmental monitoring system of air quality parameters for livestock breeding is developed based on a wireless sensor network (WSN). The system could accomplish distributed measurement, data storage, and centralized management of air quality parameters such as temperature, humidity, ammonia and carbon dioxide, and could also automatically control the work of fan and wet curtain. As the piggery environment is a multi-variable, nonlinear and time-varying system, there are coupling effects among them. It is difficult to establish a precise mathematical model. Therefore, neural decoupling fuzzy control technology is employed. Fuzzy controllers are designed to control temperature and humidity, respectively. In order to reduce the coupling effect between them, a compensation decoupling controller is designed, which adjusts weight coeffients of neural networks and compensates for the effects of mutual coupling. Experimental results indicated that, compared with the data from an air quality analyzer, measurement
    Keywords: intelligent monitoring; environmental control; livestock breeding; WSN; wireless sensor network; fuzzy control; decoupling control.

  • An energy-saving wireless sensor network based model for monitoring of ammonia concentration   Order a copy of this article
    by Chong Chen, Xingqiao Liu, Chengyun Zhu 
    Abstract: The ammonia concentration in the piggery plays a key role in the growth of fattening pigs in livestock breeding. In this study, we propose an intelligent environmental monitoring system for piggery based on a wireless sensor network, which achieves multi-point, large-scale environmental monitoring in piggery. Specifically, a model has been developed to predict environmental parameters in the server. Due to the determining role of ammonia concentration on pig health, this model facilitates the determination of environmental control strategy and evaluation of environmental quality in piggery. To optimize its prediction accuracy, this model was designed based on least squares support vector regression (LSSVR) with chaotic mutation to improve the estimation of distribution algorithm (CMEDA) for searching of the optimized parameters, which are γ
    Keywords: energy-saving; prediction model; LSSVR; least squares support vector regression; chaotic mutation; EDA; estimation of distribution algorithm; WSN; wireless sensor network.

  • Enhancing Reliability by Adoptive Graph traversals for Backbone assisted Communication in Wireless Sensor Networks   Order a copy of this article
    by Itu Snigdh, Nisha Gupta 
    Abstract: Some applications in wireless sensor networks (WSN) require dedicated backbones for effective and faster delivery of data under the stringent network conditions. An inherent flaw of such networks is that with the failure of a relay node the entire structure is affected and needs restructuring in the worst case failures. In this article we try to enhance the capabilities of such type of networks in terms of reliability by providing spare routes for robust data delivery. Our approach is twofold. Firstly, we estimate the reliability of the network through Markov model to show the improvement and effect on the reliability due to employing ad hoc spare routes in the existing backbone structures. Secondly, we also propose and implement an adoptive relay selection algorithm (ARS) to confirm the improvement in the percentage of data delivery at the sink. Our analysis shows the improvement on the packet delivery ratio and the reliability under different failure conditions.
    Keywords: Wireless sensor network; communication reliability; Markov model; packet delivery ratio; Breadth First Search; Minimum spanning trees.

  • Computational model for the recognition of lower limb movement using wearable gyroscope sensor   Order a copy of this article
    by Tahir Hussain, Hafiz Farhan Maqbool, Nadeem Iqbal, Mukhtaj Khan, Salman ., Abbas Deghani Sanij 
    Abstract: Human activity recognition using inertial sensors, has enabled many applications in different fields, especially healthcare and biomedical engineering. In this regard, an activity recognition system is proposed using the signals of a single gyroscope sensor placed at the shank. Principal component analysis method was utilized to exclude the redundant features from the feature set. Furthermore, different classifiers such as probabilistic neural network, k-nearest neighbour and support vector machine were used for recognition walking activities. K-fold cross validation and four performance parameters namely accuracy, sensitivity, specificity, and Matthews correlation coefficient were used to inspect the performance of the recognition model. The proposed model yielded encouraging recognition accuracy 98.4 % compared to the existing activity recognition systems. It is realized that the proposed system will potentially be utilized in the control of lower limb prosthesis and be useful tool for the gait analysis applications.
    Keywords: principal component analysis; human activity recognition; gyroscope; support vector machine; classification.

  • Controlling Interferences in Smart Building IoT Networks using Machine Learning   Order a copy of this article
    by Per Lynggaard 
    Abstract: The coexistence of many IoT networks in smart buildings poses a major challenge because they interfere mutually. In most settings this results in a greedy approach where each IoT node optimizes its own performance parameters like increasing transmit-power, etc. However, this means that interference levels are increased, battery powers are wasted, and spectrum resources are exhausted in high dense settings. To control interference levels, share spectrum resources, and lower the overall power-consumptions this paper proposes a centralized control scheme which is based on a nonlinear cost function. This cost function is optimized by using machine learning in the form of a binary particle swarm optimization algorithm. It has been found that this approach shares the spectrum in a fair way, it saves power and lowers the interference levels, and it dynamically adapts to network changes.
    Keywords: Smart buildings; IoT networks; interferences; fading; machine learning; BPSO; transmit power regulation; centralized control scheme.

  • Dynamic Load Tuning for Energy-Hole Avoidance in Corona Model for a Wireless Sensor Network   Order a copy of this article
    by Krishna Pal Sharma, T.P. Sharma 
    Abstract: In wireless sensor networks, communication load varies from region to region. Nodes near to sink communicate more than nodes at farther away from sink. This load imbalance causes an energy hole around sink and affects overall network lifetime. Therefore, an approach is proposed to enhance networks lifetime by balancing energy depletion rate across network and avoiding energy hole around sink. The approach is based on corona model for wireless sensor networks to balance energy depletion in inter-corona as well as in intra-corona. Initially, we prove that in a corona model the energy depletion rate across coronas can be sub-balanced by assigning communication ranges in decreasing order from outermost corona towards innermost corona. Thereafter, we prove that if communication ranges are assigned in decreasing order to coronas from outermost towards innermost for a non-uniformly deployed network with increasing density towards sink, balanced energy depletion can be achieved by deploying lesser number of nodes as compared to a non-uniformly deployed network with fixed and uniform communication ranges. Additionally, relay load is tuned over the nodes during data transmission in order to balance intra-corona energy depletion rate of nodes with in a corona. The dynamic relay load tuning is done through a function of relay load and remaining energy of nodes. The proposed scheme can be used for networks with both uniform and non-uniform nodes deployment. In order to validate the results, the approach is compared with baseline approaches through ns-2 based simulation.
    Keywords: Communication range; Energy holes; Network lifetime; Non-uniform deployment; Wireless Sensor Networks.

  • Information and Communication Technologies for the improvement of the Irrigation Scheduling   Order a copy of this article
    by Mohamed Ali Fourati, Walid Chebbi, Mounir Ben Ayed, Anas Kamoun 
    Abstract: Nowadays, agricultural water management is becoming a disturbing issue where scarce resources are increasingly affected by global warming and economic current changes. Indeed, actual irrigation systems are not perfectly adjusted to real contexts or they do not support farmers to carry out all their activities and needs. In order to redress existent deficiencies, information and communication technologies (ICT) are well proposed to minimize losses and improve decisions against unforeseen water shortages. We present in this work a precision irrigation application based on wireless and decision support technologies. Our application provides real-time remote control of decisions and optimizes allocation of exactly required water quantities (beginning and stopping, interruption and recovery times). The objective of such an application is to facilitate the manipulation of the irrigation process and offer flexible opportunities for better watering scheduling and reserves distribution. Therefore, efficient yield will be raised and better profits in time and money will be achieved. We had developed and deployed the system in a real case study where results are evaluated and well judged as regards the objectives.
    Keywords: Precision irrigation; information and communication technologies; wireless sensors network; decision support system; optimization; remote control.

  • Data Retrieval for Deadline-based Multi-request in MIMO Wireless Networks   Order a copy of this article
    by Ping He, Weidong Li, Shufu Cao 
    Abstract: Data retrieval problem is an important issue that generates an access pattern for downloading a request with multiple data items among the parallel channels such that the efficiency of wireless networks is improved. Although many related papers have discussed data retrieval problem that clients require to retrieve one request under the condition of clients equipped with one and multiple antennae, but there are few study on this problem that clients equipped with multiple antennae require to retrieve multiple requests, in particular that each request has time constraint. The problem has great value in the aspects of theory and applications in MIMO wireless networks, such as data sharing and e-business. So, this paper proposes an algorithm that schedules the suitable antennae to find a retrieval sequence (to access data items) about these requests for keeping the balance of access latency among the antennae. For retrieving each request, we develop an efficient scheme that adopts maximum match to generate an access pattern for downloading all requested data items so that the access latency of each request and deadline miss ratio are minimized. Through experiments, the proposed algorithms keep good performance.
    Keywords: Mobile computing; Data broadcast; Indexing; Data scheduling; Datarnretrieval; Data grouping.

  • SIMAS: Smart IoT Model for Acute Stroke Avoidance   Order a copy of this article
    by Samaleswari Prasad Nayak, Surajit Das, Satyananda Champati Rai, Sateesh Kumar Pradhan 
    Abstract: The rate of growth in chronic heart diseases increases the necessity of health care nursing and its constant monitoring. But different constraints have bound us to always avail a physical medical health care for every single person. Technical advancement in medical sciences and Internet of Things provides an opportunity to monitor patients health conditions from anywhere at any time and by anyone efficiently. As a primary diagnostic assessment, correct heart rate and body temperature are the common parameters to be measured. We have developed an IoT based healthcare model SIMAS, which is capable of monitoring the patient heart rate and body temperature locally as well as remotely with almost negligible error rate and includes surrounding factors affecting a patients health condition. Specially by considering the conditions of patients and sports persons who are either in a state of unconsciousness or are unaware of their current health status while in the an intense physical activity the integrated designed healthcare model will work in such a way that the persons required information can be retrieved efficiently and effectively by the authorized users to avoid from critical situations as well as major mishaps.
    Keywords: Heart rate; Humidity; IoT; NodeMCU; Real-Time; Remote patient monitoring.

  • Energy Analysis and Optimization in Amplify-and-Forward Wireless Relay Networks   Order a copy of this article
    by Hai Zhu, Mengmeng Xu, Hengzhou Xu 
    Abstract: In this paper, three kinds of transmission strategies are dealt with: 1) direct transmission without relay; 2) relay transmission without a direct link; and 3) relay transmission with a direct link. We first develop an analytical framework for the total energy consumption per transferred payload bit in amplify-and-forward wireless relay networks. In this framework, analytical expressions of the total energy consumption per transferred payload bit are derived for the three transmission strategies, which are functions of the following network parameters: packet length, transmission energy, coding rate, modulation,rntransmission distance, relay location. Then, the three transmission strategies are optimized to minimize the total energy consumption with respect to the packet length and the transmission energy. By simulation, we compare the three transmission strategies in terms of the total energy consumption per transferred payload bit under different network scenarios. Our results provide several insights into the choice of transmission strategy if some network parameters are given.
    Keywords: Wireless Relay Network; Direct Transmission; Relay Transmission; Energy Analysis and Optimization; Amplify-and-Forward.

  • Low Complexity Versatile Video Coding for Traffic Surveillance System   Order a copy of this article
    by Zhaoqing Pan, He Qin, Xiaokai Yi, Yuhui Zheng, Asifullah Khan 
    Abstract: Due to the diversity of data flows in traffic and mobile data environments, there have been some challenges to both transmission speed and bandwidth. In response to the upcoming challenges of video data compression, the Joint Video Exploration Team (JVET) explored and developed the next-generation video coding standard-Versatile Video Coding (VVC). Compared with HEVC, VVC achieves an improvement of 40\\% compression efficiency, but the coding complexity is very high, and it needs further optimization. In this paper, we propose an efficient Motion Estimation (ME) algorithm for improving the encoding complexity of VVC. First, we propose a content oriented adaptive search range setting algorithm, where the search range size of the children Coding Units (CUs) can be adaptively set by using the best motion vector information of their parent CU. In addition, based on the high spatial correlation and similar characteristics, we design a fast reference frame direction decision method to further reduce the ME encoding complexity. The simulation results show that the encoding complexity saving performance of the proposed algorithm is quite well, i.e. the total encoding time is saved by an average of 34.27%, and the ME encoding time is saved by an average of 40.79%. At the same time, there will be a certain decline in video quality performance which can be ignored.
    Keywords: Search Range; Reference Frame; Motion Estimation; Low Complexity; Versatile Video Coding; Traffic Surveillance System.

  • Performance Analysis of Wireless Sensor Networks and Priority Queueing Systems   Order a copy of this article
    by Jaime Chen, Eduardo Cañete, Manuel Diaz, Bartolomé Rubio, Jose M. Troya 
    Abstract: Wireless sensor and actor networks (WSANs) have been acknowledged as one of the most promising technologies of the 21st century. For example, recently, the use of WSAN technology has been proposed for Critical Infrastructure Protection (CIP). These new applications make more demanding requirements of WSANs and therefore it is necessary to develop algorithms that can manage the communications and route the information from each sensor node to the sink nodes while providing QoS requirements, principally priority and reliability. Many different routing protocols, programming abstractions and middlewares have been proposed in order to provide the QoS, however, WSANs are still far from becoming a mainstream technology. In order to assess the current state of WSANs in terms of performance and QoS support, this article empirically studies the performance of a network with different workloads in terms of reliability, delay and queue occupation with common network topologies. The performance of common queue-based priority mechanisms is also tested. The results reveal the intrinsic lack of reliability and low throughput that this technology currently exhibits, the importance of matching network level capabilities to data link layer capabilities and different behavior of a PQS in terms of reliability and delay depending on the congestion of the network.
    Keywords: Wireless Sensor and Actor Networks; Performance evaluation; Priority queueing system; MAC layer; Network layer.

  • Joint Topology Control and Routing Design for Reconfigurable Ring-Tree Networks   Order a copy of this article
    by Chih-Min Yu, Chun-Chyuan Chen 
    Abstract: This article presents a joint topology control and routing design of Reconfigurable Ring-Tree (RRT) topology for Bluetooth non-uniform networks. The non-uniform network consists of one dense and many other sparse regions. In the dense area, the RRT builds a ring-shaped topology as a backbone subnet in a distributive manner, which is expanded by a tree-shaped topology to other, more sparse areas. For various sizes of networks, the size of the ring subnet is controlled by the trade-off between the network performance and the construction cost. Because corresponding nodes in the ring subnet do not procure the global computation situations, obtaining the optimal ring size is an NP-complete problem. In seeking to finalize the optimum ring size, an empirical max-search strategy is provided to attain the preferred cost-performance ratio. The max-search strategy is a methodical decision policy, carried out by three working elements: the topology construction, the packet routing and the maximum decision elements. The topology construction element engenders the ring-tree topology, the packet routing element processes the routing performance with a uniform traffic model, and the maximum decision element utilizes a decision-making criterion to discover the optimum ring size. Experimental values demonstrate that the optimum ring size can be resolved by the max-search scheme for various sizes of networks, and the RRT delivers a better throughput performance than that of the conventional BlueHRT and Bluetree networks.
    Keywords: Bluetooth; Sensor networks; Topology configuration; Routing scheme.

  • Target Detection and Tracking Based on Information Geometry for Sensor Networks   Order a copy of this article
    by Hao Xu, Huafei Sun, Yongqiang Cheng 
    Abstract: Information geometry is becoming an important research field due to the various applications in statistical inferences, signal processing, neural networks and so on. In this paper, the application of information geometry is explored in anrnattempt to gain a better understanding of sensor system issues for target resolution and tracking in ground-to-air sensor networks. In particular, the Fisher information distance between two targets is used to measure the target resolvability in the region covered by the sensor system and is approximately calculated when it is close enough. And then, the accumulative information is proposed to analyze the resolution of the target on the same detection cellrnwith respect to one measurement model. Furthermore, the single step tracking method is presented based on Fisher information with single bearings-only sensor. The preliminary analysis results presented in this paper provide evidence that information geom- etry is able to offer consistent but more comprehensive means to understand and solve sensor network problems which are difficult to deal with via conventional analysis methods.
    Keywords: Information geometry; information resolution; target detection; accumulative information; target tracking.

  • Design of Human Behavior Recognition Algorithm based on Wearable IMU Sensor   Order a copy of this article
    by Wei Zhuang, Yi Chen, Jian Su, Baowei Wang, Chunming Gao 
    Abstract: In recent years, with the rapid development of Inertial Measurement Unit (IMU) technology, wireless body area network and pattern recognition theory, human motion recognition based on wearable technology has gradually gained the attention of researchers. In this paper, the human behavior recognition method based on wearable sensor motion information fusion is studied. On the existing wearable system platform, the time domain analysis and frequency domain analysis of human motion information are used to distinguish the daily behavior of the human body, and based on the human motion data acquisition experiment, time domain features, frequency domain features and attitude angles of the human motion data are used as identification features. On the basis of it, multi-classification behavior recognition algorithm based on support vector machine is proposed and human motion pattern recognition is carried out. The experimental results show that the system can accurately identify the daily behavior of the human body.
    Keywords: IMU; Wearable technology; Attitude angle estimation; Posture recognition; Support vector machine.

  • Local Outlier Detection Based on Information Entropy Weighting   Order a copy of this article
    by Lina Wang 
    Abstract: As a key research area in data mining technologies, outlier detection can expose data inconsistent with the majority in the dataset and therefore is applicable in extensive areas. The addition of entropy weighting to the spatial local outlier measure (SLOM) and local distance-based outlier factor (LDOF) algorithms in outlier data mining, i.e. the adoption of entropy in the calculation of weighted distance is taken into consideration, leads to enhanced accuracy of outlier detection and produces more expense of time. The algorithm of entropy-weighted LDOF is more optimized than that of entropy-weighted SLOM in terms of detection accuracy. The superiority of the entropy-weighted algorithm is verified through experimental results.
    Keywords: Local outlier; detection; information entropy weighting; SLOM; LDOF.

  • Driving behavior recognition based on orientation and position deviations   Order a copy of this article
    by Wei Sun, Xiaorui Zhang, Xu Zhang, Xiaozheng He, Guoce Zhang 
    Abstract: This paper proposes a driving behavior recognition method, which applies vehicle orientation and position deviations to warn the driver against possible dangers. We integrate a gradient reinforcement method based on the linear discriminate analysis (LDA) to reinforce lane edges. An improved Canny operator based on adaptive threshold segmentation is exploited to extract the lane edges reliably. Based on an improved Hough transform algorithm, the reinforced lane edges help the detection of polar angle and polar radius of lanes that are used to calculate the vanishing point position. After that, the proposed method predicts current-frame lane parameters based on the previous-frame parameters through using the Kalman filter. Combining deviation angle and deviation distance, the proposed method categorizes vehicle lane-keeping behavior into three states: normal, left deviation, and right deviation. Experimental results of a variety of traveling scenes show that the proposed method outperforms other existing methods in precision.
    Keywords: lane detection; vanishing point; Hough transform; Kalman filter; slope angle; offset distance; fatigue driving.

  • Multi-scale Residual Network for Energy Disaggregation   Order a copy of this article
    by Wan'an Liu, Liguo Weng, Min Xia, Yiqing Xu, Ke Wang, Zhuhan Qiao 
    Abstract: Energy disaggregation technology is a key technology to realize real-time non-intrusive load monitoring. Current energy disaggregation methods use the same scale to extract features from the sequence, which makes part of the local features lost, resulting in lower recognition accuracy of electrical appliances with low using frequency. Aiming at the low accuracy of non-intrusive energy disaggregation with low-frequency sampling, a non-intrusive sequential energy disaggregation method based on multi-scale residual network is proposed. Multi-scale residual network extracts multi-scale feature information through multi-scale convolution, and uses residual learning to deepen the network structure to further improve the performance of the algorithm. Through multi-scale convolution, more scale features are captured, and improve the recognition accuracy of electrical appliances with low using frequency. Sequence-to-Sequence energy disaggregation method can improve the disaggregation efficiency and ensure the efficiency of the algorithm. The experimental comparison results show that the model can get a better disaggregation effect, and can effectively identify the start-stop state of electrical appliances.
    Keywords: Energy disaggregation; Deep learning; Residual learning; Multi-scale convolution; Non-intrusive load monitoring.

  • Distributed Spectrum-Sharing in Cognitive Ad Hoc Networks using Evolutionary Game Theory   Order a copy of this article
    by Yifei Wei, Bo Gu, Yali Wang, Mei Song, Xiaojun Wang 
    Abstract: As smart portable devices are becoming increasingly popular and autonomous, decentralized ad hoc networks have been widely applied. Meanwhile, to address the critical problem of spectrum congestion and inefficiency, cognitive radio (CR) are introduced. Cognitive radio ad hoc networks (CRAHNs) consist of a group of cognitive radios nodes connected in an ad hoc manner. Due to its decentralized architecture, unstable network topology, highly fluctuated available spectrum and various transmitting requirements, CRAHNs impose unique difficulties for spectrum sharing among CR users. A challenging and open question is how CR users could share vacant spectrum reasonably in CRAHNs without centralized control. Hence, we formulate the vacant spectrum sharing among CR users in CRAHNs with Evolutionary Game Theory (EGT). In the proposed game, we define the payoff of each CR user as a function consist of the achieved transmit rate and the produced interference to primary users. And we use replicator dynamics to model the strategy adaptation process. Simulation results suggest that the evolutionary equilibrium can be obtained through strategy adaptation and convergence is sensitive to the information latency. The fairness can be guaranteed in the EGT-based spectrum sharing.
    Keywords: Spectrum-Sharing; Cognitive Networks; Evolutionary Game Theory.