International Journal of Wireless and Mobile Computing (35 papers in press)
Expected-mode augmentation method for group targets tracking using the random matrices
by Yun Wang, Guoping Hu, Hao Zhou
Abstract: In order to improve the estimation performance of interactive multiple models (IMM) tracking algorithm for group targets, a new EMA-VSIMM tracking algorithm is proposed in this paper. Firstly, by using the expected-mode augmentation (EMA) method, a more proper expected mode set has been chosen from the basic model set of group targets, which can make the selected tracking models match up to the unknown true mode availably. Secondly, in the filtering process of variable-structure interactive multiple model (VSIMM) approach, the fusion estimation of kinematic state and extension state have been implemented by using classical weighting method and scalar coefficients weighting method, respectively. We use the trace of the corresponding covariance matrix of extension state to calculate the weight coefficient. We calculate the prediction value of the extension state parameter by using a fuzzy reasoning approach to improve the estimation accuracy of the covariance matrix, which takes the elliptical area of extension and its change ratio as the input of the fuzzy controller. The performance of the proposed EMA-VSIMM algorithms is evaluated via simulation of a generic group targets manoeuvring tracking problem.
Keywords: interactive multiple models; expected-mode augmentation; group targets; maneuvering tracking.
Group maintenance strategy for stochastic selective maintenance
by Jianchao Zeng, Jing Zhao
Abstract: Selective maintenance is often applied to many industrial environments in which the maintenance actions are performed between sequence missions. When the length of maintenance or work mission time is stochastic and there are multiple maintenance workers with different capacities, previous works that determine maintenance and mission time are unsuitable. Assuming that the time of work mission is random and maintenance time is determined, to maximise system reliability of the next work mission, this paper proposes a stochastic model for a two-state system comprising several series-parallel components with multiple maintenance workers under the maintenance time limit. The optimal maintenance plan is obtained by hybrid intelligent optimisation algorithm. The validity and feasibility of the model are verified by the simulation.
Keywords: stochastic model; stochastic selective maintenance; mission time; maintenance time; group maintenance.
A hybrid harmony search algorithm for node localisation in wireless sensor networks
by Zhaolu Guo, Shenwen Wang, Baoyong Yin, Songhua Liu, Xiaosheng Liu
Abstract: Harmony search (HS) has been widely used in the field of wireless sensor networks. However, the search strategy of the basic HS has excellent exploration capability but weak exploitation capability. To enhance the search capability of HS, this paper presents a hybrid harmony search algorithm (HBHS) for node localisation in wireless sensor networks. The proposed HBHS employs the best solution to enhance the exploitation capability. Moreover, HBHS uses an adaptive search step-size scheme to further enhance the search capability. To verify the search performance, HBHS is compared with two HS algorithms on a suite of classical benchmark problems. The comparisons confirm that HBHS can achieve better performance than the compared HS algorithms on the most of the benchmark problems. Further, HBHS is applied for node localisation in wireless sensor networks.
Keywords: wireless sensor networks; localisation; harmony search; hybrid strategy.
Global minimisation of fuzzy level set for image segmentation
by Guoqi Liu, Chenjing Li, Ming Deng
Abstract: Level set is an important method in image segmentation, and some models based
on level set method have obtained great success, such as Chan and Vese (C-V) and its convex formulation, local binary fitting (LBF) model. However, these models have two drawbacks to be simultaneously solved. One is the non-convexity of energy function; the other difficulty is segmenting objects in the background of inhomogeneous intensity. In order to simultaneously cope with these shortcomings, a fuzzy level set energy function is proposed. In order to robustly deal with intensity inhomogeneity, a fuzzy factor is introduced in the original LBF model to describe the intensity inhomogeneity. Besides, the edge information is also integrated into the proposed model to improve the robustness of extracting objects. Finally, a regularisation optimisation method is introduced to obtain the global minimisation of the proposed model. Experimental results with quantitative evaluation confirm that the proposed method could segment objects in images with inhomogeneous intensity, and they also show that the proposed method is robust to initialisation because of the convexity of the proposed energy function.
Keywords: image segmentation; fuzzy level set; global minimisation; intensity inhomogeneity.
Investigation framework of web application vulnerabilities, attacks and protection techniques in structured query language injection attacks
by Nabeel Salih Ali
Abstract: Web security has become a great challenge in the recent years. Structured Query Language Injection Attack (SQLIA) is a prevalent and dominant class of the serious web application attacks. A crafter can easily get illegal access to the underlying database in the web application thereby gaining full control of the system and causing millions of dollars loss for corporations. In this paper, provides a comprehensive study of web applications and investigation their vulnerabilities, attacks, and protection techniques against Structured Query Language Injection Attacks (SQLIAs). The study includes presenting a taxonomy of the SQLIAs investigation framework, conducts a detailed review of the various SQLI attacks previous protection techniques. As well, summary and analysis of a critical review (strengths and weaknesses) of the detection and prevention techniques that have been done to address such attacks. Finally, highlights and focus on the critical and important directions or protection approaches that require more studies by future researchers.
Keywords: investigation framework; SQL injection; protection techniques; detection and prevention; web attacks; web applications; web vulnerabilities.
Cellular automata-based model of formation of aerobic granular sludge
by Benzhai Hai, Jie Yang, HaiLei Wang, Zongbo Qiu
Abstract: In this paper, aerobic granules were developed in a sequencing batch reactor (SBR) using synthetic wastewater. A cellular automata (CA) model was established to simulate the formation of an aerobic granular sludge. The results indicate that the model not only visualised the complex formation process of aerobic granules, but also allowed qualitative and quantitative study of the aerobic granules. Thus, the CA model is suitable for simulation of the formation process of aerobic granules cultivated in a granular SBR.
Keywords: cellular automata; aerobic granular sludge; wastewater treatment.
A fault tolerance based route optimisation and data aggregation using artificial intelligence to enhance performance in wireless sensor networks
by Vinod Kumar Menaria, S.C. Jain, A. Nagaraju
Abstract: In the on-demand usage of wireless sensor networks (WSN) over the internet, fault tolerance is an exigent task to improve the overall performance of service computing. In the proposed research work, an attempt has been made to make use of an artificial bee colony approach to find data aggregation for providing fault tolerance in WSN and to make effective use of the existing resources over the internet. In this paper, it is tried to apply quadratic minimum spanning tree (Q-MST), which is an artificial intelligence technique to provide fault tolerance along with data aggregation in WSN. Q-MST is used to improve the fault tolerance in WSN to transmit data packets from the source node to sink node. Ant colony, PRIMS and Particle Swarm Optimisation (PSO) algorithms are used to generate the minimum spanning tree (MST), which can be used for data aggregation. The Q-MST is an improved version of the MST, where ordered pairs of distinct edges would be considered for implementing an alternative edge for the existing edge failure in MST.
Keywords: WSN; data aggregation; fault tolerance; PSO; MST; Q-MST; ABC.
Trajectory planning algorithm and simulation of 6-DOF manipulator
by Jiabing Hu, Ying Sun, Gongfa Li, Guozhang Jiang, Jianyi Kong, Hegen Xiong, Zujia Zheng, Du Jiang
Abstract: In order to solve the problem of joint acceleration mutation in the cubic polynomial trajectory planning algorithm, the algorithm of the quintic polynomial trajectory planning is studied. The results show that the calculation of the quintic polynomial trajectory planning algorithm is relatively heavy, and it can ensure the continuity of angular acceleration and stable operation of motor. Trajectory planning is carried out in Cartesian space by using the spatial line and the spatial arc interpolation algorithm. MATLAB robotics toolbox is used to model the motion system and simulate the motion, which verifies the correctness and feasibility of the linear interpolation and circular interpolation algorithm.
Keywords: manipulator; trajectory planning; Cartesian space; joint space; motion simulation.
Analytical analysis and effect of scrambling on inter-relay interference in a tri-sectored LTE-A network
by Mehboob Ul Amin, Javaid A. Sheikh, Shabir A. Parah, G.M. Bhat
Abstract: With exponential increase of traffic in LTE (Long Term Evolution), LTE operators face the problem of interference. The interference is considered to a threat to the technology of wireless networks and LTE advanced is no exception. Various techniques and methods have been proposed to mitigate the interference in 4G LTE advanced access networks. Mitigation and coverage extension are the major challenges associated with the design of 4G-LTE networks. The incorporation of Relay Nodes (RNs) in LTE networks for coverage and capacity enhancement generates some additional interference, known as inter-relay interference. To mitigate this interference in 4G-LTE-A standard, nodes need to be synchronised. In this paper, a new scrambling technique is used to synchronise the nodes in order to mitigate the effects of inter-relay interference. This makes the proposed technique unique, in the sense that intra-relay distance becomes immaterial, unlike the existing techniques where it is mandatory to maintain a specific distance between the RNs of the same sector. A new analytical model for tri-sectorised hexagonal cellular networks is presented. The expectation values for Signal-to-Interference Noise Ratio (SINR) and throughput capacity for all the positions of access links are derived. Simulations are carried out on Matlab software to validate the analytical analysis. The Cumulative Distribution Function (CDF) curves for both SINR and throughput capacity of each link depict the performance of the proposed model.
Keywords: interference mitigation; relay nodes; fourth generation–long term evolution-advanced; signal-to-interference noise ratio; throughput capacity.
Artificial bee colony algorithm for energy efficiency optimisation in massive MIMO system
by Fatma Bouchibane, Messaoud Bensebti
Abstract: This paper deals with antenna selection for multi-user massive MIMO systems, with the aim of maximising energy efficiency. Massive MIMO technology, by employing a large number of antennas at a base station, provides huge improvements in throughput and energy efficiency. However, the increased number of antennas leads to additional energy consumption due to RF chains and signal processing circuit. The main purpose of the paper is to determine the optimal subset of antennas at the base station that should be activated to serve a given number of active user devices. This idea is implemented using an artificial bee colony algorithm, which has proven its efficiency by specifying the best control parameters.
Keywords: 5G; massive MIMO; energy efficiency; antenna selection; artificial bee colony.
Retaliation based secured model for enhanced weighted clustering algorithm in mobile ad-hoc network
by Naghma Khatoon, Mrs. Amritanjali
Abstract: Mobile Ad-hoc Networks (MANETs) are wireless networks without any fixed infrastructure, consisting of autonomous mobile devices that are interconnected via wireless media. In the recent years, MANETs have become one of the most prevalent areas of research because of their inevitable characteristics. However, resource limitations, energy efficiency, scalability and security are the major challenging concerns in MANETs. In this paper, we propose a novel methodology for secure and fair weighted clustering algorithm using a combination of four parameters; stability factor, degree deviation, sum of distances to a nodes neighbours, and energy depletion. Our proposed algorithm revolves around three important benefits for clustering. The first is to eliminate the non-eligible nodes to become a cluster head at the initial stage of clustering, which reduces the computation and communication overhead. The second is to elect that node to be a cluster head, which retains its neighbourhood for a longer time period, which maintains cluster stability. The third is to retaliate a selfish node in case of any misbehaviour or non-cooperation using the punishment algorithm and enforcing it to behave normally rather than simply blacklisting that node. The simulation results demonstrate the efficacy of the proposed algorithm compared with the other existing algorithms in terms of different performance metrics.
Keywords: ad-hoc networks; clustering; cluster head; retaliation model; stability factor; selfishness factor; punishment factor.
Knowledge network system based on hybrid production process of iron and steel
by Guozhang Jiang, Li Wang, Le Yang, Gongfa Li, Xiaowu Chen
Abstract: Iron and steel production process is a hybrid production process. The production schedule and production scheduling are out of control and disorder owing to the 'asynchronous' connection of the production processes, which is difficult to solve in production management. This paper analyses the production mode of the hybrid flow enterprise, and constructs the integrated control model and the knowledge network system model of the steel production process. By using network design and C# and Matlab mixed programming technology, a prototype of a knowledge network system for the steel production hybrid process based on B/S three tier architecture mode is implemented, and its effectiveness is verified by case simulation.
Keywords: iron and steel production; hybrid process; knowledge network system.
Design of a new compact meandered circular electromagnetic band gap antenna with a shorting pin for wireless communications
by El Amjed Hajlaoui
Abstract: This paper is devoted to the design of a compact meandered Electromagnetic Band Gap (EBG) circular antenna in a multilayered configuration. The patch is short-circuited at its edge with a shorting pin, and several slots are cut in the patch to force the excited patch surface current to travel a much longer path. The introduction of new EBG multilayered substrates above this antenna will improve its electromagnetic characteristics. The designed antenna has the capability to resonate at 4.115 GHz frequency. This new technique will be used to carry two benefits (beamforming, and creating zero radiation beams) and filtering characteristics of the resonator (spatial filtering, and increased directivity, misalignment) owing to the resonant structure itself. The simulation approach is considered to be a new way to avoid cross-coupling elements already present. The analysis provided confirms successfully the various proposed structures and interest occupied by these types of antenna in wireless telecommunications networks. Two approaches, one introduced by a one-layered antenna design with some change in material configuration and the other produced by multilayered structures with different dielectric constants in the EBG resonator, are simultaneously used as key controllers of directivity enhancement. Initially, the paper describes the concept and the realisation of an antenna using an EBG material. The analysis and simulation results are presented for an antenna operating at 4.115 GHz. Next, the defect frequencies of the unit cell of the EBG cover, and those with high directivity for the EBG antenna, are compared to validate the proposed design scheme.
Keywords: electromagnetic waves and propagation; electromagnetic band gap resonator; meandered circular patch antenna; shorting pin; polarisation diversity.
A formal method for software architecture analysis based on aspect orientation
by Xinxiu Wen, Hong Zheng, Zeping Yang
Abstract: Software architecture (SA) analysis plays an important role in the software development lifecycle. However, from the view of separation of concerns, there is an absence of both structural description and behavioral analysis of SA. This paper proposes an aspect-oriented software architecture (AOSA) to reduce the complexity of software models and improve the reliability of software systems. Aspect-oriented architecture description language based on XML is defined for structural description of SA, while aspect-oriented statechart and temporal logic are used for its behavioral analysis. A carousel case illustrates the formal method.
Keywords: software architecture; structural description; behavioral analysis; aspect-oriented statechart.
Design and simulation of MIMO and massive MIMO for 5G mobile communication systems
by Arun Kumar
Abstract: This study concentrates on the investigation of signal detection by using Zero Forcing (ZF), Minimum Mean Square Error (MMSE) and Beam Forming (BF), for a MIMO and massive MIMO system. Outcomes show that ZF is the modest method for signal detection, though BF gives better Bit Error Rate (BER) performance with some constraints in implementation, but MMSE implementation complexity is reduced by avoiding the matrix inversion in the receiver while sustaining the optimal performance as compared with other conventional methods. For several antennas at the base location, it is too difficult to implement the weight matrix for ZF, still, it is appropriate for BF with the benefit of decent digital signal algorithms. The performance of MIMO and massive MIMO using equalisation is investigated to get the throughput of the system (BER vs SNR). Results indicate a significant improvement in BER for 16x16 as compared to 4x4, 8x1 and 8x8 antennas at the base station. The other parameters such as capacity and signal interference noise ratio are likewise examined for massive MIMO system.
Keywords: beam forming; zero forcing; MMSE; MIMO; massive MIMO; 5G.
Hyperheuristic for steelmaking casting rescheduling based on strong disturbance
by Guozhang Jiang, Xiaowu Chen, Bingze Wu, Feng Xiang, Gongfa Li
Abstract: The stochastic disturbance in steelmaking casting often leads to failure of the production plan. So rescheduling is necessary for the manufacturing system. In the background of the increasing scale of scheduling and the numerous kinds of disturbance, the different algorithms are designed for different rescheduling problems, which will greatly increase the workload of the scheduling. In this paper, the disturbances are classified as strong disturbance and weak disturbance, and a rescheduling driven rule of strong disturbance is designed. Especially, the various disturbances that impact on the production plan could be reflected as time changes, so a rescheduling model under time strong disturbance is established. In this paper, a hyperheuristic genetic algorithm (HHGA) is designed based on the heuristic rules and genetic algorithms. The simulation results of the rescheduling model under strong time disturbance indicate the success with the HHGA.
Keywords: steelmaking casting; strong disturbances; rescheduling; hyperheuristic.
Cascading failures to attack on edges in interdependent networks
by Dan Cui, Charles Shen, Feniosky Peña-Mora, Jianguo Chen
Abstract: Networks mostly interact with and depend on each other in modern society, operating and functioning as interdependent networks. In this paper, the cascading process of the interdependent networks against the attack on edges is studied. Then the robustness of the heterogeneous coupled networks and homogeneous coupled networks under random attack and targeted attack is fully explored. It is found that for heterogeneous and homogeneous coupled networks, no particular type always performs better than others. However, in general, homogeneous coupled networks exhibit better robustness than the heterogeneous ones in most cases. It is also found that under targeted attack, no matter what the type of the interdependent network, it gets more vulnerable as the value of the alpha parameter increases, whereas the similar feature cannot be observed in the case of random attack.
Keywords: interdependent networks; edge attack; cascading failures; network robustness.
Safety evaluation for evacuation system under serious emergency from interdependent network perspective
by Huafu Jiang, Youhong Dong
Abstract: In urban safety risk management, many systems can be represented by networks. Major emergency evacuation risk control systems, which are complex systems, are affected by some factors including subjective personnel movement, obstacles, environment and subgroup coordination, which makes the action of the crowd in the system uncertain. It is difficult to examine the risk management related problems under such an uncertain environment. In this paper, the personnel evacuation system is selected as the research objective, and the cascade failure process is modelled and simulated through the complex network theory. Firstly, based on the mutual influence of individual decision-making and overall behaviour in the network, cascading failure processes due to attacks on edges are explored and the S-C-B load redistribution model is innovatively proposed. Secondly, based on the random attack and intentional attack scenarios, we explore the effectiveness of the S-C-B model at improving the system security compared with the original allocation mechanism. Through simulation, we find that owing to the correlation between two subsystems, such as people, objects and management, the tiny destruction in the subsystem will spread rapidly across the network, which will lead to the large-scale destruction of the evacuation network. In addition, for the three subsystems [human, material, management], improvement of any one of these three, can effectively cut off the accident domino and improve the effectiveness of evacuation.
Keywords: interdependent network; serious emergency; risk management; evacuation; safety.
Task selection strategies of self-interested robots in skill games
by Hao Wang, Minglan Fu, Baofu Fang
Abstract: This paper focuses on task allocation for skill games involving a coalition of multiple self-interested robots. When the self-interested robots select tasks, it is difficult for them to maximise both their individual revenues and the system revenue simultaneously. But through reasonable distribution of the tasks utilities, the two matrices can be kept consistent to some extent. Based on this idea, an algorithm is proposed to allocate tasks to self-interested robots that can be in conservative state or radical state. Conservative task selection strategies ensure that the game will converge to Nash equilibrium. Radical task selection strategies help the algorithm to jump out of the Nash equilibrium. Tasks utilities are distributed according to the robots powers in the game. Meanwhile, the algorithm endows each robot with a numerical value, called patience, so that the robots can transform between these two states. Finally, the simulation results verified the effectiveness of the algorithm.
Keywords: multi-robot coordination; self-interested robot; coalition skill games; Nash equilibrium; best response strategy.
Adaptive video contrast enhancement with low noise amplification via local data analysis
by Zhi Dou, Shixun Wang, Hualei Shen, Guoqi Liu
Abstract: Unlike global enhancement methods, the proposed algorithm analyses an image in local areas to take full advantage of the local information, and enhances it in two channels to obtain the exact result. Furthermore, the algorithm can restrain noise amplification by virtue of local statistic characteristics analysis. To enhance videos, the proposed method can make use of temporal information regarding the Kullback-Leibler distance between frames to reduce computational complexity. Experimental results show that the resultant images from the proposed algorithm are comparable or better than those from previous state-of-the-art methods. On the other hand, the computational complexity of the proposed method is much lower than the current local-data-based contrast enhancement algorithms.
Keywords: contrast enhancement; local-data-based; low noise amplification.
Robust pedestrian tracking via multi-cue-based joint particle filter
by Wei Gao, Chenglin Zhang, Yongfei Zhang, Dunbo Cai
Abstract: This paper presents a multi-cue vision system for pedestrian tracking which is designed to be robust in complex environments where severe occlusions exist. The system represents a pedestrian by multiple regions of her/his appearance and the relationships among the regions. The robustness is achieved by both a multi-cue-based joint particle filter and the use of multiple relative regions to represent the nonrigid pedestrian in videos. Specifically, the spatial HSV (Hue, Saturation and Value) histogram and edge features are adaptively integrated into each frame to model the target's appearance. The geometric features of multiple regions of the pedestrians are constructed to simulate their body movements. To co-inference on multiple cues of pedestrians, a joint particle filter on multiple regions is constructed. In the sampling step of joint particles, a prior estimation based Markov chain Monte Carlo method is employed. Results with synthetic data and experiments on real video sequences show that the proposed algorithm is robust in the tested scenarios.
Keywords: computer vision; pedestrian tracking; multi-cue; Markov chain Monte Carlo.
Research on the logistics robot task allocation method based on improved ant colony algorithm
by Xue Fei, Tingting Dong
Abstract: In this paper, multiple logistics robot task allocation in an intelligent warehouse system is studied. First, the task allocation model is developed based on two objectives of time balancing among logistics robots and the correlation between tasks. Then, the model is solved using the improved ant colony algorithm which innovates from the updating rule of pheromone and the setting of the heuristic function. Finally, by simulation experiment, on the one hand, the improved ant colony algorithm can be used to solve the logistics robot task allocation and get a task allocation scheme. On the other hand, the validity of the improved ant colony algorithm can be verified by comparing the improved and original ant colony algorithms.
Keywords: task allocation; logistics robot; intelligent warehouse system; ant colony algorithm.
A improving clustering algorithm for order batching of an e-commerce warehouse system based on logistic robots
by Xue Fei, Tingting Dong, Zixiang Qi
Abstract: In this paper, the batching model and strategy of orders in an e-commerce warehouse system based on logistic robots are studied. First, different order-picking patterns are put forward by analysing the operation process of logistic robots in the e-commerce warehouse system. Then the order-batching model is established based on the two objectives of the minimisation of the total picking and travelling time of logistic robots and the minimisation of the longest picking time used among all the picking stations. The model is solved using the improved clustering algorithm. Finally, the results show that the picking pattern of batching first and combining last has the advantages of higher put-out-storage efficiency by simulating experiment and the comparison analysis of order-picking efficiency corresponding to different order-picking patterns.
Keywords: logistic robots; e-commerce warehouse system; order batching; clustering algorithm.
Optimal decisions of a retailer-owned dual-channel supply chain with demand disruptions under different power structures
by Song Huang, Shuting Chen, Hui Li
Abstract: This paper studies the optimal restoration policy when a retailer-owned dual-channel supply chain experiences unexpected demand disruptions. We study three different supply chain power structures: the manufacturer Stackelberg (MS) game, the retailer Stackelberg (RS) game and the vertical Nash (VN) game. By defining a virtual production cost function, we establish an easy way to restore the supply chains performance. We find that the original wholesale price is robust with demand disruptions if the disruption is mild and the disruption costs are borne by the retailer. In the VN market, the optimal adjusted wholesale price when the disruption costs are borne by the manufacturer is higher than that when the disruption costs are borne by the retailer, if and only if the demand disruption is negative. Moreover, the supply chain power structures have no impact on the robustness of the original production quantity in the decentralised dual-channel supply chain.
Keywords: dual-channel supply chain; disruption management; power structures; game theory.
RJDSS: a storage system for released judicial data
by Yun Teng, Dong Guo
Abstract: A storage system for released judicial data (RJDSS) plays an important role in a judicial publicity visualisation analysis system, which can provide convenience for the government and the general public. Released judicial data presents a number of challenges to the underlying storage system owing to the characteristics of huge amount, wide variety and rapid growth. In order to improve the performance of the upper visualisation system, RJDSS is designed and implemented based on the access characteristics of the released judicial data. Furthermore, it is optimised in system expansion, storage density and read/write performance through such key technologies as scalable distributed storage organisation strategy, highly reliable data dynamic layout mechanism, and exponential-decay-based cache algorithm. The experimental results of effectiveness analysis, data import and video playback show that the RJDSS system achieves performance improvement by 3.5x, compared with the existing judicial data storage systems.
Keywords: released judicial data; storage system; scalability; I/O performance.
Demand forecasting for transportation service network of food cold chain based on a combined model of trend double exponential smoothing and improved grey methods
by Xing Xu, Ren-wang Li, Yun Zhao, Xin-li Wu, Timo Nyberg
Abstract: In a competitive market, the accurate forecasting of short-term transport demand is critical to the transportation service network of a food cold chain. In this paper a model that combines trend double exponential smoothing and improved grey forecasting methods is proposed to predict the short-term cold chain transport demand of transportation service networks of a food cold chain, showing changes in trends and seasonal fluctuations having irregular periods. The combined model is constructed to fit the changing trends and the featured seasonal fluctuation periods. In order to improve forecasting accuracy and model adaptability, the combined model is modelled repeatedly to fit the remnant tail time series of the main combination model until forecast accuracy is achieved. The modelling approach is applied to the freight companies engaged in the transportation of the food cold chain in China. The results demonstrate that the proposed modelling approach produces acceptable forecasting results and goodness of fit, also showing good model adaptability in an uncertain environment. This fact makes the modelling approach an option for predicting the short-term transportation demands of the food cold chain transportation service network.
Keywords: demand forecasting; transportation service network; combination model; time series analysis.
PAPR reduction in OFDM using various coding techniques
by Priyanka Mishra, Mehboob Ul Amin
Abstract: OFDM is a transmission scheme that offers diversity in frequency selective fading environments but suffers from high PAPR. Many researchers have studied the appropriate coding scheme to reduce PAPR and offer good error control properties as well. This article reviews the major results obtained up to date for the reduction of PAPR. Various scrambling techniques, such as PTS and SLM, have been combined with clipping-filtering and DCT to reduce PAPR. The various coding techniques, such as block coding and Reed Muller codes, have been evaluated in this paper to obtain significant reduction in PAPR. CCDF graphs depict the performances of the proposed techniques.
Keywords: OFDM; PTS; SLM; CCDF;DCT.
Optimal cluster head selection by hybridisation of firefly and grey wolf optimisation
by T. Senthil Murugan, Amit Sarkar
Abstract: Clustering is one of the fundamental techniques for prolonging the life expectancy of a WSN. However, cluster head selection still remains the major challenge in WSN with respect to energy stabilisation. This paper proposes the Firefly Cyclic Grey Wolf Optimisation (FCGWO) to simulate the optimal cluster head selection framework. The main objective of this paper is to select the cluster head optimally by focusing on the stabilisation of energy, minimisation of the distance between nodes and minimisation of delay. It hybridises the Firefly (FF) and Grey Wolf Optimisation (GWO) algorithms to attain the best performance. After the simulation, it compares the performance of the FCGWO-based cluster head selection over the traditional algorithms, such as Genetic Algorithm (GA), Group Search Optimisation (GSO), Artificial Bee Colony (ABC), Fractional Artificial Bee Colony (FABC), FF, Firefly with Cyclic Randomisation (FCR) and GWO-based cluster head selection. The performance comparison appears to analyse the network lifetime, energy efficiency and statistics of dead nodes. The simulation outcomes show that the proposed cluster head selection model is more efficient to prolong the lifetime of the network.
Keywords: WSN; clustering; cluster head selection; hybridisation; FCGWO.
Research on parallelisation of collaborative filtering recommendation algorithm based on Spark
by Yang Yongli, Ning Zhenhu, Cai Yongquan, Liang Peng, Liu Haifeng
Abstract: More and more people become conscious of the recommendation system to make good use of the data through their inherent advantages faced with the large amount of data on the internet. The collaborative filtering recommendation algorithm cannot avoid the bottleneck of computing performance problems in the recommendation process. In this paper, we propose a parallel collaborative filtering recommendation algorithm, RLPSO_KM_CF, which is implemented based on Spark. Firstly, the RLPSO (Reverse-learning and local-learning PSO) algorithm is used to find the optimal solution of particle swarm and to output the optimised clustering centre. Then, the RLPSO_KM algorithm is used to cluster the user information. Finally, effective recommendations are made to the target user by combining the traditional user-based collaborative filtering algorithm with the RLPSO_KM clustering algorithm. The experimental results show that the RLPSO_KM_CF algorithm has a significant improvement in the recommendation accuracy and has a higher speedup and stability
Keywords: collaborative filtering recommendation algorithm; RLPSO algorithm; K-means algorithm; Spark.
Energy consumption prediction model of plane grinder processing system based on BP neural network
by Yan Zhou, Hua Zhang, Wei Yan, Feng Ma, Gongfa Li, Wentao Cheng
Abstract: According to the processing characteristics of high energy consumption and low efficiency of China's CNC surface grinding machine, this paper studies the process parameters influence on the energy consumption of the processing system, determines the wheel speed, feed speed of worktable and grinding depth as the main parameters, in which the grinding depth has the greatest influence on energy consumption. Then, the prediction model of the energy consumption of the processing system based on BP neural network is established, and the above three main factors are used as input and the additional load loss power. After training the model, the energy consumption ratio of the grinding machine system can be predicted. The prediction results show that the accuracy of the model is high, and it can predict the energy consumption of the grinder in the process well.
Keywords: processing system energy consumption; process parameters; BP neural network; prediction model.
Optimal base station location in LTE heterogeneous network using non-dominated sorting genetic algorithm II
by Ouamri Mohamed Amine, Zenadji Sylia, Khellaf Sylia, Azni Mohamed
Abstract: The main objective of radio network planning is to provide a cost-effective solution for the radio network in terms of coverage, capacity and quality of service. The network planning process and design criteria vary from region to region depending on the dominant factor, which could be capacity or coverage. However, the optimisation of base stations is an important and crucial process in cellular network planning. It represents a major challenge for mobile operators and is considered an NP-hard problem. In this work, we study the placement of the base station and configuration with an optimisation approach. In addition, a mathematical model based on set covering problem is suggested to solve the base station positioning. The main objectives of model are to maximise the coverage and minimise the financial cost. Non-Dominate Sorting Genetic Algorithm (NSGA II) is applied to find a suitable solution. Simulation results and discussions on the performance of suggested algorithm are provided.
Keywords: base station; cellular network planning; antenna; genetic algorithm;.
A feature selection method based on effective range and SVM-RFE
by Yifei Mao, Yuansheng Yang
Abstract: Identification of discriminative features from information-rich data with the goal of clinical diagnosis is crucial in the field of biomedical science. Support Vector Machine Recursive Feature Elimination (SVM-RFE), an efficient feature selection method, has been widely applied in the domain and has achieved remarkable results. However, biological data are usually class-imbalanced and contain outliers, which largely affect the feature ranking in SVM-RFE. This paper proposes a new feature selection method based on SVM-RFE and Effective Range (SVM-RFE-ER). The proposed method ranks the features by means of combining the SVM weight and the feature weight based on the effective ranges. Experiments on the simulated and real datasets have shown that SVM-RFE-ER is robust, especially against outlier and imbalanced data, and it is effective in identifying biologically meaningful biomarkers for disease study.
Keywords: feature selection; imbalanced data; outlier data; effective range; SVM-RFE.
Performance analysis of efficient power consumption protocols in wireless sensor networks using RELSEP and TSEP
by Ramkrishna Ghosh, Dipak Kumar Jana
Abstract: Wireless sensor networks have attracted worldwide attention. Wireless Sensor Networks (WSNs) and Wireless Multimedia Sensor Networks (WMSNs) consist of wirelessly interconnected sensor nodes organised arbitrarily or deterministically in an environmental area, which can gather, distribute and route information in different application areas. Power consumption in these networks is the foremost problem. WSNs are a recent generation of sensor networks and have an ample range of applications, and their expansion and application will have an extensive impact in human life and construction of all areas. Some of the applications include diaster management, landslide detection, glaciar monitoring, wildlife tracking, health care, military applications, environmental monitoring, security surveillance, industrial process control and a large number of applications to robotics, including Internet of Things (IOTs) projects. This paper will demonstrate the elementary description of WSN followed by different energy-efficient power consumption protocols. Here we have performed the comparative performance analysis of different energy-efficient protocols. In our work we have compared MODLEACH with TSEP followed by TSEP with RELSEP.
Keywords: WSN; LEACH; MODLEACH; SEP; TSEP; cluster head.
Research on conservation planning strategy of historic and cultural site islands in Shanghai based on analytic hierarchy process
by Zhen Wei, Wei Zhang
Abstract: With the continuous development of Shanghai in the process of internationalisation city construction, a variety of urban problems have emerged. The emergence and conservation planning of historic and cultural landscape site islands in Shanghai has become an important research object in academic circles. This paper applies the analytic hierarchy process (AHP) to the conservation planning strategy of historic and cultural site islands in Shanghai. By establishing a protection hierarchy of historic and cultural sites and calculating the weight of elements in each layer, various important factors are ranked. The corresponding protection planning strategy is put forward, which not only provides support for the conservation planning of historic and cultural site islands in Shanghai, but also offers reference for the conservation planning of historic and cultural site islands in other cities.
Keywords: analytic hierarchy process; conservation planning; historic and cultural landscape site islands; Shanghai.
Efficient analytic discrete cosine harmonic wavelet transform OFDM with denoising
by M.N. Suma, S.V. Narasimhan
Abstract: Present wireless implementation needs efficient modulation methods and signal processing at physical layer to enhance system performance. OFDM is one such modulation and multiplexing method preferred for such implementation. OFDM implementation using harmonic wavelets has better performance and implementation simplicity. In this paper, denoising is applied to Analytic Discrete Cosine Harmonic Wavelet Transform OFDM (ADCHWT OFDM), which is implemented in harmonic wavelet domain. ADCHWT is simple and shift-invariant as it does not involve explicit decimation, interpolation and associated filtering and delay compensation. The reduced leakage and computational simplicity offered by ADCHWT provides an interesting opportunity to explore its application for the OFDM system. By application of denoising, BER performance is further improved in shift-invariant ADCHWT OFDM compared with its earlier version DCHWT OFDM
Keywords: OFDM; analytic DCHWT; BER; PAPR.