International Journal of Computer Applications in Technology (86 papers in press)
Non-linear modified equation modeling in dynamical systems (Case Study research on Long Jump patterns)
by Farzad Sharifat
Improving Arabic Text Categorization using FA Words with K-Nearest Neighbor and Centroid-Based classification algorithms
by El-Sayed Atlam, M.E. Abd El-Monsef, O. El-Barbary
3D Scanning Machine and Additive Manufacturing: Concurrent Product and Process Development
by Ismet P. Ilyas
Simulation and visualisation approach for accidents in chemical plants
by Feng Ting-Fan, Tan Jing, Liu Jin, Deng Wensheng
Abstract: A new general approach to lay the foundation for building a more effective and real-time evacuation system for accidents in chemical plants is presented. In this work, we build the mathematical models and realise automatic grid generating based on the physical models stored in advance with several algorithms in jMonkeyEngine environment. Meanwhile, the results of the simulation data through finite difference method (FDM) are visualised coupling with the physical models. Taking fire as an example, including fire with single and multiple ignition sources, shows the feasibility of the presented approach. Furthermore, a coarse alarm and evacuation system from fire have been developed with a multiple SceneNode and roam system, which also includes the making and importing of the physical models. However, to improve the accuracy of the mathematical models, adaptability and refinement of the grids and universality of the evacuation system is the direction of efforts.
Keywords: simulation; chemical accidents; alarm and evacuation system; jMonkeyEngine.
Detecting occluded faces in unconstrained crowd digital pictures
by Chandana Withana, S. Janahiram, Abeer Alsadoon, A.M.S. Rahma
Abstract: Face detection and recognition mechanisms, a concept known as face detection, are widely used in various multimedia and security devices. There are significant numbers of studies into face recognition, particularly for image processing and computer vision. However, there remain significant challenges in existing systems owing to limitations behind algorithms. Viola Jones and Cascade Classifier are considered the best algorithms from existing systems. They can detect faces in an unconstrained crowd scene with half and full face detection methods. However, limitations of these systems are affecting accuracy and processing time. This project proposes a solution called Viola Jones and Cascade (VJaC), based on the study of current systems, features and limitations. This system considered three main factors: processing time, accuracy and training. These factors are tested on different sample images, and compared with current systems.
Keywords: face detection; unconstrained crowd digital pictures; face recognition.
MPF-LEACH: modified probability function for cluster head election in LEACH protocol
by Khalid Nahar, Ra'ed M. Al-Khatib, Malek Barhoush, Alfian Abdul Halin
Abstract: In this research we enhance the LEACH protocol by updating Cluster Head (CH) election probability function (thresholds). More probability was given to out-of-service CH to be elected again. The idea is to get benefit from CH residual energy in order to extend the network lifetime. A new threshold was introduced which guarantees a non-zero probability value of a CH. We propose a newly developed research technique to enhance the original LEACH protocol. The enhancement focuses on extending a WSN's lifetime, and increasing its throughput. It is achieved by giving more probability to re-elect the expired CH that has been removed from CH list because of its insufficient residual energy. Several experiments were conducted to evaluate the efficiency of our proposed MPF-LEACH approach. From the experimental results, a remarkable enhancement in the network lifetime and throughput are achieved. We have improved the election probability threshold for the original LEACH protocol by benefiting from the CH residual energy. As a result, the whole network lifetime was increased due to the extra chance that is given to a CH to be elected again.
Keywords: routing protocol; wireless sensors networks; energy awareness; LEACH protocol; cluster head; threshold election probability.
Online MRAC method using neural networks based on variable learning rate for nonlinear systems
by Sabrine Slama, Ayachi Errachdi, Mohamed Benrejeb
Abstract: This paper presents a new algorithm of a neural network model reference adaptive controller that uses a variable learning rate. It illustrates how the learning rate affects directly training speed. The neural network training algorithms, such as backpropagation, suffer from low convergence speed and are time-consuming. While increasing the learning rate may help to proceed much faster, it can result in unstable training in terms of weights divergence. Therefore, we propose a neural controller training algorithm using a variable learning rate which is capable of speeding up the learning process significantly and it can provide simultaneously stability of the learning process. The results of simulation show that using variable learning rate has better effects both on response time and on tracking performance.
Keywords: model reference adaptive control; neural network controller; adaptation mechanism; variable learning rate; nonlinear systems.
Improved Bayesian regularisation using neural networks based on feature selection for software defect prediction
by R. Jayanthi, M.Lilly Florence
Abstract: Demand for software-based applications has grown drastically in various real-time applications. In order to develop the capital growth of industries, software quality, reliability and customer satisfaction are highly recommended to the software-based industries. However, software testing schemes have been developed that include manual and automatic testing. Manual testing requires more human effort and the chances of error may still affect the quality of the software and hence manual testing cannot be implemented. To overcome this issue, automatic software testing techniques have been developed which are mainly based on machine learning techniques. In this work, we focus on the machine learning scheme for early prediction of software defects using LevenbergMarquardt algorithm (LM), backpropagation (BP) and Bayesian regularisation (BR) techniques. Bayesian regularisation achieves better performance in terms of bug prediction. However, this performance can be enhanced further. Hence, we developed a novel approach for attribute selection to improve the performance of BR classification. An extensive study is carried out with the PROMISE dataset repository using MATLAB simulation tool, where we considered KC1 and JM1 datasets. Experimental study shows that the proposed approach achieves better performance in predicting the defects in software
Keywords: defect prediction model; machine learning techniques; software defect prediction; software metrics.
Integrated safety and economic factors in a sand mine industry: a multivariate algorithm
by Reza Babajani, Mohammad Abbasi, Ahmad Taher Azar, Mahdi Bastan, Reza Yazdanparast, Mahdi Hamid
Abstract: Performance evaluation and optimisation of safety factors is a crucial need for almost all industries. This need is highlighted in some critical industries such as the mining industry. Health and economic losses in mining accidents are significant, especially in developing countries that lack modern technologies. This paper applies resilience engineering as a relatively new approach toward safety analysis in a sand mine in Iran. To this end, a comprehensive framework is proposed based on resilience engineering, fuzzy data envelopment analysis (FDEA) and statistical methods. FDEA is applied to measure efficiency scores of resilience engineering factors. Sensitivity analysis on obtained efficiency scores is employed using statistical tests. The obtained results are validated and verified. The results indicate injury severity score, number of injuries and accident frequency rate as the most effective factors on safety performance.
Keywords: resilience engineering; mining industry; fuzzy data envelopment analysis; safety; performance optimisation.
Congestion control scheme using network coding with local route assistance in mobile adhoc networks
by Navneet Kaur, Rakesh Singhai
Abstract: A mobile adhoc network is a wireless infrastructure-less and self-organised disseminated network. Congestion is a critical issue which occurs in a network when traffic offered to the network exceeds the resource accessibility. This research addresses a novel approach for congestion control for the performance improvement of basic routing protocol AODV. The congestion is discovered based on parameters node queue length, channel utilization and residual energy. To reinforce the congestion control, a network coding mechanism is employed in event of high congestion to reduce the quantity of transmissions among nodes. Local route assistance by neighbouring non-congested node is performed in case of medium congestion to avoid condition of high congestion and to reduce recovery time. The proposed technique shows enhanced performance in performance parameters routing overhead, packet delivery ratio, number of congested nodes, end to end delay, residual energy and normalized routing load. This algorithm is highly adaptive and scalable and mainly reduces packet loss due to congestion and energy consumption of nodes in high mobility cases.
Keywords: MANET; network coding; route recovery; congestion control; congestion status packet CSP; AODV.
Hybrid multiple objective evolutionary algorithms for optimising multi-mode time, cost and risk trade-off problem
by Duc Hoc Tran, Duc Long Luong, Phong Thanh Nguyen
Abstract: Identifying and minimizing the risks associated with time and cost factors in construction projects are the main challenges for all parties involved. As the project duration is shortened, to reduce total cost, the total float is lost resulting in more critical or nearly critical activities. This, in turn, results in reducing the probability of completing the project on time and increases the risk of schedule delays. The objective of project management is to complete the scope of work on time, within budget and deliver a quality product in a safe fashion to maximize overall project success. This research presents a new hybrid multiple objective evolutionary algorithm based on hybridization of artificial bee colony and differential evolution to facilitate time-cost-risk tradeoff problems (MOABCDE-TCR). The proposed algorithm integrates core operations from differential evolution (DE) into the original artificial bee colony (ABC) in order to enhance the exploration and exploitation capacity of the optimization process. A numerical construction project case study demonstrates the ability of MOABCDE-generated, non-dominated solutions to assist project managers to select an appropriate plan to optimize TCR problem, which is an operation that is typically difficult and time-consuming. Comparisons between the MOABCDE and currently widely used multiple objective algorithms verify the efficiency and effectiveness of the developed algorithm.
Keywords: time-cost-risk tradeoff; construction management; multi-objective analysis; artificial bee colony; differential evolution.
EverSSDI: blockchain-based framework for verification, authorisation and recovery of self-sovereign identity using smart contracts
by Tong Zhou, Xiaofeng Li, He Zhao
Abstract: In the current identity management process, the problems such as increasingly differentiated identities, fragmentation and centralization of identity information have been exposed. In this paper, a framework, named EverSSDI, using smart contracts in an Ethereum-based blockchain is constructed, which provides a unique identifier to normalize differentiated user identities. A fine-grained authorization mechanism is designed based on the Hierarchical Deterministic Protocol (HD protocol) to solve the fragmentation problem of identity information. Besides, a reliable information verification mechanism is elaborated to enhance the credibility of the digital information. In addition, two novel methods are proposed for identity recovery, one of which is based on Social Networking Services (SNS) authorization and the other Ethereum Oracles. The final implementation and discussion show that the user has self-sovereign management authority, which enables the user to become the real dominant of his/her identity rather than only a prover of digital identity.
Keywords: digital identity; blockchain; smart contract; ethereum; IPFS; self-sovereign; decentralization; hierarchical deterministic protocol.
A modular cloud-based ontology framework for context-aware EHR services
by Anas AlSobeh, Rafat Hammad, Abdel-Karim Al-Tamimi
Abstract: Healthcare providers in a heterogeneous distributed environment depend on many non-functional requirements of the system which affect the entire Healthcare Information System (HIS). Such requirement specifications are called crosscutting concerns, which lead to changes in many distributed application modules for the healthcare services offered to HIS, such as logging, tracing, security handling, monitoring, and transactions. Implementing the crosscutting concerns potentially can lead to high-impact error-prone integrity of Electronic Health Records (EHR), as well as code quality, i.e. tangling and scattering problems. Aspect-Oriented Programming (AOP) encapsulates the crosscutting concerns to improve the overall quality of software design and implementation by reducing the code that is required to be part of collection modules. In this paper, we propose a novel approach, which uses AOP and service context information to improve the elasticity of the entire cloud-based EHR services. We propose an abstract framework for cloud-based HIS, which provides EHR interoperability through incorporating crosscutting concerns obliviously. This framework eliminates the need to ever touch the core code by easily adding high-level abstractions to complex cloud applications. We implement a prototype of the proposed framework to validate its ability in both separation of concerns and reuse of EHR services.
Keywords: aspect-oriented software development; aspect-oriented programming; modularisation; crosscutting concern; cloud computing; ontology; electronic health record.
An integrated resilience engineering algorithm for performance optimisation of electricity distribution units
by Seyed Hossein Iranmanesh, Mahdokht Tavakoli, Kiomars Heydari, Mahdi Bastan, Reza Yazdanparast
Abstract: This study proposes an integrated fuzzy data envelopment analysis algorithm performance assessment of the resiliency of electricity distribution units in Iran. The proposed algorithm introduce integrated resilience engineering factors including network length, number of employees, transformers capacity, imports, cost and pollutant as input variables, while number of customers, total electricity sales and exports are considered outputs variables. The required data are gathered from Iranian distribution units from 1978 to 2014. The obtained results indicate the most efficient year of electricity distribution in Iran. The results also indicate that the cost of electricity power and imports are most influential factors affecting electricity distribution efficiency. The proposed algorithm can help managers to increase the overall performance of electricity distribution units in strategic level.
Keywords: fuzzy data development analysis; electricity distribution units; performance assessment; efficiency frontier analysis.
Solving a stochastic multi-objective and multi-period hub location problem considering economics aspects by meta-heuristics: Application in public transportation
by Mahdi Hamid, Mahdi Bastan, Mojtaba Hamid, Farrokh Sheikhahmadi
Abstract: The hub location problem in public transportation (HLPPT) is related to a strategic decision which simultaneously determines the location of hub nodes and allocation of demand nodes. Cost and service quality are among the key issues in the current competitive business environment, which have been regarded in HLPPT. Considering the economic aspects as well as the overall processing time in hub nodes can increase the service quality and decrease the total costs. Furthermore, it is useful to consider the possibility of disruptions in hub network to reduce of their disruptive events in hub network. Hence, in this study, a multi-objective scenario based mathematical model is presented for the capacitated hub location problem in public transportation considering economic and investment aspects. Three objective functions are regarded in the presented mathematical model. The first one aims to minimize the total costs considering the possibility of investing the unused budget at each period. The second one aims to minimize the total processing time in the hub network at each period. The last one aims to minimize the maximum distance between each pair of origin-destination nodes in the network. To solve the model, three multi-objective meta-heuristic algorithms are developed, namely S metric selection evolutionary multi-objective optimization algorithm (SMS-EMOA), multi-objective imperialist competitive algorithm (MOICA) and non-dominated sorting genetic algorithm (NSGA-II). Finally, developed algorithms are compared to each other based on several comparison measures using a relatively novel statistical approach. Computational results showed that as far as the mean ideal distance and diversification metrics are concerned, SMS-EMOA outperforms the other algorithms. However, MOICA and NSGA-II can perform better regarding spacing metric and number of non-dominated solutions, respectively.
Keywords: hub location problem; multi-period hub location problem; uncertainty; economic aspects; multi-objective evolutionary algorithms.
Combining structural and semantic cohesion measures to identify extract class refactoring
by Mustafa Hammad, Mohammad Alnabhan, Sarah Al-Sarairah
Abstract: Class cohesion is a major design factor that affects the quality of classes. Classes that have related methods are easy to comprehend and maintain. Classes with many responsibilities are refactored by extracting some methods to new classes. This paper investigates class metrics to identify extract class refactoring opportunities to increase the degree of cohesion. An approach is presented that combines both the structural and the semantic metrics of classes to determine methods that need to be extracted in new classes. A case study is presented to evaluate the proposed approach. The aim of the study is to compare results obtained from applying semantic metrics, structural metrics, and combined metrics together. Results revealed that the proposed approach can provide a valuable set of extract class refactoring suggestions to improve class cohesion.
Keywords: class cohesion; extract class refactoring; SCOM metric; Cosine distance; LOCM metric; Levenshtein distance.
A multi-states continuous time Markov chain model for secondary spectrum access in dynamic spectrum access networks
by Hui Sun, Chuang Yang, Rui Wang, Sabir Ghauri
Abstract: Dynamic Spectrum Access (DSA) networks are vulnerable to hackers who normally pretend themselves to be the primary users; this is called the Primary User Emulation Attack (PUEA). Research communities have already reported a vast use of PUEA in the existing research. Other potential attackers, such as greedy users, should not be ignored when investigating the dynamic spectrum access networks. In this paper, we propose a multi-states Continuous Time Markov Chain (CTMC) model to describe the behaviour of DSA, analysis of the channel states and discussion on the impacts of normal, normal greedy and greedy malicious users in DSA network. The CTMC model is simulated and the simulation results are discussed and validated by comparing with the existing models. Finally, it is proved that CTMC model is an improved method to analyse the performance of the DSA networks when PUEA occurs.
Keywords: dynamic spectrum access; continuous time Markov chain; primary user emulation attack; malicious user; greedy user.
Research on grid-connected photovoltaic inverter based on quasi-Pr controller adjusting by dynamic diagonal recurrent neural network
by Zhenxiong Zhou, Bingshen Liu, Wenbao Wang, Hongxi Wang
Abstract: The single-phase grid-connected photovoltaic inverter system is studied in this paper. In view of the nonlinear and time-varying characteristics of this system, the three-closed-loop control strategy consisting of DC voltage outer loop, grid-connected current inner loop and capacitive current inner loop based on quasi-PR control is proposed. Since the quasi-PR controller of fixed parameters is unable to adapt to changes of parameters in the power network, a quasi-PR control method of dynamic self-tuning based on a dynamic diagonal recurrent neural network (DRNN) is presented. DRNN is based on the recursive prediction error (RPE) algorithm with second-order gradient, which has faster convergence than the BP algorithm. The simulation and experiment results prove that the grid-connected photovoltaic inverter the above control algorithm have a good quality of the output current and fast performance in dynamic response.
Keywords: photovoltaic inverter; Quasi-PR; DRNN; RPE; grid-connected inverter.
A reliable route repairing scheme for internet of vehicles
by Mustafa Banikhalaf, Ahmad M. Manasrah, Ahmed F. AlEroud, Nabhan Hamadneh, Ahmad Qawasmeh, Ahmed Y. Al-Dubai
Abstract: Recently, the Internet of Vehicles (IoV) has been recognised as a key solution for vehicular communications. Connected vehicles, personal smart devices, and infrastructures roadside units have been shaping the underlying architecture of IoVs technology, where the conventional routing protocols cannot facilitate reliable and efficient communication for dynamic IoV topologies. Hence, this technology is highly susceptible to frequent network fragmentations, thus exposing communication channels to regular failure problems. Reliable communication between vehicles requires adopting the existing routing strategies along with the current requirements. This paper, thus, introduces a novel routing repair strategy, referred as Reliable Route Repairing Strategy (RRRS) to tackle routing failure problems. Repairing the operation of channel communications between the source and destination pairs is prioritised according to stability degree of the connected vehicles. The RRRS defines three routing priority zones classified based on the angular values between source, vehicles and destination, and privileges repairing a broken link to the high-active zone only. The RRRS features are combined with the traditional AOMDV protocol, and a comparison study has been conducted to compare the AOMDV, the RRRS-AOMDV and the HM-AOMDV protocols. The simulation results demonstrate that the RRRS-AOMDV achieves better performance, about 30% to 45% in terms of packet transmission overhead, packet repairing overhead and average data packets latency.
Keywords: IoVs; Internet of Things; AOMDV; broken links.
Adaptive neural-fuzzy and backstepping controller for port-Hamiltonian systems
by Ahmad Taher Azar, Fernando E. Serrano, Marco A. Flores, Sundarapandian Vaidyanathan, Quanmin Zhu
Abstract: In this article, a novel control strategy is shown for the stabilisation of dynamic systems in the form of port-Hamiltonian systems. This hybrid approach composed by a neural-fuzzy and backstepping controller is implemented to stabilise the port-Hamiltonian system by dividing it into two blocks in order to separate the variables and yielding an efficient control strategy. Many kinds of dynamical system, such as power, electrical, mechanical and fluid systems, can be represented in port-Hamiltonian form. Thus, it is important to develop new control strategies to stabilise port-Hamiltonian systems considering that this is not a simple task, especially to increase the robustness, to deal with the uncertainties and to improve the system performance. The proposed control strategy consists in an hybrid approach formed by a neural-fuzzy and backstepping controller. A four-layer neural-fuzzy controller is implemented to stabilise the port-Hamiltonian system, where fuzzification, fuzzy rules inference system and defuzzification layers are considered. The neural-fuzzy controller consists of two steps: an offline training implementing a gradient descent algorithm and an online training by a Lyapunov stability approach. The backstepping controller is designed by a recursive method considering the port-Hamiltonian system properties and implementing a Lyapunov stability approach. Along with the proposed control strategy, a neural-fuzzy observer is implemented to estimate the port-Hamiltonian system states considering the properties of the system representation. Finally, a cart-pendulum example is shown to verify the effectiveness of the proposed observer and controller along with a comparative analysis
Keywords: neural-fuzzy system; backstepping control; observer design; port-Hamiltonian system.
A simulation model of pedestrian wayfinding behaviour in familiar environments
by Amina Bouguetitiche, Foudil Cherif, Fabrice Lamarche
Abstract: In this paper, we present a novel approach to simulate wayfinding behaviour of pedestrians familiar with their environment. This approach is inspired from spatial cognition and space syntax domains in order to achieve naturally crowd navigation. Therefore, the proposed wayfinding process is incremental; route choice decisions are made at every street junction, taking into account spatial configuration and individual knowledge of the environment as well as individual preferences. An adequate environment description is provided; it is a graph automatically generated, informed with pre-calculated data, that is used by agents to quantify the benefit cost of a route choice. The environment description is also used to endow agents with mental maps that contain the regions supposed to be experienced by them without going through a learning phase. Obtained results demonstrate that, under our model, agents calculate paths that have the same characteristics as those chosen by pedestrians familiar with their surroundings.
Keywords: environment description; wayfinding; mental map; space syntax; cognitive science; urban environment.
Real-time high speed 5-D hyperchaotic Lorenz system on FPGA
by Ismail Koyuncu, Murat Alcin, Murat Tuna, Ihsan Pehlivan, Metin Varan, Sundarapandian Vaidyanathan
Abstract: Chaotic systems have several engineering applications, such as cryptology, random number generators, image processing and secure communication. A basic structure used in these studies is a chaotic oscillator design that produces a chaotic signal. In this paper, the 5-D hyperchaotic Lorenz system (Hu, 2009) has been implemented on FPGA using Heun algorithm to improve the chaos-based embedded engineering applications. The 32-bit IEEE-754-1985 floating point format has been used in the Heun-based design. The design has been coded in VHDL. The maximum operating frequency of FPGA-based 5-D hyperchaotic Lorenz system reaches 430.146 MHz. In addition, a real circuit realisation of the 5-D hyperchaotic Lorenz system has been performed using analogue circuit elements. The results of the new FPGA-based 5-D hyperchaotic Lorenz system have been compared with the results of computer-based numerical simulation and then the error analyses (MSE and RMSE) have been carried out.
Keywords: hyperchaos; FPGA; VHDL; Heun algorithm.
DANP-based method for determining the adoption of hospital information system
by Khuram Shahzad, Zeng Jianqiu, Asma Zubedi, Xin Wen, Lei Wang, Muhammad Hashim
Abstract: Hospital information system (HIS) is an integrated electronic system that provides comprehensive information regarding every aspect of the hospital and patients whenever it is required. In Pakistan, the diffusion of HIS is in the early stages and the rate of adoption is very slow. The primary purpose of this study is to identify the essential factors that are significantly driving or hindering the decision to adopt HIS. For better understanding, this study proposed the initial theoretical model that integrates Technology Organization Environment (TOE) framework, Human Organization Technology (HOT) fit model and institutional theory. Hence, the initial model consists of four main dimensions and 13 variables, which are the most frequently used in prior literature and are essential for the investigation of HIS adoption. The data were collected from healthcare experts who have full knowledge of HIS. Accordingly, the recently developed DANP (Decision Making Trial and Evaluation Laboratory (DEMATEL) based Analytic Network Processes (ANP)) method is employed for assessing interdependency and give weights to dimensions and criteria. According to the experts' knowledge and experience, the results indicate that perceived technical competence, compatibility, top management support, and vendor support are found to be the most essential variables for the successful adoption of HIS concerning people, technology organisation and environment, respectively. Hence, the finding of this study has contributed theoretically, and the practical implementation of this integrated model will give deep insight to healthcare providers for the successful implementation of HIS.
Keywords: hospital information system; public hospitals; TOE framework; HOT-fit model; institutional theory; DANP method; public hospitals; Pakistan.
Experimental and modelling study of flow characteristics on large-scale roughness bed
by Zhang Jianmei, Han Zhengguo, Che Quan, Zhu Feng
Abstract: Large-scale roughness bed flow is a special flow condition that is commonly seen in mountainous rivers. In this paper, a gradient flume experiment is carried out with plum-blossom distributed cubic obstruction blocks with 50 mm edge, and large-scale roughness bed flow characteristics including free surface morphology, flow velocity distribution, turbulence propagation and friction head loss coefficient are studied by high precision measuring and statistical analysis. In addition, a CFD-based numerical model is established to verify the experimental results and provide a prediction method to expand the experiment. After that, a modified empirical formula of friction head loss coefficient considering more important influence factors is proposed, based on the experimental results and the numerical simulation results, which provide a practical calculation method of friction head loss coefficient for engineering reference.
Keywords: large-scale roughness; plum-blossom distribution; flow characteristics; numerical simulation; empirical formula.
VoiCon: A Matlab GUI-based tool for voice conversion applications
by Sanghamitra Nath, Nabadip Borah, Aparajita Gohain, Utpal Sharma
Abstract: Voice conversion finds applications in a wide variety of areas, such as customisation of text to speech systems, voice editing and dubbing, voice restoration systems, in addition to its initial applications of speaker conversion and conversion of speaking styles. The basic steps required for the implementation of voice conversion however remain the same. The alignment of features using various techniques, such as dynamic time warping and time sequence matching, the development of the mapping function, and finally the conversion of features, are computation intensive and require the researcher to have in-depth knowledge of the various techniques used. In this work, in an effort to reduce the tasks of a researcher interested in using voice conversion for his application, a Matlab GUI-based tool has been designed, implemented and tested for carrying out spectral feature conversion for three applications, conversion of female to male speech, whispered to normal speech and synthetic to natural speech. The tool not only provides an easy to use interface by carrying out feature conversions using various mapping functions but also enables the user to view, save and compare his results graphically.
Keywords: voice conversion; mel cepstral coefficients; fundamental frequency; Gaussian mixture model; vector quantisation; artificial neural networks; root mean square error; mel cepstral distortion.
Hand-drawn electronic component recognition using deep learning algorithm
by Haiyan Wang, Tianhong Pan, Mian Khuram Ahsan
Abstract: Hand-drawn circuit recognition plays an increasingly important role in circuit design work and electrical knowledge teaching. Hand-drawn electronic component recognition is an indispensable part of hand-drawn circuit recognition. Accurate electronic component recognition ensures accurate circuit recognition. In this paper, a hand-drawn electronic component recognition method using a convolutional neural network (CNN) and a softmax classifier is proposed. The CNN is composed of a convolutional layer, an activation layer and an average-pooling layer and is designed to extract features of a hand-drawn electronic component image. The kernel function for the CNN is obtained by a sparse autoencoder method. A softmax classifier is trained for classification based on the features extracted by the CNN. The recognition method can identify rotating electronic components because of the added rotated image and achieve 95% recognition accuracy.
Keywords: electronic component recognition; convolutional neural network; sparse autoencoder.
FCM: a component-based platform with explicit support of crosscutting and dynamic features
by Abdelhakim Hannousse
Abstract: Dealing with crosscutting and dynamic features in component software is a longstanding problem, primarily owing to the nature of used components: components may be available only as black box software units and their implementations may be protected against alteration. Aspect-orientation provides a valuable means to deal with crosscutting features in different paradigms; however, existing endeavours to use aspects in component software have several limitations, such as the lack of suitable design of aspects and the absence of proper aspect runtime weaving mechanisms. In this paper, we contribute by proposing a new aspect component model to solve such problems. In the proposed model, components and aspects are first-class entities that remain separated from design to implementation; and aspects can be added and removed at runtime. We also developed a tool support for the model in Java. We demonstrate the viability of the model through the implementation of a running example.
Keywords: component-based software engineering; aspect-oriented programming; crosscutting and dynamic features.
Correlation-based Search for Time Series Data
by Ibrahim A. Ibrahim, Abdullah Albarrak
Abstract: Exploration of time series data based on correlation is a key ingredient ofrnvarious analysis tasks. However, such exploration entails massive CPU and I/O costsrndue to the quadratic nature of the exploration space. Searching for a time sub-intervalrnin which all time series pairs are correlated within certain values is one aspect of timernseries exploration and has various applications in many domains. Consequently, in thisrnpaper, we formulate the Targeted Correlation Matrix Search problem where the goal is tornnd an optimal sub-interval with a correlation matrix that maximizes the closeness andrnsimilarity to targeted pairwise correlation values. We show the computational hardness ofrnthis problem, and propose the RELATE scheme to address the associated challenges byrnutilizing the incremental property of correlation. Further, we propose two-level pruningrntechniques for the RELATE scheme to minimize the associated computational and I/Orncosts. These techniques enable RELATE to avoid exhaustively traversing the search spacernby pruning unqualied candidate queries, and avoid computing pairwise correlation ofrnevery time series pair wherever possible.We demonstrate by experiments the performancerngains of RELATE against state-of-the-art algorithm with real and synthetic datasets.
Keywords: correlation; time series; search
Comparison among different tools for tolerance analysis of rigid assemblies
by Wilma Polini, Andrea Corrado
Abstract: Tolerance analysis is an important task to design and to manufacture high precision mechanical assemblies; it has received considerable attention in the literature. Actually, there are some different tools used or proposed in the literature to make the tolerance analysis of an assembly, but none of them is completely and univocally accepted. A comparison between a Computer Aided Tolerancing (CAT) tool for geometry assurance and some methods proposed in the literature is discussed in this work. Therefore, the aim of this work is to solve, by a CAT software, five case studies that were already solved by different methods in the literature. The potentialities and the limits in using a CAT technique for geometry assurance are highlighted.
Keywords: geometry assurance; tolerance analysis; computer aided tolerancing.
Cloud-based electricity consumption analysis using neural network
by Nand Kumar, Vilas Gaidhane, Ravi Kant Mittal
Abstract: In recent years, optimisation of resource usage is very much required to analyse and understand energy consumption patterns. This analysis has previously been carried out using algorithms, which needs many assumptions, and meeting all the assumptions in practice is a very difficult task. However, there are other methods available to analyse and understand energy consumption. In this paper, an efficient approach for energy consumption pattern analysis is proposed. It is based on the Levenberg-Marquardt algorithm-based neural network (LMNN) and clustering technique. The energy consumption data is collected from the educational institute building using a smart system. The various experimentations are carried out on the collected real time database. The experimental results illustrate that the proposed approach is effective and computationally efficient for consumption pattern classification. The performance of the presented approach is found superior to existing clustering approaches.
Keywords: educational institute building; Levenberg-Marquardt algorithm; neural network; classification; confusion matrix; ROC curve.
A novel ANN-based four-dimensional two-disk hyperchaotic dynamical system, bifurcation analysis, circuit realisation and FPGA-based TRNG implementation
by Sundarapandian Vaidyanathan, Ihsan Pehlivan, Leutcho Gervais Dolvis, Kengne Jacques, Murat Alcin, Murat Tuna, Ismail Koyuncu
Abstract: This paper describes the modelling, bifurcation analysis, circuit realisation and FPGA implementation of a novel ANN-based four-dimensional two-disk dynamical system exhibiting hyperchaos and hidden attractor. This paper provides a detailed analysis of the multistability, coexisting attractors and bifurcation properties of the novel system. The system does not possess any rest point pinpointing the presence of hidden hyperchaotic attractor. We realise the dynamic equations of the two-disk hyperchaotic dynamical system with a real circuit. Next, we build, design and implement the ANN-based two-disk hyperchaotic dynamical system on FPGA. Finally, using the FPGA-based implementation, we design and implement a novel high speed True Random Number Generator (TRNG).
Keywords: circuit design; hyperchaos; hyperchaotic systems; artificial neural network; FPGA implementation; TRNG.
Prediction modelling of exhaust characteristics of a marine engine for SCR urea dosing calibration
by Zhuo Zhang, Mingwei Shi, Zibin Yin, Defeng Wu, Leyang Dai
Abstract: The International Maritime Organization (IMO) issued Annex VI of the MARPOL Convention to control the serious exhaust pollution of marine diesel, specifying the NOx emission limitation requirements. In this paper, the exhaust characteristics of a marine diesel engine were tested, and the exhaust characteristic model was established by a BP neural network, which has been verified via learning ability and generalization ability. The relative errors of the exhaust flow, NOx concentration and exhaust temperature prediction are within 6%, which can be used to predict the exhaust performance of a marine diesel engine in steady state. The calibration for urea dosing of an SCR system was based on an ammonia-nitrogen ratio of 1:1, whose data are predicted by the exhaust characteristic model.
Keywords: Marine engine, Exhaust characteristics, BP neural network, Modelling, SCR system
A Comparative Study of meta-heuristic Optimization Techniques for Prioritization of Risks in Agile Software Development
by Prakash B, Viswanathan V
Abstract: Risks are in general termed as threats or uncertainties that influence the project performance and its outcomes to the greater extent. To ensure software quality and project success, every organization should enforce a proper mechanism to efficiently manage the risks irrespective of the development model they follow. Risk prioritization is a most critical step in risk management process that helps the organization to resolve the risks in shorter duration of time. In this paper, a comparative study about different meta-heuristic optimization techniques for prioritizing the risks in agile environments is presented. The five most effective meta-heuristic optimization algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Grey Wolf Optimization (GWO) and Analytical Hierarchy Process (AHP) are considered and the results are evaluated based on four key criterion attributes such as error rate, accuracy, reliability, and running time. The result proves that GWO outperforms other four meta-heuristic optimization techniques for the prioritization of risks in agile environment.
Keywords: Risk Management; Risk Prioritization; Agile Software Development; Meta-Heuristic Optimization; Project Management.
Reasoned bargaining protocol in construction contracts using a novel Bayesian game
by Vu Hong Son Pham
Abstract: The objective of this paper is to provide new insights on some dimensions of the bargaining process – asymmetries and uncertainties in particular- by using a Novel Expert Bayesian Game (NEBG). We develop decision support system for determining the price in construction supply contract. Our results confirm that uncertainty affects negotiators’ behavior and modify the likelihood of a self-enforcing agreement to emerge. The validation analysis revealed that a novel approach to BN construction by combining domain knowledge from experts possessing incomplete observation data substantially improved the estimation ability of negotiators. Exhibiting a high success rate and profit as well as low negotiation time, the proposed model is superior to those reported in previous research.
Keywords: decision support system, expert system, Bayesian game.
Manifold multi-view learning for cartoon alignment
by Wei Li, Huosheng Hu, Chao Tang, Yuping Song
Abstract: Cartoon alignment is a key to retrieve cartoon characters and synthesize new cartoon clips. To successfully achieve the tasks, it is necessary to extract visual features that comprehensively denote cartoon characters and to align the feature points accurately between cartoon characters. In this paper, Speed Up Robust Feature (SURF) and Shape Context (SC) are introduced to characterize the cartoon character from multi-view. The two features are complementary to each other, and each feature set is thought as a single view. To increase accuracy rate of cartoon character alignment, traditional methods, such as semi-supervised alignment and Procrustes alignment, require predetermining the correspondence. However, this is a tedious task. To overcome the flaw, we propose a manifold multi-view learning (MML) to align cartoon characters. MML learns a projection that maps data instance (from cartoon characters with different dimensionality) to a lower dimensional space, which simultaneously matches the local geometry and preserves the neighbourhood relationship within each cartoon character. The matching relationship can be obtained from local geometry structure. Experimental results show the good performance, such as a matching accuracy rate of more than 90% and processing time of average 100 milliseconds that is only 30% of the traditional algorithm in the certain dataset. Hence, MML can also be potentially applied in mobile devices.
Keywords: Cartoon Alignment; Manifold; Multi-view; Speed Up Robust Feature; Shape Context
Special Issue on: BDCA'17 Computer Science and Information Technology
Dynamic VM allocation and traffic control to manage QoS and energy consumption in cloud computing environment
by Mohamed Hanini, Said El Kafhali, Khaled Salah
Abstract: In the last few years, cloud computing technology has revolutionised the IT industry and its popularity has increased, owing to its economic benefits for both the cloud providers and the users. Despite the benefits that this new paradigm offers, it poses major challenges for providers. Among these challenges is the guarantee of the desired Quality of Service (QoS) for the users defined in the Service Level Agreement (SLA) document. Moreover, power consumption control can significantly benefit providers. In this paper, we propose a mechanism combining a scheme for Virtual Machine (VM) use in a given Physical Machine (PM) with a mechanism to control the access for incoming requests to the Virtual Machine Monitor (VMM). The number of activated VMs in the PM is defined according to the workload, and the control access is based on the number of requests in the system. The studied mechanism is described by a mathematical model, and the performance parameter expressions are derived. In addition, a power consumption model is described and evaluated. Numerical examples evaluating these parameters are given. In particular, in terms of QoS, we analyse the behaviour of loss probability, mean number of requests, throughput and mean requests delay while varying the incoming request arrival rate. Moreover, the impact of the proposed mechanism on the behaviour of energy consumption is evaluated. Analysis of the obtained results shows the positive impact of the proposed mechanism on the QoS parameters and on power consumption.
Keywords: queueing theory; cloud data centre; virtual machines; performance analysis; quality of service; energy consumption.
A comparison of text classification methods using different stemming techniques
by Mariem Bounabi, Karim El Moutaouakil, Khalid Satori
Abstract: In the retrieval of information, two factors have an important impact on the system's performance: the extract features and the matching process. In this work, we compare three well-known stemming techniques: Lovins stemmer, iterated Lovins, and snowball stemmer. Concerning the classification phase, we compare, experimentally, six methods: BNET, NBMU, CNB, RF, SLogicF, and SVM. Basing on this comparison, we propose a new retrieval system by calling the voting method, as a matching tool, to improve the performance of the classical systems. In this paper, we use the TF-IDF algorithm to extract features. The envisaged systems are testing on two databases: BBC NEWS and BBC SPORT. The systems based on Lovins stemmer and on the voting technique give the best results. In fact, for the first databases, the best accuracy observed is for the system Lovins +Vote with a recognition rate of 97%. Concerning the second database, the system snowball +Vote gives 99% as the recognition rate.
Keywords: NBMU; SVM; RF; NB; SLogiF; CNB; voting technique; classification; stemmer; weighting term.
Performance prediction of pharmaceutical suppliers: a comparative study between DEA-ANFIS-PSO and DEA-ANFIS-GA
by Rohaifa Khaldi, Abdellatif El Afia, Raddouane Chiheb
Abstract: Selection of pharmaceutical suppliers is a critical task within a hospital, because dealing with the wrong supplier may plague the overall healthcare supply chain, and possibly risk patients' lives. Thereby, this study investigates the feasibility of using DEA in conjunction with ANFIS-PSO and ANFIS-GA, to evaluate and predict supplier performance. This investigation is a comparative study between ANFIS-PSO and ANFIS-GA. To our best knowledge, it fills the gap in the literature by assessing the benchmarking capabilities of the two proposed models. DEA-BCC was applied to evaluate the efficiency scores of the selected suppliers. ANFIS-PSO and ANFIS-GA were applied to learn DEA patterns and to predict the performance of unseen suppliers. To determine the accuracy of those models, a statistical analysis was performed. Besides, the results were compared with ANFIS-Hybrid model. According to RMSE and R, the results revealed that ANFIS-PSO model yields the best prediction abilities. Thus, this model can be considered as a promising decision support system at the operational and strategic levels.
Keywords: adaptive neuro-fuzzy inference system; genetic algorithm; particle swarm optimisation; data envelopment analysis; benchmarking; prediction; performance; pharmaceutical suppliers; healthcare supply chain.
Estimate of stochastic model parameter of exchange rate using
machine learning techniques
by Mostafa El Hachloufi, Hamza Faris, Mohammed El Haddad
Abstract: In this paper we present a new approach for estimating the stochastic model parameter of exchange rate using genetic algorithms and neural networks.This approach takes into consideration the minimisation of exchange rate risk that is measured by the conditional value at risk (CVaR) in the estimation procedure of this parameter. The objective of this approach is to provide a tool of decision for the exchange market managers.
Keywords: exchange rates; estimation; risk; CVaR; stochastic model; genetic algorithms; neural networks.
Using data taxonomy to achieve security in cloud storage
by Ennajjar Ibtissam, Tabii Youness, Benkaddour Abdelhamid
Abstract: Cloud computing is a broad concept pertaining to different service models following the utility computing model. Owing to its numerous advantages, such as high resource elasticity, time improvement, IT maintenance cost reduction, and simplicity, the cloud computing paradigm interests many customers. However, notions introduced by the cloud, such as the multi-tenancy concept, computation outsourcing, and distributed resources, increase the security concerns and make trust in cloud providers a critical security challenge. This paper suggests a new method for enhancing the confidentiality of data in the cloud. As data in the cloud has not the same sensitivity, encrypting it with the same algorithms can lead to a lack of security or of resources. The paper proposes to classify data according to a sensitivity level in order to give a suitable security model for each data. By this process, we optimise the use of security mechanisms and the resources consumption.
Keywords: cloud computing; data security; classification; cloud storage; sensitivity level; confidentiality.
Special Issue on: ICMIC2016 Computer Applications in Technology
Sensorless high-order super-twisting sliding modes vector control for induction motor drive with adaptive speed observer.
by Horch Mohamed, Boumédiène Abdelmadjid, Baghli Lotfi
Abstract: This paper proposes a high-order super-twisting sliding mode control method applied to an induction motor fed by a power voltage source without speed sensor. Based on the vector control principle, high-order super-twisting sliding mode controllers in speed loop and flux loop are designed, respectively. The super-twisting sliding mode control is utilized to improve the response speed and robustness of motor control systems. Meanwhile, high-order sliding modes are adopted to eliminate the chattering phenomenon. We also present the mechanism of adaptive super-twisting speed and rotor flux observers with the only assumption that from stator voltages and currents are measurable. The objective is to improve the speed control, the rotor flux control under load torque disturbances and parameter variations.The simulation results prove clearly a good robustness against load torque disturbances, the estimated fluxes and the rotor speed converge to their real values. Our study is close to reality; all carried out simulations are based on real models simulated within the Matlab Sympower system environment in continuous time.
Keywords: non-linear systems; field-oriented control; induction motor drive; robustness; super-twisting sliding mode; sensorless drive.
Novel adaptive iterative observer based on integral backstepping control of a wearable robotic exoskeleton
by Brahim Brahmi, Maarouf Saad, Cristobal Ochoa-Luna, Mohammad Habibur Rahman, Abdelkrim Brahmi
Abstract: In this paper, an integral backstepping control combined with an iterative estimator and a Jacobian observer of external forces is used to take into account the dynamics' uncertainties of an exoskeleton robot arm. The exoskeleton robot carries the upper limb of the subject to perform a passive physical therapy. The users force is thus considered as an external disturbance. Additionally, an accurate modelization of the dynamic model of the 7-DOFs exoskeleton robot is not available owing to its complicated mechanical structure. In such a case, the system may be subject to modelling uncertainty. In order to reduce the uncertainties and external disturbances effects, a robust integral backstepping control and a Jacobian force observer are used. A Lyapunov function is selected to prove the closed loop stability of the system. Experimental results show the effectiveness and feasibility of the designed controller to the uncertain robot system.
Keywords: backstepping control; iterative control; force observer; physical therapy; uncertainty.
Modelling, simulation and control of a class of hybrid dynamic systems using hybrid automaton, APROS and mixed integer quadratic optimisation algorithm
by Mohamed Fouzi Belazreg, Khaled Halbaoui, Djamel Boukhetala, Mohamed El-Hadi Boulheouchat
Abstract: This paper presents modelling, simulation and control of a class of hybrid dynamic systems. The hybrid automaton is used for modelling a transition system with continuous dynamics. The framework consists of a finite set of state and transition for modelling a discrete dynamics who called control mode. Each control mode describes continuous dynamics. Using APROS tools, it permits to simulate the behaviour of the hybrid system approaching in the experiment case: actuators, pumps and valves. The mixed logic dynamics formalism allows to describe the both dynamics defined by logic rules, continuous dynamics and constraints. These are described by linear dynamic equations subject to linear inequalities involving continuous, discrete and auxiliary variables. This model is used to synthesise a predictive control law under constraints. The controller requires online mixed-integer quadratic programming solution to an optimization problem. Simulation was performed to illustrate performances and efficiently of these methods and tools.
Keywords: hybrid systems; hybrid automaton; non-linear systems; nodalisation APROS; mixed logical and dynamical; stateflow; model predictive control; mixed-integer quadratic programming; OPC interface.
Hybrid chaotic synchronisation between identical and non-identical fractional-order systems
by Abir Lassoued, Olfa Boubaker
Abstract: In this paper, a chaotic hybrid synchronization (HS) for multiple fractional-
order (FO) systems coupled with ring connection is proposed. For such schema, the
complete synchronization (CS) and the complete anti-synchronization (AS) should coexist in the same time under designed control laws. It will be ascertain, that integer-order controllers are adept to synchronize asymptotically coupled FO systems. In order to prove the effectiveness of the synchronization schema, two cases studies are considered where multiple FO identical systems and multiple non-identical FO systems are synchronised, respectively. Finally, numerical simulations illustrate the good achievement of the synchronisation problem under the designed control laws.
Keywords: fractional order; hybrid synchronisation; ring connection; anti-synchronisation; complete synchronisation.
Chaos Synchronisation of two different PMSMs via a fractional-order sliding mode controller
by Amina Boubellouta, Abdesselem Boulkroune
Abstract: This paper deals with the design problem of a fractional-order sliding mode control to synchronise two different chaotic permanent magnet synchronous motors (PMSM). By constructing fractional-order sliding mode surfaces, it is proved that the corresponding synchronisation errors are Mittag-Leffler stable. The simulation results show that the proposed controller is strongly robust against the parametric variations, modelling uncertainties and unknown external disturbances, and can significantly reduce the chattering level.
Keywords: fractional-order sliding surface; sliding mode control; chaotic PMSM; chaos synchronisation.
Intelligent power system controller design
by Saoudi Kamel, Bouchama Ziyad, Ayad Mouloud, Benziane Mourad, Harmas Mohamed Naguib
Abstract: In this paper, a type-2 fuzzy-based adaptive sliding mode power system controller is proposed for damping low frequency oscillations with the aim to enhance power system stability despite modele uncertainties introduced by variations of system parameters and external disturbances. Addressing these latter, Type-2 fuzzy systems approximating properties are used to approximate unknown power system nonlinear dynamics. Furthermore, to achieve more robustness, the proposed controller design is combined with sliding mode approach. The latter and Lyapunov synthesis approach are incorporated in an adaptive fuzzy control scheme such that the derived controller is robust, closely tracking any changes in power system operating conditions and guaranteeing stability while a PI control term is added to mitigate chattering. Proposed stabilizer robustness has been tested on a single machine infinite bus system and a multi-machine power system. Nonlinear simulation studies show good performance of the proposed stabilizer and confirm its superiority over conventional PSS and some other types of power stabilizers.
Keywords: power system; sliding mode control; type-2 fuzzy system; adaptive control; Lyapunov;.
Towards compact swarm intelligence: a new compact firefly optimisation technique
by Lyes Tighzert, Cyril Fonlupt, Boubekeur Mendil
Abstract: Firefly algorithms (FA) is a recent and promising swarm intelligence algorithm. It is inspired by the modelling of brightness and attractiveness manifested by fireflies. Like other population-based algorithms, it has the drawbacks of high computational cost and memory storage. This paper deals with this problem and introduces a compact firefly optimisation technique with minimal computational and memory requirements. So, we present four new variants of compact firefly algorithms that require only a minimal computational cost. The swarm is compacted and represented by a probability of density function (PDF). This idea is inspired from compact evolutionary algorithms (cEAs). Two solutions of memory storage of the population are presented and analyzed. The first is based on normal PDF and the second on uniform PDF. Furthermore, two versions of compact L
Keywords: compact firefly algorithms; compact swarm intelligence; Lévy flight; optimisation; gymnastics; humanoid.
Secure communication scheme using chaotic time-varying delayed system
by Benkouider Khaled, Halimi Meriem, Bouden Toufik
Abstract: In this paper, we are interested in the study of discrete-time delayed chaotic communication system where the delay is injected into the state vector of the chaotic system in order to increase the security of the transmission. Our aim is reconstructing the encrypted information at the receiver. However, this reconstruction requires knowing the delay. We propose a method based on the use of unknown input polytopic observers, to estimate the unknown delay and the encrypted information. The obtained simulation results show the effectiveness of this method.
Keywords: chaotic communication; secure transmission; synchronization; delayed system; LPV system; unknown input observer.
Special Issue on: Computational Intelligence and Applications
Intelligent game-based learning: an effective learning model approach
by Tanzila Saba
Abstract: Game-Based Learning (GBL) broadly refers to the use of video games applications to support teaching and learning processes. This research focuses on the concept of GBL in the context of stimulating interest in the field of computer science education specifically. In contrast to theoretical learning, GBL is a practical learning approach that is meant to teach and be enjoyed at the same time. Additionally, a GBL model with visual features has been proposed and tested. Promising feedback has received from learners through the post conducted surveys. The research findings exhibit that GBL is an effective methodology in transferring knowledge, enhancing learning, and making the learning a more enjoyable process in computer science studies than just the theoretical approach.
Keywords: binary games; game-based learning; logical games; theoretical learning.
Special Issue on: Emerging Trends in Computer Applications in Technology
Cooperative evaluation mechanism based on the optimal decision of DE-CA-CR
by Cui Guotao, Zhang Ying
Abstract: University-enterprise deep cooperation is the important measure to overcome the vocational education development bottleneck. However, owing to the restrictions of factors such as market environment, school-enterprise cooperation system and mechanism, the school-enterprise cooperation is just at a junior level and forms no benign interaction between them. This paper establishes a comprehensive evaluation index system of university-enterprise through the stakeholder analysis method, and builds a university-enterprise deep cooperation evaluation model combined with DEA-CCR optimal decision. Besides, based on previous research and survey, the paper tries to analyse the problems in school-enterprise cooperation and influencing factors from the perspective of enterprise, and thus to establish the school-enterprise cooperation performance evaluation model and conduct verification through living examples.
Keywords: industry-education integration; school-enterprise cooperation; confidence mechanism; data envelopment analysis.
Improved data envelopment analysis model based on geometric mean model
by Feng Yanhong, Wang Zhongfu
Abstract: According to the characteristics of the circular economy conception of underdeveloped regions under new economy situation, the improved data envelopment analysis model based on geometric mean model is reconstructed in this paper through increasing unexpected input (waste recycling quantity) in the input and increasing unexpected output (waste discharge quantity) in the output. Then, the improved data envelopment analysis model was adopted to evaluate the efficiency of the circular economy of the underdeveloped regions in 31 provinces and cities in China during 2002-2012. The research result shows that the efficiency of the circular economy of underdeveloped regions in China is generally improved, and the scale efficiency gradually tends to reach the optimum efficiency, but the obvious insufficiency of pure technical efficiency has influenced the development level of the circular economy of underdeveloped regions in China.
Keywords: underdevelopment; circular economy of underdeveloped region; efficiency evaluation; DEA model.
Improved face recognition with accelerated robust features improved by means of mean shift K-means clustering
by Jiao Ding, Minfeng Zhang, Tianfei Zhang, Haiyan Long, Meiyu Liang
Abstract: To improve the precision of the heterogeneous face recognition model, a heterogeneous face recognition model method based on binary multilayer Gabor extreme learning machine (GELM) is proposed in this paper. Firstly, a random weighted Gabor feature extraction scheme is proposed based on pixel weight. It propagates the locally geometric input image sub-block to the hidden node, and embeds the extracted Gabor feature into the hidden layer. Moreover, it conducts random weighting and sum using a group of Gabor kernels so as to realise a convolution operation of the nonlinear activation function of the propagated pixel. Then, it estimates the output layer by means of linear weighting that is similar to extreme learning machine (ELM). At last, the performance of heterogeneous face recognition method of the proposed algorithm is verified through BERC VIS-TIR database and CASIA NIR-VIS 2.0 database.
Keywords: mean shift; K-means clustering; face recognition; precision.
Analysis of system implementation effect based on Bayes analysis of imbalanced measures
by Huang Liumei, Lian Huijie
Abstract: In order to enhance the analysis effectiveness of the compulsory education policy implementation situation of regional economic development, this paper puts forward an analysis method based on inconsistent-measurement Bayes of the compulsory education policy implementation situation. Firstly, it researches the evaluation index model of compulsory education resource allocation, and establishes the evaluation model based on the relevant indexes, such as education infrastructure, teacher resources and appropriation for education; secondly, it puts forward a Bayes filtering algorithm based on forward-backward compression to effectively handle the noising and uncertainty problems that exist in the compulsory education policy implementation, realise the effective estimation on the equilibrium model of compulsory education policy implementation, and deeply analyse the influence of compulsory education policy in domestic compulsory education policy implementation trend. The research result verifies the effectiveness of the method.
Keywords: regional economy; compulsory education; policy implementation; Bayes filtering.
Effect of cognitive need and purchase involvement on information processing in the online shopping decision-making
by Liu Chuanlei, Chen Baishu, Huang Dijian
Abstract: This research uses the information board technology to simulate the network shopping decision task. Through the experiment, it discusses the influence of cognitive needs and purchase involvement on information processing in network shopping decision-making. The experimental results show that (1) cognitive needs have a significant influence on information processing in shopping decision; (2) purchase involvement has a significant impact on information processing in the decision-making process of network shopping; (3) cognitive needs and purchase involvement have a significant impact on information processing in the network shopping decision making process.
Keywords: cognitive need; purchase involvement; online shopping; information processing.
Analysis and prediction of autistic children's game characteristics
by Liu Chuanlei, Han Yuanfei, Li Jiao
Abstract: Children with autism showed great defects in the playing of games, and the study of autistic children games is unseen. This research, designated "CePingBiao of autistic childrens game ability", undertook a study on the game features of 130 children diagnosed with autism, and gave 69 children rehabilitation training game ability for 4 months. The results showed: (1) the autistic children's ability of game on the game type has high and low points; the best body level of game development, followed by the structure of the game, again is a symbol of games, social games, and the lowest development level is the rules of the game; (2) age significantly influences the ability of children with autism; with the increase of age, autism game levels increase, and 5 to 6 years old is the rapid development period; after this period, the game development level of children with autism begins to fall; the development curve is inverted U type; (3) the training time influences the structure of the autistic children's game development level; the structure game level of children with autism increased with the increase of training time, but the training period needs to be at least 9 months; (4) language is the important factor affecting the development of autistic children's game-playing ability; the language level obviously promotes the development of this ability. The paper concludes that there are differences in game type, age, training duration and language level of children with autism, and their game levels can predict their overall development level.
Keywords: autistic children; play ability; assessment.
Development mode of circular economy industrial cluster based on game theory
by Feng Yanhong, Wang Zhongfu
Abstract: Policy decision is an external driving factor for cyclic economy, under the perspective of global value chain, the subjects of the interested parties are the cyclic economy enterprises and the government responsible for supervision. In the development process of enterprises, the cyclic economy industry cluster development mode is the important link for development, which has an important position in the environmentally friendly and resource conservation development goal of the Chinese economy, and is an urgent problem needing to be solved. This paper, based on the game analysis method, establishes a model for the relationship between enterprise and government under the cyclic economy industry cluster development mode, and conducts game analysis on the cyclic economy innovation process under the driving of the government, thus to research the relationship between the cyclic economy enterprise and the government responsible for supervision under the cyclic economy development mode from the perspective of theory. Finally, based on the model establishment and analysis results, the paper puts forward reasonable suggestions on strengthening the cyclic economy industry cluster development.
Keywords: cyclic economy industry cluster; industry cluster; innovation game; external factor.
Community discovery method based on complex network of data fusion based on super-network perspective
by Li Pei
Abstract: To enhance the computational efficiency and precision of community discovery, a community discovery algorithm with the mixed label based on the minimum description length (MDL) of information compression is proposed in this paper. Firstly, the community detection is converted into an information compression problem of seeking an effective network structure, and the quality evaluation function is constructed based on MDL criterion. Secondly, the community discovery algorithm with heuristic mixed label movement is constructed based on the label node movement algorithm and Louvain community addition algorithm so as to reduce the quality evaluation function. At last, the simulation experiment in the standard test set and API capture Sina microblog dataset shows that the proposed algorithm is superior to the selected comparison algorithm in computational efficiency and precision.
Keywords: super-network perspective; label movement; complex network; heuristic; community discovery.
Relay protection method based on decentralised control logic based on two-sided active excitation detection
by Wei Bin, Xu Chong, Wu Xiaokang, Gao Chao, Xu Jinxing
Abstract: To lower the wireless charge coil losses and leakage level of magnetic field for electromobiles, a relay method with decentralised control logic based on active excitation detection by a secondary side is proposed. It is guaranteed to only excite the primary coil below electromobile that raises charging requirement, and to realise precise localisation and local power supply. The method fully multiplexes the primary/secondary power coil and needs no additional sensing unit or centralised signal line. The active excitation detection circuit is configured as a series connection compensating network and uses its characteristic of frequency splitting to enhance the intensity of the detecting signal and avoid the overflow of the secondary side's active excitation. A reasonable relay control flow is designed to reduce the power needed for detection and avoid the conflict of simultaneous excitation of primary and secondary sides. Lastly, a simulative experiment was conducted to verify the feasibility of relay method proposed.
Keywords: segmental; electromobile; wireless charging; offset compensation.
A dynamic modelling method for dynamic wireless charging system of electric vehicles based on dual LCL non-resonant compensation
by Huang Xiaohua, Wei Bin, Gao Chao, Wu Xiaokang, Xu Jinxing
Abstract: This paper starts with the mutual inductance coupling model and takes the compensation structure of "series connection-series connection" as an example to respectively deduce the mathematical expressions for system output power and efficiency when traditional resonant compensation strategy and non-resonant compensation strategy are adopted, so as to compare the advantages and disadvantages of the two strategies from the aspect of the influence of coupling coefficient change on the two strategies output power and efficiency. Then, to cope with the drawbacks of the non-resonant compensation strategy, a dynamic modelling method for electromobile's dynamic wireless charging system based on dual-LCL non-resonant compensation is proposed to analyse the structure's function of improving the system's power transmission capacity and determination method of different component parameters from the perspectives of circuit equivalence and impedance conversion. Models are built for the rectifier and filter circuit, the DC-DC circuit and the loading battery pack for the receiving terminal, and the characteristics of each link are expressed by mathematical expressions.
Keywords: LCL non-resonant compensation; electromobile; wireless charging; dynamic modelling; rectifying and filtering.
An image segmentation algorithm based on combination of slope width reduction and cross-cortical model
by Zhang Zhen
Abstract: An image segmentation algorithm based on the ramp width reduction combined with an intersecting cortical model (ICM) is proposed to resolve the problem that ICM in the segmentation of images with weak edges produces geometric distortion. By virtue of prewitt boundary operator and edge ramp model, the algorithm defines the objective edge point, adjusts the gray level of edge pixel, and reduces the width of image edge. On this basis, the paper uses 2D histogram to expand the cross entropy to 2D space so as to obtain the optical segmentation threshold of ICM. The experiment indicates that the algorithm not only overcomes the impact of edge blur and segments the image with weak edge accurately, but also improves the processing speed greatly.
Keywords: automatic local ratio; Chan-Vese model; image segmentation; boundary operator; intersecting cortical model; cross entropy.
A comprehensive evaluation model based on fuzzy meta-association rules
by Qian Hao-yun
Abstract: At present, the attention to the competitiveness of the provincial-level administrative region in China has become a hot topic. Based on principles of scientificity, systematicness, comparability and feasibility, this paper establishes a set of index systems for comprehensively evaluating the comprehensive competitiveness of provincial-level administrative regions, and proposes a kind of fuzzy meta-association rule method based on hierarchy theory. It carries out binary fusion extraction of the meta-rules for urban development evaluation element knowledge by making use of the similar structure of the data stored by the development branch in each city, with no need to process the entire dataset. It is able to obtain results/modes from a single database to reduce the time required for rule mining. Finally, the comprehensive competitiveness level of the provincial-level administrative regions all over the country is analysed and evaluated from the microcosmic level and macroscopic level through factor analysis and clustering analysis.
Keywords: hierarchical analysis; association rules; competitiveness; comprehensive evaluation.
A risk preference model for teaching resource allocation based on functional link fuzzy neural network algorithm classifier
by Wei Tongpeng, Chen Li
Abstract: The volleyball teaching resource allocation model based on the functional link fuzzy neural network algorithm is proposed to improve the effectiveness of volleyball teaching resource allocation in the course arrangement process. Firstly, the risk preference allocation model for volleyball teaching resource allocation is designed based on the functional link fuzzy neural network algorithm classifier, and the risk is divided into existing and non-existing risk preferences and the functional link fuzzy neural network algorithm classifier is used to achieve the training and data optimisation of the volleyball teaching resource allocation dataset. Secondly, considering that the functional link fuzzy neural network algorithm classifier may break down in the volleyball teaching resource allocation prediction process, the functional link fuzzy neural network algorithm is used to achieve optimisation of the volleyball teaching resource allocation process. Finally, the stimulation research on the volleyball teaching resource allocation model example shows that a more reasonable volleyball teaching resource allocation model can be obtained by the proposed algorithm, reflecting the effectiveness of the algorithm.
Keywords: volleyball teaching resource; neural network; functional link; resource allocation.
Fusion algorithm for information interaction control of multi-UAVs based on intelligent algorithm
by Chen Guangming
Abstract: This paper is devoted to designing a kind of UAV robust information interaction detection and tracking control system suitable for external interference suppression. Firstly, it models the UAV rotor as a linear parameter-varying system (LPV), takes it as an objective to make system design, and considers information interaction an detection and isolation scheme through an observer library, to detect and isolate sensor information interaction. Then, the paper improves a kind of existing adaptive variable space algorithm, introduces the algorithm thought into improvement of particle swarm optimisation, and when the evolutional generation of population reaches an integer multiple of a preset period, automatically expands or shrinks the size of the search space according to the improved adaptive variable space algorithm, which automatically searches for proper search space, improves convergence rate and accuracy, and effectively prevents premature convergence of particle swarm optimisation. Finally, the effectiveness of the algorithm is verified through experiment in a simulation model.
Keywords: information interaction; control fusion; particle swarm optimisation.
Computer-based outdoor sport sustainable development using wavelet neural network
by Chen Shan
Abstract: In order to enhance the effectiveness of research on sustainable square dancing under the background of national fitness, this paper puts forward a research method based on a wavelet neural network, applies the time series prediction theory of the wavelet neural network into the prediction on sustainable square dancing, obtains the LF approximate part and HF approximate part in the sustainable square dancing data through wavelet decomposition and restructuring, and then, based on analysing the good and bad models, selects the most effective model or model combination to establish the prediction model for researching sustainable square dancing. Finally, it conducts model simulation by aid of the actual sustainable square dancing data, and the result shows that the model can effectively enhance the prediction precision of sustainable square dancing.
Keywords: square dancing; wavelet analysis; neural network.
Design of extensible multi-source signal acquisition device based on DSP and STM32
by Li Bo, Xiong Di, Guohua Chen
Abstract: As large numbers of multi-type sensor signals are required to be collected in numerical control machine tool tests, a general acquisition system could only collect certain types of sensor signal, which did not have extended function, so that its application scope in machine tool test was limited, leading to excessive acquisition equipment categories and insufficient compatibility with each other. Aiming at above-mentioned problems, a kind of extensible multisource acquisition system based on DSP and STM32 was designed, which took a microcontroller as DSP and STM32. DSP microcontroller was mainly used to complete data acquisition and output functions, while STM32 microcontroller was mainly used to complete data storage and communication functions. The acquisition system had a multisource signal interface, which could automatically identify analogue signal type and support parallel acquisition for temperature, vibration, voltage, current, pressure, displacement, and other types of signal. Extended functions could be realised so as to increase the multichannel analogue input interface through a removable extended module so as to satisfy acquisition requirements for different machine tool tests.
Keywords: numerical control machine tool; extensible; multisource acquisition.
Design and implementation of LTE physical layer on FPGA
by V. Venkataramanan, S. Lakshmi, A. Vineet Kanetkar
Abstract: Changing trends in the communication industry pertain to the configuration of devices and their processing for maximised result. Each device needs a processing unit comprising a microcontroller or a Field Programmable Gate Array (FPGA). This paper deals with the use of FPGAs and how they can be configured as hardware in loop (HIL) for validation along with Simulink and Xilinx System Generator (XSG). Further, their compatibility is mentioned for long term use and durability in communication. The comparison of related work in the field of communication is done with the FPGA implementation of a Long Term Evolution (LTE) physical layer with different modulation schemes, different antenna configurations and different signal-to-noise ratio systems implemented on Virtex and Spartan FPGA boards. On the other hand, the simulation is carried out with Xilinx Vivado Design suite to analyse the power, resource use, timing summary, and memory use.
Keywords: field programmable gate array; hardware co-simulation; LTE; MIMO; OFDM; 3GPP.
Innovation mechanism of cluster industry based on weighted time-varying multi criteria and similarity evaluation method
by Feng Yanhong, Wang Zhongfu
Abstract: In order to enhance the effectiveness of the cooperation mechanism of cluster industry innovation of booming megalopolises, this paper puts forward a research method based on weight time-varying multi-criteria and similarity evaluation method for the cooperation mechanism of cluster industry innovation of booming megalopolises. Firstly, under the background of internet+, it researches the indicator system establishment of the cluster industry innovation platform of booming megalopolises, establishes the comprehensive evaluation system with five first-level indicators in scientific research innovation capability, intelligent production and service support capability, information transmission capability, infrastructure and environment support capability and platform system establishment capability. Secondly, it puts forward the weight time-varying multi-criteria and similarity evaluation method to enhance the industry innovation cooperation mechanism recommendation precision, combined with the wight time-varying process, deeply considers the criteria weight of different times periods to enhance the decision scientificity of the industry innovation cooperation mechanism; finally, it verifies the effectiveness of algorithm through empirical analysis.
Keywords: megalopolises; industry innovation; cooperation mechanism; weight time-varying; similarity evaluation.
Collaborative sparse unmixing using variable splitting and augmented Lagrangian with total variation
by Nareshkumar Patel, Himanshukumar Soni
Abstract: Linear Spectral Unmixing (LSU) is a widely used technique in the field of Remote
Sensing (RS) for the estimation of fractional abundances of endmembers and their
spectral signatures. Large data size, poor spatial resolution, non-availability of pure endmember signatures in dataset, mixing of materials at various scales and variability in spectral signature make LSU a challenging and inverse-ill posed task. Broadly there are three basic approaches to manage the LSU problem: geometrical, statistical and sparse regression. The first and second approaches are types of blind source separation (BSS). The third approach assumes the availability of some standard
publicly available spectral libraries, which contain signatures of many materials
measured on the Earth's surface using advance spectroradiometry. In the sparse re-
gression approach, the problem of LSU is simplified to finding the optimal subset
of spectral signatures from the library known in advance. In this paper, the con-
cept of collaborative sparse regression is incorporated to improve the performance
of the existing SUnSAL-TV algorithm. SUnSAL-TV is a recently proposed Total Variation(TV) spatial regularisation based approach. Our simulation results conducted
for standard and publicly available synthetic fractal dataset show 10 to 15%performance improvement in signal to reconstruction error for different
data cubes. Simulation is also performed for a subset of real cuprite data cube and
compared with the outcome of recent algorithms.
Keywords: linear spectral unmixing; sparse regression; augmented Lagrangian; ADMM; hyperspectral unmixing; total variation; collaborative.
Special Issue on: ISMIC 2018 Information Processing and Control Technologies
Automatic selection of lexical features for detecting Alzheimer's disease using bag-of-words model and genetic algorithm
by Gang Lyu, Aimei Dong
Abstract: Early detection of Alzheimer's disease is the key to treatment. Neuropsychological testing has the advantages of being non-invasive and low-cost, but the need for manual selection of features and expert diagnosis is not conducive to the popularity of this method. This paper proposes an approach for automatically extracting and selecting features from texts. First, it uses the bag-of-words model of natural language processing technology to extract all the vocabulary features in the texts. Secondly, unlike the manual selection of features by t-test, it uses the genetic algorithm to select lexical features automatically. We tested the new approach with the DementiaBank database. Its classification accuracy for Alzheimer's disease is 79%, close to the best value of the hand-crafted-feature-based method. The new approach also has the ability to process data quickly and automatically, which can greatly help clinicians improve their work.
Keywords: bag-of-words model; genetic algorithm; hyperparameter; machine learning; naïve Bayes algorithm; Alzheimer's disease.
A new topology and power control of grid-connected photovoltaic array
by Li-ping Zhong
Abstract: Under the partial shading, the series photovoltaic modules will generate additional power loss and present a multi-peak power-voltage curve that causes difficulties for the maximum power point tracking. By using a novel grid-connected topology and power control method presented in this paper, the photovoltaic array can be connected through a full parallel structure and thus the shortcomings mentioned above can be overcome. A higher voltage, required for grid connection, also can be obtained through the topology without the need of any step-up transformer or boost circuit. Furthermore, the output power of the photovoltaic array can be adjusted by controlling the phase of the modulated wave. As a result, the maximum power point tracking can be achieved with a simple method. The simulation and experiment results verified the validity of the proposed topology and control method.
Keywords: partial shading; multi-peak power-voltage curve; CLC immittance converter; phase control; maximum power point tracking.
Low frequency structure-borne noise refinement based on rigid-flexible coupling model of powertrain mounting system
by Rang-Lin Fan, Zhen-Nan Fei, Cheng-Cheng Feng, Fang Yin, Wei-Cun Zhang
Abstract: The refinement of low frequency structure-borne noise generated by automotive powertrain mounting system usually adopts transfer path analysis (TPA) and vehicle body plate optimisation, which requires a lot of simulation or experiment work based on the vehicle body. This paper proposes a simple method to provide an optimal mount stiffness target for the refinement of structure-borne noise. The method is based on rigid-flexible coupling model of powertrain mounting system, and the model requires complete and accurate parameters such as mass, inertia, position and stiffness. The excitation forces of engine which are used as input of the rigid-flexible coupling model are identified by an indirect semi-experiment method. Based on this model, the direction of mount stiffness with maximum sensitivity to the dynamic characteristics of the powertrain mounting system is identified. Then the low frequency structure-borne noise is refined by changing the mount stiffness in this direction.
Keywords: structure-borne noise; rigid-flexible coupling model; rigid modal; powertrain mounting system; automotive.
An adaptive multi-threshold segmentation algorithm for complex images under unstable
by Wei Ding, Yanfang Zhao, Reilei Zhang
Abstract: Images acquired from the actual manufacturing
Keywords: threshold segmentation; peak distribution; gray level probability density; prior knowledge.
Multiple cell tracking by generalised labelled multi-Bernoulli filter
by Jian Shi, Mingli Lu
Abstract: Cell detection and tracking in microscopy images are of great importance to medical research and related fields. In this paper, a generalised labelled multi-Bernoulli (GLMB) track-before-detect (TBD) filter is proposed for the tracking of multiple cells. In this filter, GLMB based on random finite set (RFS) theory is used to jointly estimate the positions and the numbers of cells in images, and TBD is adopted to track cells without an explicit detection step. The experimental results indicate that the proposed method can accurately discover cells and maintain their tracks in low contrast image sequences.
Keywords: cell tracking; random finite set; track–before–detect; generalised labelled multi-Bernoulli.
Broad learning system for human activity recognition using sensor data
by Aiqiang Yang, Xinghong Yu, Tingli Su, Xuebo Jin, Jianlei Kong
Abstract: In a multi-sensor environment, it is efficient to record and reflect peoples information of activities, using the large amount of data. However, the data cannot directly display the form of activity itself so that it is necessary to do the further job of exploration and processing. Deep Learning (DL) methods have attracted more attention and have shown some superior performance, while they have the problem of structural complexity. Therefore, this paper creatively used Broad Learning System (BLS) method for human activity recognition. We use sliding window to get the data segmented. The weights involved in are fine-tuned by pseudo-inverse and ridge regression algorithms, and we achieve an accurate classification of activities. The method is verified by using OPPORTUNITY dataset. The results show that this method can greatly shorten the learning time and improve the accuracy, as well as the performance in comparison with traditional method.
Keywords: broad learning system; human activity recognition; sensor data; sliding window processingrn.
Crowd counting via scale-adaptive convolutional neural network in extremely dense crowd images
by Ran Yan, Shengrong Gong, Shan Zhong
Abstract: Crowd counting, a high accuracy and high-speed technology, has been applied in new retail, shopping mall, underground, rail station and vehicle surveillance systems. However, owing to the inconsistent sizes of human heads, there are a lot of counting errors and instability of the crowd density estimation in extremely dense crowd images. Therefore, a scale-adaptive CNN (Convolutional Neural Network) architecture is proposed by introducing residual network on the basis of multi-column CNN. In the process of model training, joint learning is proposed in this paper. Through alternating training for residual network and multi-column CNN, network parameters with the best accuracy are selected after iteration. Joint learning helps to enhance the modelling ability for massive scale transformation and the scale self-adaptability of the network. The proposed method is tested on public dense crowd datasets. Experimental results prove that scale-adaptive CNN shows better counting capability than the current state-of-the-art method.
Keywords: crowd counting; density estimation; convolutional neural network; scale-adaptive; joint learning.
Design of a new type of float flowmeter and remote monitoring system based on ARM microcontroller
by Wu Qian, Shen Bingbing, Jiang Ling, Zhao Fengsheng, Hua Liang
Abstract: A new type of float flowmeter remote monitoring system developed in this paper has been optimised in terms of system integration, software and hardware. The system uses a cost-effective 32-bit ARM microcontroller as the control processing unit to achieve accurate measurement of liquid fluid flow. The host computer monitoring system and the client computer of flowmeter controllers use the prescribed communication protocol for data interaction. Each port number of the IP address of the host computer server can be connected to 255 devices. The system makes full use of the advantages of the IOT (Internet of Things) wireless communication technology, ARM technology and communication protocol, which realises the complete separation between the monitoring terminal of the host computer and the field device of the actual industrial site conditions. The user can remotely monitor the float flowmeter at the job site on any PC. The instrument has features of high precision, efficiency, explicit structure, and better maintainability and it has broad prospects in application.
Keywords: remote monitoring; ARM microcontroller; wireless communication; communication protocol.
An operation sequence based temporal multilayer networks model for production process in flexible manufacturing systems
by M.E.I. Dai, Zhicheng Ji, Yan Wang
Abstract: This paper focuses on flexible manufacturing systems, which are typical discrete event systems. Unlike traditional approaches for the system, this paper provides a multidimensional, ordinal, temporal topology structure to depict the dynamic production process from the perspective of complex networks. Firstly, the factors of a real manufacturing environment are mapped to the multilayer networks and the framework of the model is given. The proposed temporal multilayer networks model includes intra-layers and inter-layers, involving two aspects of task and resource nodes and time-dependent graphs, respectively. Secondly, the generation rules to construct the networks are presented. A novel approach for evaluating bottleneck resources in time blocks is presented to develop the networks model. Finally, the temporal multilayer networks model is generated based on the theory of constraints. The proposed model based on operation sequence presents superiority in measurement of time-dependent performance of the system. Moreover, an application case for energy consumption evaluation confirms that the proposed model can support exploratory analysis of the time-related performance criterion in discrete manufacturing.
Keywords: multilayer networks; temporal networks; flexible manufacturing system; generation rule; bottleneck identification.
Analysis of an approach to reducing drops of secondary user on primary user emulation attack
by Hui Sun, Chuang Yang, Rui Wang, Sabir Ghauri
Abstract: In this paper, we propose a method for a Cognitive Radio Network (CRN) to reduce the drop probability of a secondary user (SU) due to the primary user emulation attack (PUEA) by a malicious user (MU). Instead of abandoning the current channel, a novel method called a pause approach is used if the primary user (PU) or an MU accesses the same channel. This method helps the SU to find the attack behaviour of the MU and increases the network throughput performance. Also, we analyse the channel states with a Continuous Time Markov Chain (CTMC). The simulation results validate the proposed method based on Matlab.
Keywords: cognitive radio; PUEA; Markov chain; drop probability.
A model for target acquisition and edge detection under complex scenes
by Fei Wang, Jihong Zhu
Abstract: For the needs of target acquisition and edge extract under complex scenes, a method based on layer by layer segmentation is proposed. First of all, morphological reconstruction technology is combined with an ant colony edge detection method to execute pre-segmentation to get the foreground that targets locate. Secondly, a line scanning technique is used to divide the foreground into several parts for further segmentation. And then, mean difference characteristic is adopted to determine the seed point, with which a region-growing algorithm is applied to find true targets. Finally, an active contour method is used to find the edges of targets. The experimental results show that our method is effective in finding targets and extracting edges under complex scenes.
Keywords: morphology reconstruction; ant algorithm; active contour.
Multi-threading parallel reinforcement learning
by Qiming Fu, Yiyi Kang
Abstract: With respect to the problem of the slow convergence of the traditional reinforcement learning algorithm in practical applications, we proposed a novel multi-threading parallel reinforcement learning (MPRL) algorithm. MPRL is mainly composed of two parts. One is the FCM-based reinforcement learning multi-threading partitioning method, which transforms the multi-threading partitioning problem into a clustering partition problem to obtain the optimal multi-threading partitioning solution. Another is the parallel reinforcement learning framework, which makes the parallel execution between the policy evaluation and the interaction with the environment. In the learning process, the experience replay is adopted to update the value function, which can also solve the problem of the non-convergence in the off-policy evaluation. Experimentally, the MPRL algorithm is applied to the windy grid world problem and the cart pole problem, and compared with Q-Learning, Sarsa and KCACL. The experimental results show that MPRL has a faster convergence rate and better convergence performance.
Keywords: reinforcement learning; multi-threading technology; thread partitioning; parallel reinforcement learning; experience replay.
Rapid freshness prediction of crab based on a portable electronic nose system
by Peiyi Zhu, Yulin Zhang, Lu Ding
Abstract: In this paper, an automatic freshness prediction system for the living Chinese mitten crab was explored, which was formed from an electronic nose based on seven metal oxide semiconductor sensors. The prediction system acquired test data from the characteristic compounds in the headspace of the crab, and then was dealt with four different dimension reduction algorithms including PCA, LDA, KPCA and LE to reduce dimensions and extract effective features of sensor scores. Experimental results illustrated that the prediction system sensitively responded to crabs. PCA and LDA results failed to differentiate the response data of the living crabs. LE and KPCA were able to identify the different response data of crab samples. Back propagation neural network was used as a prediction model after dimension reduction. The model based on LE-BPNN reached a high identification rate of 90.6%. The simulation and experiment results was clarified the prediction system can estimate the freshness of the living crab.
Keywords: Chinese mitten crab; electronic nose; Freshness; Laplacian eigenmaps; back propagation neural network.
Research on robot location based on improved method of map feature matching
by Mao Limin, P.U. Yuhuan, Wang Liangyu
Abstract: With respect to robot self-positioning, this study reports that the map feature extraction algorithm based on Euclidean distance is improved, the processing of outliers and class division points in line segment landmark fitting is added, and the slope and intercept of the line are added. The aggregation step reduces the influence of class over-segmentation of the map feature extraction. According to RANSAC feature matching, a map matching method based on corner points and line segment landmarks is proposed.
Keywords: straight line fitting; map feature matching; data point classification; feature extraction; line landmark.
Development of shipbuilding safety information monitoring and management system
by Qing Zhang, Liang Hua, Xiaojie Tian, Zijun Tang, Lubing Nian
Abstract: Production safety accidents in the shipbuilding industry in China frequently occur nowadays. Based on the key technology of the internet of things as the communication basis, this paper designs a new type of shipbuilding safety monitoring system based on wireless communication. The wireless heterogeneous network is organically formed to solve the special problem of shielding a ship's airtight steel structure, as well as the problem of networking in bad working conditions. It adopts multi-sensor coordination and pattern recognition technology to achieve reliable collection and intelligent processing of environmental information and human body information. The system adopts network management to realise the connection between all items and networks, based on multi-sensor real-time monitoring and intelligent analysis of various physiological conditions and surrounding environmental conditions, combined with RFID technology and ZigBee technology for identity security identification. Finally, the collected information is transmitted to the host computer and Android client through WiFi.
Keywords: shipbuilding; safety monitoring system; internet of things system; multi-sensor; host computer management system.
Special Issue on: Advances in Computer Graphics and Imaging
Research on the Design of Visual Interface in Information Visualization
by Guangtao Ma, Tao Liu, Yang Zhou, Jun Li
Special Issue on: Machine Vision and Computational Intelligence in Recent Industrial Practice
Robust Skin Segmentation using Color Space Switching
by Ankit Chaudhary, Ankur Gupta
Special Issue on: Xxxx
A Query Driven Method of Mapping from Global Ontology to Local Ontology in Ontology-based Data Integration
by Haifei Zhang