Computer Engineering and Networking

In contrast to conventional radars, phased array radars have the capability to switch the direction of the radar beam very quickly without inertia. The measurements from phased array radar can contribute to many application fields such as data and intelligence process and radar performance evaluation. However, it often costs more than we can bear to obtain phased array radar measurements. It is necessary to model and simulate the phased array radar, especially for missile-borne phased array radar. This chapter lays a strong emphasis on the search and simulation technology of missile-borne phased array radar. According to operational theory of phased array radar, this chapter focuses on the functional modeling and simulation techniques. It contains three parts: beam arrangement of phased array radar in sin coordinate, parameter optimization of missile-borne phased array radar, and a function simulation model of missile-borne phased array radar.

A second-order algorithm is presented to calculate the parallel projection of a parametric curve onto a parametric surface in this chapter. The essence of our approach is to transform the problem of computing parallel projection curve on the parametric surface into that of computing parametric projection curve in the two-dimensional parametric domain of the surface. First- and second-order differential geometric characteristics of the parametric projection curve in the parametric domain of the surface are firstly analyzed. A marching method based on second-order Taylor Approximation is formulated to calculate the parametric projection curve. A first-order correction technique is developed to depress the error caused by the truncated higher order terms in the marching method. Several examples are finally implemented to demonstrate the effectiveness of the proposed scheme. Experimental results indicate that both the computational efficiency and accuracy of the presented method have dominant performance as compared with the first-order differential equation method.

Looking-up standard processing time table is a commonly used and important determination method of processing time. However, the large error in nonstandard nodes brings adverse effect on its accuracy. In view of the problem, a computation method of processing time based on back propagation neural network (BPNN) and genetic algorithm (GA) is proposed. Several key technologies of BPNN based on Matlab, including computation of the number of neurons in hidden layer, determination of training algorithm, and affecting factors of generalization ability, are researched in depth. In order to improve the training efficiency of BPNN, GA is used to optimize its connection weights and thresholds. The encoding method, selection operation, crossover, and mutation operation of GA are discussed in detail. The higher computation precision and faster operation speed of the proposed method is demonstrated through application cases.

Integral sliding mode control is formulated with respect to an uncertain continuous network control system with the state delay. The parameter uncertainty is assumed to be the norm-bounded and satisfy the sliding mode matching requirements. The switching function is presented which include the integral term of the state feedback gain and the sliding mode compensator. The sliding mode controller is designed which is divided into the equivalent controller and switching controller, so the reachability of the sliding surface is ensured. A sufficient condition is derived by means of linear matrix inequality such that the asymptotical stability of the closed-loop system is guaranteed. The validity and feasibility of the proposed approach is investigated via the corresponding numerical simulation.

In array pattern synthesis, it is often designed to achieve the desired radiation pattern. In this paper, real-coded genetic algorithm (RGA) optimization method is presented to optimize the value of weights of each antenna element to minimize side lobe level of the uniform spaced linear array geometries with a certain main beam width. The optimization program is done by using MATLAB. It compared with the conventional analytical methods such as Chebyshev and Taylor through radiation patterns with different number of elements and intervals of each element. The simulation results show that the optimization results are of little difference when

d

≥

λ

/2, but GA can get a more optimal result when

d

<

λ

/2. The application of genetic algorithm for pattern synthesis is found to be useful.

In order to solve the prediction problem of interaction between proteins, we use a wavelet coefficient combined with artificial neural network method, improving the prediction accuracy of the problem of protein–protein interactions. By introducing the Biorthogonal Wavelet 3.3 coefficients as the feature extraction method and the three-layer feedforward neural network as a classifier, we solve the problem of protein interaction effectively. Using the Human dataset verifies the validity of this method. Through testing the Human dataset, using Biorthogonal Wavelet 3.3 coefficient combined with the three-layer feedforward neural network, solve the prediction problem of protein interactions with well results. This combination of wavelet coefficients and the three-layer feedforward neural network to predict protein interaction problem is an effective method. At the same time, compared with other prediction methods, this method performs at least 4 % higher accuracy than the better accuracy of auto-covariance (11) combined with PNN on the same dataset.

In order to improve the efficiency of slot utilization of Link16, the traditional binary tree model of timeslot allocation is improved, and the performances of three kinds of timeslot assignment algorithm are compared and analyzed in the average transmission delay and message sending failure rate. When the user is determined and the message arrival rate is lower, the fixed timeslot allocation is simple and effective. When there are a lot of users or the user is uncertain, and the message arrival rate is very low, competitive slot allocation should be more efficient. When the number of users is small, the user message arrives sudden, and arrival rate is high, the performance of the dynamic slot allocation has a very distinct advantage. Finally, based on the characters of different timeslot assignment algorithms, the overall timeslot allocation scheme is presented for Link16 network.

Routing algorithm is one of the key technologies of wireless sensor network. Based on the LEACH algorithm, the second choice of cluster head algorithm for WSN is proposed. It takes into account the cluster head as residual energy and distance to BS, and then it chooses a senior cluster head. The improved algorithm avoids direct communication between BS and the cluster head which has low energy and is far away from BS, which prolongs the lifetime of network and enhances the ability of data collection. The experiments show that this technique of event-driven can cut down the data transmission and further extend the lifetime of network.

Dynamic voltage scaling is an effective energy minimization technique by conjointly changing the supply voltage and the operational frequency during run-time. In this chapter, an improved ant colony system is presented for distributed systems consisting dynamic voltage scalable processing elements. The energy saving can be obtained by using the DVS algorithm on the schedule obtained by the presented scheduling algorithm. The pheromone information of the ants and the heuristic information inspired by the list heuristic rule and energy consumption are combined together to guide the ants search. The parameter value of heuristic is varied from higher value to lower value to lessen its impact on ants search, while the parameter value of pheromone information is increased during the run of ant algorithm. And the elitist solution is discarded if it cannot be improved from generation to generation. By cooperating several generations of artificial ants, the ants search for the path with a minimum energy consumption cost, and the quality of the solution can be improved for minimizing the energy consumption. Experiments are implemented to demonstrate the performance of the algorithm.

Task scheduling problem is a major issue of distributed system. An improved ant colony algorithm is presented to solve the scheduling problem in heterogeneous distributed system whose complexity is known to be NP-complete in general cases. To speed up the converging rate of the algorithm, elite initial solutions are generated by using an adaptable list heuristic algorithm which is a good tradeoff between the computation complexity and solution quality. A novel representation of pheromone can make effectively use of the task fitness value to accumulate pheromone in ACO (ant colony optimization) algorithm. To improve the self-adaptability of the algorithm, the ACO algorithm and the heuristic rule are combined together to adjust the searching space on the progress of the algorithm. Finally, local and global neighborhood searching are performed on the best solution obtained in iterations. Simulation results show that the performance of the improve ACO algorithm is better on finding optimal or near-optimal solutions than general genetic algorithm.

The green agri-food supply chain network (GASCN) design is critical to reduce the total transportation cost for efficient and effective supply chain management. This paper proposes a new solution based on particle swarm optimization (PSO) to find optimal solution for GASCN problem. PSO adopts transforming operator to modify particles in the population. The novelty of the transforming operator is that it can avoid applying the penalty function so that the diversity of populations is decreased. To show the efficacy of the algorithm, PSO is also tested on three cases. Results show that the proposed algorithm is promising and outperforms GA by both optimization speed and solution quality, especially when the scale of problem is large.

Power load changes periodically. And the effects of climatic (precipitation, relative humidity, temperature, wind speed) on the load should be fuzzy. In order to solve the problem, this chapter presents a method combining wavelet transform, fuzzy set concept, and neural networks for short-term load forecasting. Through the wavelet transform, the load sequence decomposes into subsequences consisting of different wavelet coefficients. On the other side, by the fuzzy neural network, the samples of five meteorological factors influencing power load are transformed into fuzzy input with the subsequences, and then, the suitable RBF neural networks for the forecasting are selected. Finally, the load forecasting sequence is obtained by the reconstruction of the forecasted results from the subsequences. The simulation results demonstrate the proposed method possesses validity and practicability with a mean absolute error below 1.5 %.

The objective of wireless sensor networks is to extract the synoptic structures (spatiotemporal sequence) of the phenomena of ROI (region of interest) in order to make effective predictive and analytical characterizations. Energy limitation is one of the main obstacles to the universal application of wireless sensor networks. Recently, adaptive sampling strategy is regarded as a much promising method for improving energy efficiency. In this paper, we dedicate to investigating how to regulate sampling frequency of sensor nodes in different clusters dynamically following the change of signal frequency. The adaptive frequency-based sampling (FAS) algorithm proposed in this literature is to measure periodic signal frequency online in different clustered region, afterwards regulate signal sampling frequency following with minimal necessary frequency criterion; as a result, the previous desired level of accuracy is achieved, and the energy consumption is decreased. The simulation results are compared with that of fixed sampling rate approach with respect to energy conservation.

To further solve the problem that using the positioning algorithm directly based on received signal strength (RSS) in WiFi technology has lower positioning accuracy because the characteristics of the wireless channel affect the signal attenuation largely and randomly, an indoor three-dimensional positioning algorithm based on difference received signal strength (DRSS) is proposed through the analysis that two close propagation paths have similar interference. It can reduce the large positioning error caused by time-varying interference and directly use the fluctuant values of the interfered received signal. Meanwhile, the wireless signal attenuation model with a parameter of time-varying environment factor is used. A method of real-timely estimating and modifying the parameters by least square estimation (LS) and the way of average is proposed to solve the problem that the model cannot describe the real-time changes of signal attenuation accurately. The environmental test results show that this method not only can obtain a more accurate model but also has higher positioning accuracy in three-dimensional multi-interference environment.

This study investigates the universal approximation capability of three-layer feedforward double flexible approximate identity neural networks in the space of continuous functions with two variables. First, we propose double flexible approximate identity functions, which are a combination of double approximate identity functions and flexible approximate identity functions as investigated in our previous studies. Then, we prove that any continuous function

f

with two variables will converge to itself if it convolves with double flexible approximate identity. Finally, we prove a main theorem by using the obtained results.

Direct use of the hand as an input device is an attractive method for providing natural human–computer interaction (HCI). Computer vision (CV) has the potential to provide more natural, noncontact solutions. As a result, there have been considerable research efforts to use the hand as an input device for HCI in recent years. Hand tracking is the most important procedure for HCI. This chapter presents a novel hand tracking algorithm which can track a hand stable and is real time, and we review on the latter hand tracking research direction, which is a very challenging problem in the context of HCI. Our algorithm is based on mean-shift and we improved it to fit for robust hand tracking by using integrated GIH and skin color mask, our improved algorithm can track hand reliably even in clutter environments. Finally, we demonstrate the benefits of our approach in contrast to existing methods.

The green agricultural products supply chain network (GAP-SCN) design provides an optimal platform for efficient and effective supply chain management. This chapter proposes a new solution based on transforming quantum-inspired genetic algorithm (TQGA) to find optimal solution for the GAP-SCN problem. TQGA adopts transforming representation to convert the Q-bit representation to float-point number, and the float-point number to Q-bit representation, using transforming operator to modify chromosomes. The novelty of the transforming operator, that it can avoid the diversity of populations, is decreased. To show the efficacy of the algorithm, TQGA is tested on two cases. Results show that the proposed algorithm is promising and outperforms the classic GA by both optimization speed and solution quality.

In allusion to the shortage of hardware configuration in the mobile devices and high time-complexity of Dijkstra algorithm, the chapter comes up with a shortest path algorithm based on cut-corner for restricted searching area. This algorithm aims at shrinking the smallest searching area quickly and considers the advantages of Ellipse algorithm and Rectangle algorithm. When tested in the simulator, we find that the time-complexity of Cut-corner algorithm is reduced by 5–20 % compared with that of other conventional algorithms. Thus, it has better effect when used in navigation software of low-end mobile device.

Aimed at the problem of nodes energy imbalance, which is caused by the heavy burden of cluster heads in clustering wireless sensor networks, an uneven clustering routing algorithm based on multihop communication has been proposed for wireless sensor networks. An election algorithm is used for reasonable selection of cluster heads based on candidate threshold and time driven, the independent nodes are introduced to reduce burden of the cluster heads, and the multihop routing based on angle is applied to optimize the intercluster routing algorithm. Simulation results show that the algorithm can save the network energy effectively and balance the energy consumption.

Pseudorandom binary frequency shift keying (2FSK) sine wave signals with the frequency of 1,200 bit/s and 2,200 bit/s are produced by using the rand function in MATLAB. A noise cancel adaptive filter based on least mean square error LMS algorithm is designed and simulated on MATLAB. The relationships between the adaptive parameters (filter taps N and iterations step length μ) and the convergence speed and precision are analyzed. The optimized adaptive parameters which are not only sufficed for filtering performance but also suited for FPGA hardware implementation are found out, that is, filter taps N = 22 and iteration step length μ = 0.006. It may provide a good foundation for FPGA hardware implementation.

A density-sensitive semi-supervised affinity propagation clustering algorithm (DS-SAP) is proposed in this chapter. The DS-SAP uses supervised information of the pairwise constraints for adjusting data points distance matrix. Then we introduce a novelty similarity metric based on the characteristics of global and local data distribution. This metric can effectively reflect the reality of data distribution. The DS-SAP clustering algorithm is based on the frame of the traditional AP algorithm and has extended data processing capacity compared to the traditional AP algorithm. Experimental results show that the new algorithm is outperforming traditional AP clustering algorithm.

In order to improve the efficiency of hybrid particle swarm optimization (PSO) algorithm, a PSO merging simulated annealing and hill climbing (SAHCPSO) is implemented based on a three-level parallel model to increase its convergence speed and to decrease the operation time. SAHCPSO can enhance the diversity of the population and avoid population premature convergence. By analyzing and optimizing the SAHCPSO, we complete the task mapping on the model and make full use of CPU/GPU heterogeneous cluster resources. Optimization for parallel accessing further improves the efficiency of the algorithm. The parallel SAHCPSO implements the coarse-grained parallelism between computation nodes and fine-grained parallelism within each node, greatly reducing the operation time. The experimental results show that with the increase of particle scale, higher speedup can be obtained. The high efficiency of the parallel strategy of the model makes the parallel SAHCPSO more easily to solve large-scale problems.

Intensity modulated radiation therapy (IMRT) is a principal cancer treatment at present, and the optimization of inverse planning is the core to realize the IMRT treatment planning, so it is important to study how to optimize the inverse planning as much as possible. The optimization of inverse planning in IMRT refers to a number of parameters and requires rapid calculation speed clinically, so an improved NNIA for multi-objective is adopted in this paper. At first, according to the IMRT dose constraints of multiple targets, an average dose-based function is used, and then the optimization results compared with the SAGA algorithm under the water phantom show the feasibility and efficiency of the improved NNIA algorithm.

In this work, the proposed orchestrating and sharing system for online paper aims at managing papers from information collecting, paper editing, paper type-setting, and paper submitting to paper sharing. In the five aspects above, there are many available tools which help science researchers write papers, but these tools work separately not cooperatively. Orchestrating and sharing system for online paper integrates functions of these tools, which offers one-stop service. As an important part of this system, the recommendation for paper submission is to provide valuable information about the latest international conferences and journal for paper publication. When papers are written, our system, a context-aware solution for paper, automatically obtains the keywords from context. Given that the recommendation for paper submission is subject-oriented search, we design a recommendation system for paper submission based on vertical search engine, which enhances the search accuracy by the improved URL-based filtering algorithm and the improved content-based filtering algorithm.

With the rapid development of computers and networks, the growth of data causes the data mining increasingly difficult. To solve this problem, this paper proposes an improved SPRINT algorithm based on the Hadoop platform. By analyzing the traditional SPRINT algorithm, we improve it in three aspects: eliminate unnecessary and repetitive calculations in the processing of discrete attributes; none presort of continuous attributes and split by line directly when splitting; and add the node field for attributes list in the data structure. For illustration, a performance test of acceleration and accuracy is executed to prove the effectiveness of the improved SPRINT algorithm. Compared to the original SPRINT algorithm, experimental result shows that the improved SPRINT algorithm guarantees the accuracy and reduces the computing time for the best split point thus accelerates the speed of decision-tree construction.

In order to solve the problem of prediction for long-term web sentiment, a prediction model is built using the proposed method in this paper. First, a novel clustering method based on the extension neural network (ENN) is introduced to recognize the types of subclass of web sentiment. For each class of social events, the class model library of the development trend of web sentiment is established by cycle analysis and ENN clustering combined with nonparametric auto-regression analysis (NAR) method. Then the adaptive transformation is applied to the already known development trend of a new social event, and the min-sum of mean square error (MSE) from the library is selected to predict the future development trend of web sentiment. Empirical findings indicated that compared with the traditional methods, such as the GM (1,1) and least squares estimation (LS) method, the approach presented in this paper yields a higher correlation value in predicting the long-term development trend of web sentiment and can predict the turning points of the development trend more effectively. The ENN- and NAR-based prediction model can effectively solve the problem of prediction for long-term web sentiment.

Dynamic route guidance system is an important part of the intelligent transportation system; the core part of which is optimal path algorithm. This paper has analyzed the main influencing factors on the choice of optimal path, then provided an improved K-optimal chaos ant colony algorithm (K-CACA). The road impedance factor in K-CACA is based on the length, crowdedness, condition, and traffic load of the road sections. The optimizing procedure of the algorithm is speeded up by introducing the included angle threshold of direction. The chaos perturbation effectively refrains the algorithm from trapping into local optima. The results of simulation experiment show that K-CACA is effective and has much higher capacity of global optimization than Dijkstra algorithm and basic ant colony algorithm for optimal route choice.

This paper proposes using a certainty-based active learning framework for extractive meeting speech summarization in order to reduce human effort in generating reference summaries. Active learning chooses a selective set of samples to be labeled by annotators. A combination of informativeness and representativeness criteria for sample selection is proposed. The results of summarizing parliamentary meeting speech show that the amount of labeled data needed for a given summarization accuracy can be reduced by more than 40 % compared to random sampling. The certainty-based active learning framework can effectively reduce the need of labeling samples for training. Furthermore, compared with lecture speech summarization task, the experiments show that the proposed active learning method of meeting speech summarization is obviously more affected by choice of different kinds of classifiers.

For the problems existing in the identification process of piano tone, this paper puts forward an MFCC-based (Mel Frequency Cepstrum Coefficient) feature extraction algorithm and a new piano tone identification method with BP neural network as the matching model. Using the MFCC feature extraction algorithm to extract parameters is a good alternative, which could improve the identification rate. Regarding the improved BP neural network as the matching model of tone identification consumes moderate training time and owns high recognition rate. Simulation results show that the piano tone identification combining the BP neural network with the MFCC algorithm is simple, fast, and highly accurate.

We study the problem of implicit factoring presented by May and Ritzenhofen in 2009 and apply it to more general settings, where prime factors of both integers are only known by implicit information of middle discrete bits. Consider two integers

N

1

=

p

1

q

1

and

N

2

=

p

2

q

2

where

p

1

,

p

2

,

q

1

, and

q

2

are primes and

q

1

,

q

2

≈

N

α

. In the case of

t

log

2

N

bits shared in one consecutive middle block, we describe a novel lattice-based method that leads to the factorization of two integers in polynomial time as soon as

t

> 4

α

. Moreover, we use much lower lattice dimensions and obtain a great speedup. Subsequently, we heuristically generalize the method to an arbitrary number

n

of shared blocks. The experimental results show that the constructed lattices work well in practical attacks.

HBase is a top Apache open-source project that separated from Hadoop. As it has most of the features of Google’s BigTable system and is implemented in Java, it is very popular in days of massive data. HBase’s advantages are reflected in the massive data read and query. Loading huge amounts of data into HBase is the first step to use HBase. HBase itself has several methods to load data, and different methods have different application scenarios. This article made an exhaustive study and a performance testing of them. Also, this article achieved the custom loading data, and experiments show that it has good efficiency.

Recently, the decision tables used in decision-theoretic rough sets are the most complete decision tables and less for incomplete decision tables. Some authors use the filling method to deal with incomplete decision table of DTRS, which is filling the unknown values with all possible values. Its disadvantages are large calculated amount and noise. Therefore, incomplete decision-theoretic rough set model based on improved complete tolerance relation is proposed. First, improved complete tolerance relation and tolerance class were constructed. Then, the decision degree between object and target concept were also computed. Finally, the decision degree was served as conditional probability; the probability positive region, negative region, and boundary region were computed, and it would make three-way decision. The example shows that this model has less calculated amount and noise, and it more accords with practical application.

A new association rule mining algorithm based on matrix is introduced. It mainly compresses the transaction matrix efficiently by integrating various strategies. The new algorithm optimizes the known association rule mining algorithms based on matrix given by some researchers in recent years, which greatly reduces the temporal complexity and spatial complexity, and highly promotes the efficiency of association rule mining. It is especially feasible when the degree of the frequent itemset is high.

Several conference proceedings have been infiltrated by fake submissions generated by the SCIgen computer program. Due to the fictional content the chapter “Decoupling Interrupts from the Internet in Markov Models” by “Jinwen Ma, Jingchun Zhang, and Jinrong Guo” has been retracted by the publisher. Measures are being taken to avoid similar breaches in the future.

Feature selection is an important research topic in machine learning and pattern recognition. It is effective in reducing dimensionality, removing irrelevant data, increasing learning accuracy, and improving result comprehensibility. However, in recent years, data has become increasingly larger in both number of instances and number of features in many applications. Classical feature selection method is out of work in processing large-scale dataset because of expensive computational cost. For improving computational speed, parallel feature selection is taken as the efficient method. MapReduce is an efficient distributional computing model to process large-scale data mining problems. In this paper, a parallel feature selection method based on MapReduce model is proposed. Large-scale dataset is partitioned into sub-datasets. Feature selection is operated on each computational node. Selected feature variables are combined into one feature vector in Reduce job. The parallel feature selection method is scalable. The efficiency of the method is illustrated through example analysis.

In order to do formal verification of interlocking system, which is complicated but safety critical, we choose formal specification language Maude for modeling and verification based on membership equational logic and rewriting logic. In this chapter, a method is proposed to show how the initial state can be modeled and contains important information of specific interlocking system. And a case of Tongji Test Line is reported to illustrate this method in detail. The verification results show that Maude can be applied to formal object-oriented specification and model checking of railway interlocking system successfully using the proposed modeling method.

In order to reduce redundant information in data classification and improve classification accuracy, a novel approach based on nonnegative matrix factorization and harmonic functions (NMF–HF) is proposed for semi-supervised learning. Firstly, we extract the feature data from the original data by nonnegative matrix factorization (NMF) and then classify the original data by harmonic functions (HF) on the basis of the feature data. Empirical results show that NMF–HF can effectively reduce the redundant information and improve the classification accuracy compared with some state-of-the-art approaches.

Interactions among operating system, drivers, and peripheral devices are important for users to understand data communication at low system level, system architecture, and hardware programming. In this chapter, we study low-level data communication and resource management by conducting the development of a USB device driver. A reverse engineering approach has been adopted in this study, and we focus on exploring the USB protocol and developing a device driver for the Linux operating system. We have performed various experiments to evaluate the device driver from different aspects, and all testing results are remarkably good. We believe this work can provide users a clear practical understanding of data communication from the hardware level to user space applications as well as theoretical foundations to reproduce any unsupported peripheral hardware devices.

Tuberculosis is a chronic infectious disease which can make serious hazard to human health and cause large social and economic burden on a country. So for experts and researchers, tuberculosis is one of the biggest public heath challenges. The cause of this disease can be effectively studied by precise analysis of the spatial distribution of the disease. This chapter demonstrates that using existing health data, spatial analysis and GIS in conjunction with epidemiological analysis can identify tuberculosis transmission. This chapter also demonstrates some of the valuable results of GIS in disease surveillance and mapping. The decision-makers could master the epidemic of tuberculosis dynamically and then take better measures to control tuberculosis. Moreover, this study may add some value to traditional and molecular epidemiology and provides an alternative method that may give insight into the transmission of tuberculosis.

Traditional control flow graph of exception handling lacks an explicit description of exception handling and propagation and cannot be used to well analyze the exception situations and exception handling error. To solve these problems, this chapter presents a construction method of exception control flow graph (ECFG) for BPEL process. This method uses a label that is marked exception and is of power for collection computing to describe exception information of BPEL process in building the ECFG. Moreover, the experiment shows that the ECFG generated can clearly express exception information and propagation process in BPEL process.

Brain–computer interface systems are communicative mediums between human brain and external device. One of the applications of these systems is P300 speller. This application provides the ability to spell the characters on the screen for disabled people. In this study, we review the character recognition and its relation to P300 detection. Then, we used three neural networks models with flexible activation functions to detect P300 patterns from electroencephalographic signals more accurately. The obtained results have shown the accuracy of the character recognition based on the precision and recall measures.

In this information era, we are facing the knowledge explosion, and the information on the Internet is multifarious. It is not convenient enough for us to access to information directly on cell phones due to their limitation. Based on parsing a web page with regarding it as a DOM (document object model) tree, we extract the valuable information with considering three factors: structure, content, and programming habits. For illustration, 28 websites are utilized to show the feasibility of the method in web information extraction, and we design the mobile client to present the web content on the cell phones. The practice has proved that using the web page extraction technology related to this article to browse the corresponding news websites only consumed 8 % of cell phone traffic of the existing mobile phone browser. And the user experience is improved. This method can help people to get rid of costing too much on the cell phone traffic, redundant information, complicated operations, and so on.

The explosive increase of scientific data brings in the “Fourth Paradigm” research method by Jim Gray. In order to accelerate the processing speed for these big data, parallel distributed processing is needed. As the data-intensive computing requires high throughput of IO, the data transfer from different node should be cut down as much as possible. Current technologies focus more on the framework for local reliable network with homogeneous resources, but the parallel processing framework for scientific data-intensive problems such as spatial data shared with the Internet and queried by semantics is not fully studied. In this article, we proposed a new data-intensive parallel processing framework for spatial data—Robinia DSSSD (Distributed Storage and Service for Spatial Data), which provides the flexible ability to support data distribution and allocation across the Internet, and semantics query. Experiments shows that Robinia DSSSD can achieve good acceleration with low overhead, and it can well support data-intensive computing.

GUI (graphical user interface) is becoming increasingly important in the software field for the reason that it is a friendly way for the users to interact with the software through GUI. Testing in GUI, however, is faced with many challenges, due to the immense number of event interactions. Testing all possible event interactions is impossible, since the number of required test case is huge in numbers. GUI testing mainly serves two goals: First, to establish confidence in assessment of GUI; Second, to find that more software defects in GUI testing while limiting the number of test cases. For this purpose, any testing method must be better at detecting defects. This article proposed a new technique that can be used for GUI testing, which can guide the GUI testing and find more defects as soon as possible. In this chapter, it introduces an approach of GUI testing guided by Bayesian model optimization scheme, discusses the Bayesian model topology and its issues encountered in the modeling process. It presents solutions in connection with the parameters problem. In the end, a simple case verifies the validity of the model during the GUI testing.

Reverse distribution has received growing attention throughout this decade. Built on the concept of green supply chain management (GSCM), this chapter presents a mathematical programming and distribution model for reverse logistics with collection sites. Due to the complexity of the GSCM model, a heuristic solution is given and improved. The solution adds a heuristic concentration procedure, where subproblems with reduced sets of decision variables are iteratively solved to improve the optimality, and it improves the capacitated plant location problem (CPLP). Computational test demonstrates that high-quality solution is obtained with the improved model.

With the large-scale development of wind power, the storage and analysis of wind turbine’s data gradually become more and more important, there are huge amount of information about wind turbines, and the traditional relational database has been difficult to meet the demand of mass data storage and analysis. This chapter proposes the solutions that data storage is based on non-relational databases (MongoDB) and compares SQL Server with MongoDB about the storage ability and query performance. The results show that using this method can increase storage speed and query performance significantly.

This study proposes a method using mean comparison to improve the accuracy of interferometry processing correlation time delay under the low signal-to-noise ratio and low residual time delay. The method uses the means of taking the average of stripe subsection, comparing threshold and eliminating outliers, which offsets the influence of channel noise on the accuracy of the signal. Using the direct method under the same SNR strike delay simulation comparison proved that the mean comparison method can get relatively high accuracy of time delay information.

The mass data storage systems need to be coupled with efficient parallel/distributed file systems, such as Lustre and HDFS, which can effectively solve the problems of the mass data storage and I/O bottlenecks. This chapter systematically studies the performance factors and distribution of parallel/distributed file systems and proposes a valuation scheme for the classic parallel/distributed file system by capturing the changes in workload characteristics. The experiment results show that the proposed evaluation scheme can reach better accuracy and efficiency.

As critical infrastructures in the Internet, data centers play an important role in supporting large-scale distributed applications as well as data-intensive computing. This chapter presents CoCell, a server-centric architecture, which uses servers to relay packets. CoCell has several nice properties for desired data center networking. The average node degree in CoCell network is close to 3, and the longest routing path length is no larger than 7. Besides, CoCell network is able to provide high network capacity to support bandwidth-intensive applications. Leveraging the multi-paths between any pairs of servers in CoCell network, we propose relative routing schemes and make a comparison among these paths. The evaluation indicates that CoCell performs well in all-to-all traffic pattern.

Simulation test has provided a favorable method and platform for quantitatively evaluating impact on countering ARM by overcoming the disadvantage of high price and poor privacy in regard to outfield experiment. The essay tries to make a deep research on modeling simulation of active decoy and ARM and thus to formulate active decoy interference model and anti-radiation missile movement model. Then it carries through simulation process of active decoy’s effect; the simulation result validates the model’s effectiveness and accuracy, and therefore it has provided theoretical bases for designing and deploying active decoy and ARM’s base station program.

Binary phase-coded signal is usually used in pulse compression (PC) radar, which is mostly used for surveilling and tracking targets. Two prominent characteristics of the binary phase-coded signal are introduced in this article; according to these characteristics, some jamming forms are analyzed which can be used to jam binary phase-coded PC radar; simulation with MATLAB is done to test these jamming forms. The result indicates that noise jamming has less effectiveness on binary phase-coded PC radar and continuous and partial code replicated jamming have preferable effectiveness on it. Because binary phase-coded signal is compressed to narrow pulse when it is received by receiver, jamming signal can easily capture the range gate when the jamming side adopts range-gate pull-off (RGPO), and so the jamming effectiveness is perfect.

To solve the recommendation system, the prevalence of blindness, and low resistance, in this paper, an in-depth study on the personalized automatic recommendation system user model and automatic recommend technology. The paper first introduces the automatic recommendation system user model representation and update. Then the user modeling, clustering, classification, and automatic recommendation technology are combined to develop the automatic personalized document recommendation system based on clustering and classification. First, offline, the system form clustering points of interest to the article and build user model based on clustering interest points. Then realize the automatic recommendation online by recommendation algorithm which is based on classification. Theoretical analysis and experimental results show that the system can significantly improve the online response speed and efficiency.

Subspace-based palmprint recognition methods, such as principal components analysis and linear discriminant analysis, assume that each class can be grouped in a single cluster. However, this assumption is not reasonable at some situations where a class is assembled in two or more clusters. In order to solve this problem, a novel palmprint recognition method based on subclass discriminant analysis is proposed in this chapter. Each palmprint class is divided into a set of subclasses that can be separated easily in the new subspace representation. After that, the Euclidean distance and nearest neighbor method are employed as the similarity measurement. Experimental results conducted on a database of 86 hands (10 impressions per hand) show that the equal error rate (EER) of the proposed method yields promising result of EER = 0.67 % for verification rate, which demonstrates that the proposed method is effective to solve the problem mentioned above.

In order to evaluate automatic aircraft docking quality, a new method of process quality monitoring of the automatic aircraft component docking is proposed. This method is based on laser tracker measurement. The position of automatic aircraft docking component is obtained by laser tracker measurement for a plurality of feature points. By doing identification labels on the surface of docking components, component surface images replace complex docking images. Image processing technology is used to extract the features of component surface image information to evaluate docking quality so as to control the perfect match of docking component. The new method enforces automatic docking process visual monitoring and absorbs human visual and reliable monitoring advantages of manual assembly model.

Sag is one of the important parameters for operation and maintenance of the transmission lines, and its size directly affects the safe and stable operation of the line. In recent years, in order to improve the transmission capacity, many existing transmission lines allow the temperature from 70 °C to 80 °C, and then the transmission line sag becomes a major constraint for the transmission security. This chapter presents a novel sag measurement method based on image processing. Firstly it grays the collected color images and preprocesses the images with some image-denoizing methods. Secondly, special points generated by the isolation rod are extracted by the corner extraction algorithm, and the spatial coordinate values of the extracted points are identified according to the principle of binocular vision and the relationship of three-dimensional coordinate space coordinates and image coordinates. Finally via the method of the curve fitting, the actual sag of the transmission line is calculated. The experimental results show that this method is suitable for both the cases in which the height of the transmission line is equal or not, and it has good adaptability.

The ontology construction process is very expensive and time consuming when performed manually. In order to solve the problems of time consumption and high cost, a Chinese domain ontology learning method based on semantic dependency and formal concept analysis is proposed in this chapter. During the learning process, semantic dependency analysis technology is used for extracting formal context from unstructured domain texts, and Godin algorithm is used for constructing concept lattice. At last, the chapter takes a medical ontology construction as an example to verify this method. The experiment results show that the method we proposed can construct domain ontology automatically and reduce manual intervention. In addition, the ontology we got is a formal ontology, so it has more advantages in sharing and reusability.

In text classification community, k-nearest neighbor (kNN) and support vector machine (SVM) are all effective classifiers, but both of them have their own drawbacks. kNN involves high cost to classify a new document when training set is large; SVM is too sensitive to the noisy data when the noisy data is close to the hyperplane it suffers. So one hybrid algorithm based on variable precision rough set is proposed. It combines the strength of both KNN and SVM techniques and overcomes their weaknesses. Finally some experiments are carried out to compare the efficiency and classification accuracy with different classification algorithms. Results show that the proposed method achieves significant performance improvement.

In order to solve the variation of target appearance and background influence to the visual tracking, we extend the robust fragment-based tracker to an adaptive tracker by selecting features with an online feature ranking mechanism, and the target model is updated according to the similarity between the initial and current models, which makes the tracker more robust. What is more, we reposition the integral histogram’s bin’s structure and that makes our tracker quicker. The proposed algorithm has been compared with fragment-based tracker, and the results proved that our method provides better performance.

Gait features obtained by current extraction methods are easily affected by people’s walking direction, dresses, and carryings, due to which gait recognition system has not yet appeared. An extraction method based on centroid is proposed in this chapter. Segment and track the moving silhouettes of a walking figure in image sequences to calculate the silhouettes’ centroid. The complex silhouette is represented by a point to avoid the influence of dresses and carryings. Divide centroid coordinate value by the height of detecting walking figure to normalize to remove the disturbance caused by walking direction relative to the camera optical axis angle. By denoizing centroid trajectory remove the noise caused by some accidental factors to obtain regular wavelet curve whose main frequency component distribution vector is the final gait feature. Experimental results show that this approach can obtain identical gait features even when experimenters change their walking directions, dresses, or carryings, tolerating noise and low resolution.

In order to solve the problems that the intelligence algorithm falls into the local optimum easily and has a slow convergence in Bayesian networks (BN) structure learning, an algorithm based on adaptive selection operator with simulated annealing is proposed. This chapter conducts the adaptive selection rule in combination with conditional independence tests of BN nodes to guide the generation of neighbor. In order to better compare the adaptive effect, an algorithm based on selection operator with simulated annealing (SOSA) is proposed; at the same time 15 data sets in the three typical networks are accessed as learning samples. The results of the Bayesian Dirichlet (BD) score, Hamming distance (HD), and evolution time of the network after learning show that it has the quicker convergence and it searches the optimal solution more easily compared with simulated annealing (SA) and SOSA.

This chapter proposes a polar space-based method to segment the static image automatically. The proposed method aims at segmenting the object of interest by finding the optimal closed contour in the polar space, solving the long-term problem of scale in the Cartesian space. Experimental results further verify and demonstrate the efficacy of the proposed polar space-based method on the challenging datasets.

Image restoration technology can be applied in a lot of fields including image communication, image archive restoration, and image editing. In this chapter, we try to solve image restoration with a nonlocal regularization point of view. Similarity between different image pixels is measured by a nonlocal

p

-Laplace operator. We use minimum least square with a regularization term to formulate the whole procedure of image restoration. In the solvent of cost function, a linear Gauss–Jacobi iterative method is utilized for unknown pixel solvent. The complexity of iterative solvent is controlled by a step threshold. Finally, experimental results highlight our superior performance over previous methods from subject visual perception or object image quality assessment.

Based on “green building,” we use computer to make model analysis regarding building actual environment, adapting natural light, and using artificial lighting rightly to achieve green energy-saving building goal. According to living example making, we adopt 3d max and ECOTECT model analysis software to analyze and simulate building room natural light, artificial lighting and distribution. Analysis states: at the stage of the architectural sketch, we use 3d max modeling, uniting inter-room facility draft and analysis real environment data by ECORECT, getting the natural light effect, and uniting the two sides to get a more accurate room light environment data. Through this supporting data, we can get better daylight design, not only to satisfy the in-room light requirement but also to make it more beautiful and energy saving.

In order to achieve the better online gesture recognition rate, a multi-sensor fusion method is proposed in this chapter. After the dimension reduction and quantization, we first measure the performance of every single sensor in training phase and use this prior knowledge to determine the weight vector; then we do the fusion of multiple sensors according to the weight vector which indicates each sensor’s importance in recognition. The core algorithm we use for online gesture recognition is WarpingLCSS, which is demonstrated to be an efficient template matching method for gesture spotting. We do the experiments on the OPPORTUNITY Activity Recognition Datasets, and the results show that the recognition rate of multi-sensor fusion method achieves 61 %, which outperforms the single sensor’s performance about 11 %. This demonstrates that our proposed multi-sensor fusion method is efficient in improving the performance of online gesture recognition.

In order to enhance image contrast and ensure accurate identifying and tracking in intense illumination case, this chapter uses the algorithm of histogram cone-shaped, which can enhance image contrast effectively. With the algorithm of histogram cone-shaped, the scope of the gray value increases obviously. And the chapter introduces a first-order differential operator two-direction Prewitt operator to enhance the edge for image; the enhance effect is favorable, and the compute time is short. Then the algorithm of 2-D gray histogram is used to segment image. The chapter uses Hough transformation to identify the lane mark’s two edges and account its intercept and slope and then draws the midline as the last identifying result. In order to reduce the count time, the chapter uses the algorithm of area of interesting to track the lane mark. The experiment results show that the lane mark can be tracked dependably in intense illumination and the algorithms are of real time; moreover, when the tracking algorithm is a failure, the system can also recover in time and lock the tracking target accurately again.

A framework for multi-featured classification modeling of remote sensing images with sparse representations is proposed in this chapter. The problem for extracting features to build sparse learning optimal dictionaries is solved by using the remote sensing image spectral values combined with their transformation’s characteristics such as Normalized Difference Vegetation Index and K–T transformation. This framework employs sparse representation on dimensional reduction and feature refinement for the remote sensing images. Experiments show that by using our approach the classification accuracy and Kappa coefficient are greater than the support vector machine and conventional sparse representation methods. This result is based on remote sensing images from the sand lake in Pingluo County, which is located in the Ningxia Hui Autonomous Region.

In order to improve the performance of the traditional support vector machine (SVM), this chapter proposes one method referred as MR-SVM to parallelize SVM on MapReduce and mitigates the convergence problems brought by data partitioning and distributed computation. By splitting the large dataset and concurrently calculating the support vector set of each chunk across map units, MR-SVM improves the process capability and efficiency. Then the partial support vector sets are combined as the training set of the global training in reduce phase, and the current global optimum solved by reducing operations is fed back to each map units to determine whether MR-SVM should proceed with another pass. This process iterates until MR-SVM converges to the global optimum. In theory, it has been proved that MR-SVM converges to the global optimum within finite iteration size. Experimental results show that MR-SVM can improve the data processing capability and efficiency of the traditional counterpart and guarantee its high accuracy.

In order to solve the problem that the mature vehicle classification cannot meet the requirements of accuracy and speed concurrently, this chapter chooses the contour features and speeded-up robust features (SURF) features of vehicles and then adopts a hierarchical support vector machine (SVM) classifier for vehicle classification. At first, the system uses the contour features which are simple and fast in the first layer of classifier so that it will filter out the easy samples. Second, the system utilizes the rich information and stable SURF feature in the next layer of classifier. We conducted extensive experiments against a number of baseline methods; the accuracy of proposed method was increased by about 20%, and the time was shortened by 2/3, significantly outperforming the baselines. The method of double features and hierarchical SVM has a good trade-off between speed and precision.

Image splicing is a common method to construct forged image which decreases the authentication of the traditional image. Resizing operation is usually necessary to create a convinced forged image. Though the forged image leaves no visual clues, resizing operation using interpolation method destroys the relationship between neighboring pixels, thus leaving traces which can be captured by statistical feature. We first convert the traces left by resizing to feature and then feed features from enough sample images to support vector machines to train for detector. Finally, we use detector to determine whether the image is tampered and point out which parts of the image are tampered by block-wise method. Experimental results verify the effectiveness of our proposed method.

In order to improve the problem of recognition rate and inaccurate in the curve, this paper proposed a lane detection algorithm based on hyperbola model, which uses Canny operator to detect the edge of the lane and wields the Hough transform to extract lane boundary points, and utilizes extended Kalman filter to reduce road scanning range. By fitting points on pair road boundaries into the hyperbola model, and completes the lane boundary reconstruction. Some experimental studies are conducted, and the results show that the accuracy of the algorithm has reached 93.4 % and the processing speed of each image needs 77.4 ms. Our method is able to make full use of lane boundaries with existence partial occlusion, blur and low contrast. Meanwhile, it can quickly and accurately identify lane line, and it has high performance and robustness.

This paper presents some results from an experiment of image softproofing under different International Color Consortium (ICC) rendering intents, which will be available for selecting an appropriate rendering intent in printing processes. In a screen softproofing procedure, image is converted via ICC profile embedded in image itself to output device profile, then to proofing device profile. With the rendering intent of absolute colorimetry in the second conversion, the effects of different rendering intents in the first conversion have been investigated through a softproofing software developed in Matlab 7.0. A variety of testing images, including light tone, shadow detail, etc., are used for image softproofing. The color differences between original images and proofed images are calculated under the S-CIE L*a*b* color difference formulae. Comparisons and analyses on the obtained images under four rendering intents show the effects of color characteristics of original images on rendering intent selection.

Traditional dense matching algorithms for face based on region growing have a lot of flaws. To generate a better disparity map, a novel improved method is proposed in this paper. Firstly, scale invariant feature (SIFT) algorithm is adopted to detect feature points for a pair of images, which are taken from two different angles. Secondly, this paper uses normalized cross correlation (NCC) to get match points and uses random sampling consensus (RANSAC) algorithm to eliminate mismatches. Several robust seeds are generated after this step. At last, by using an improved strategy of region growing, in which seeds are evaluated to help determine the locations and sizes of the search windows dynamically, the matching relations of seeds propagate to other parts of images. Experiments show that this method can obtain a good disparity map and has high computation speed.

Short texts have played an important role in the field of text data mining. Because of the problems arousing from the complexity of Chinese semantics and data sparseness, which is an obvious characteristic of short texts, it is necessary to explore some new semantic-based methods to cluster Chinese short texts. An improved approach of feature selection based on HowNet is applied in this paper to address data sparseness of Chinese short texts. By redefining Vector Space Model in semantic level and merging generalized synonymy features, we present a new feature generation strategy. Experimental results show that by merging semantic similar feature, our method is effective in feature dimension reduction and gets better clustering performance. The proposed HowNet-based feature selection method is suitable for Chinese short texts clustering.

This paper proposes a novel Internet worm detection and classification method. The behaviors of worms are different from each other’s, and they are also different in terms of the normal Internet activities. So we can detect and classify worms by the extracted features of the network packets. At first, we sniff raw network packets from the local area network (LAN), and extract 13 features from the packet header, and then select 10 important features using the information gain algorithm. With the labeled features, we train Support Vector Machine (SVM) classifiers. The classifiers can classify the behaviors of the worm apart from the normal Internet activities. And this approach can also classify network attacks and Internet worms, although the network attacks and the Internet worms have similar behaviors. In the experiments, this approach performs well in worm detection and classification.

Fall detection is an important problem in the research of abnormal behavior recognition. In this chapter, a novel real-time method is proposed to detect human fall with a single uncalibrated camera by the changes of global orientation and human shape. Our algorithm has three basic parts: moving object extraction, fall pre-detection, and fall confirmation. The overall orientations are derived from the combination of Gaussian mixture model and motion history image. The shape deformation is quantified from the silhouettes by an approximated bounding box. Standard deviations of the overall orientation and aspect ratio of bounding box are checked to pre-detect a fall, and a fall is confirmed by unmoving shape of the bounding box with the defined fall angle. Experimental results show that our method can detect all possible types of human fall accurately and successfully.

This chapter introduces a new method to classify the SAR oil spill images. That is Deep Belief Network (DBN). Through the experimental certification, it is shown that the SAR images’ information extracted by Gray-Level Co-occurrence Matrix (GLCM) can have a better effect in classification then that extracted by Gabor wavelet features. And using DBN to classify 240 samples including oil slick, looks-like oil slick and seawater, we can reach high total classification accuracy up to 91.25 %. Finally, we get a result that the method of DBN with GLCM features can better meet the needs of the SAR oil spill images classification.

Spectral information of ground objects target in remote sensing image is complex, more noise, and highly nonlinear. It makes traditional data processing method no longer significant, effective, and efficient. The BP neural network classification-recognition method provides a more ideal solution. Using the TM remote sensing images as the example, this paper experimented the application of the BP neural network to the remote sensing image classification and recognition. Results showed that the classification precision of cultivated land was very low for both the BP neural network and traditional maximum likelihood methods because the spectrum difference between the new cultivated land and the bare land having low plant covered in this area was not significant. Maximum likelihood method wrongly regarded the bare land which had higher soil moisture content by lakeshore as water body. Except the grassland, the classification effect of the BP neural network was superior to maximum likelihood method. The overall classification accuracy by the BP neural network reached 81.79 %; however, the one by the maximum likelihood method was 79.08 %, indicating that the BP neural network classification and recognition was superior to the traditional maximum likelihood method.

In view of the problem existing in abusive using of image copy-move forgeries, this paper proposes an image forensics algorithm for detecting copy-move forgery based on improved PCA-SIFT. The present method works first by extracting features of an image and then reducing its dimensionality, and the method uses k-nearest neighbor to operate forgery detection. Owing to the similarity between pasted region and copied region, the descriptors are then matched between each other to seek for any possible forgery in images. Extensive experimental results are presented to confirm that the algorithm is able to precisely individuate the tampered image and quantify its robustness and sensitivity to image post-processing and offer a considerable improvement in time efficiency.

In order to solve the letter skeleton problem with handwriting-like attributes, a thinning model is used in this paper. By introducing improved reservation and eliminating produces, additional pixels are constrained by thinning nearest pixels. In the experiment, the proposed method is compared with others in the literature English letters by using the one pass thinning algorithm (OPTA) and Hilditch methods. Empirical results show that the proposed model can thin handwriting-like skeleton in terms of reserving topology and eliminating extra pixels.

In allusion to the urgency of the white wine identification, and the key shortcoming of PCA (Principal Component Analysis), whose all loadings are non-zero, the sparse PCA (SPCA) is employed to enhance the explanatory and remove unnecessary variables. By using elastic net Zou Presented, SPCA can seek sparse factors and explain the maximum amount of variances in the data. For illustration, a comparison of strategy between PCA and SPCA, which combined with the specified categorizer—Support Vector Machine (SVM) and Infrared (IR) Spectroscopy, is utilized. The finally classified result of white wines based on SPCA is up to 93 %, but the PCA’s 83 %, and directed categorizer’s 80 %. Obviously, the SPCA can extract characteristics more effectively, which benefits the classification accuracy of SVM.

Multi-bank mobile electronic payment system uses mobile terminals for electronic payments, which can circulate in multiple banks and cannot limit from the bank that issues the e-cash. The paper researches electronic payment on the withdrawal agreement, the pay agreement, the deposit agreement, the update protocol of the e-cash based on elliptic curve cryptography. The design of the system is more suitable for mobile payment terminals with limit of calculation capacity, storage, network bandwidth, and power supply, which meets the needs of the day-to-day transactions.

It is well known that power saving is one of the most important issues for mobile device in accessing network services. The efficient conservation of energy for longer operation times of a mobile station is vital to the success of various mobile applications. This paper proposes a systematic approach to allocate radio resources in Long-Term Evolution Advanced (LTE-A) network by considering the channel condition and QoS requirements, while the power saving is a centric issue during the scheduling process. The proposed scheme includes the selection of component carriers (CC) and the allocation of radio resource to satisfy the QoS demands while minimize the power consumption of user equipment (UE). Additionally, exhaustive simulations were performed to examine the performance of the proposed scheme. Both http and video streaming traffic models were applied during the simulations. The experimental results show that the proposed scheme achieves better performance when compared to the other scheme.

In order to solve the performance problem of the cloud computing environment, a dispatching and management model (JDRMSP), which is based on safe performance interface is proposed in this chapter. By using the performance interface integrated into the safe DPI as the original basic data capture, agent-based job scheduling algorithm as the job dispatching method, and ant colony algorithm resource scheduling strategy as the resource management method, the integrated cloud performance is enhanced. For illustration, a simulation experimental example is utilized to show the effect of the model. From the experimental results we can get the conclusion that The JDRMSP model can analyze the cloud environment performance of various cloud components distribution pattern more accurately, and this is the basis of configuration control of the performance of the entire cloud environment, ultimately achieving the purpose of enhancement of the performance of the cloud. The JDRMSP model can effectively solve the performance data capture accuracy problem and take advantage of the dispatching and management algorithm to optimize the cloud environment.

In the development of general system design approaches, the main concern is whether the approach meets the proposed system’s specifications and the ability of the system to operate for a specified period within those specifications. However, with the expansion in the field of sensitive and complex systems such as e-health monitoring systems, a greater emphasis is placed on the behaviour of the system with the presence of fault (i.e., fault-tolerance). Consequently, when the system is being built, tasks such as fault-tolerance requirements are essential to ensure the quality of the resulting reliable e-health monitoring system. By considering the fault-tolerance requirements as functional requirements in the requirement phase, the completeness of reliability requirements for an e-health system can be developed. This paper proposes a methodology that conceptually studies fault-tolerance in relation to a smart mobile e-health monitoring system. The methodology aims to contribute towards standardising the fault-tolerant requirements of a reliable e-health monitoring system.

For the present situation of shortage in GIS indoor theories and insufficiency in indoor GIS applications, a set of indoor GIS research theories, indoor map cartography specifications, and related technologies closely integrated with fire-fighting industry were proposed in this paper. The indoor map cartography specifications included technological processes of the map cartography and matched data updating mechanism, convenient for fast, accurately, timely producing professional indoor map. The key technologies, such as symbol dynamic drawing, indoor outdoor seamless integration, map updating and path analysis, were preliminary applied in fire-fighting emergency rescue platform, so as to realize functionalities such as indoor and outdoor seamless expression, POI updating periodically, and the best rescue path analysis, and improve the transparent command level of the fire rescue site, and it also may have certain reference value to other emergency rescues.

In order to improve hardware resource utilization, reduce maintenance and management costs, to build a new IT infrastructure platform for the user to provide a stable, efficient access to services. Taking the library of Harbin Institute of Technology, using VMware cloud computing solutions to build private clouds as an example, through the introduction of VMware vSphere to build a virtual architecture data center, integrate various application services, the introduction of VMware View software system provides “cloud + terminal” desktop cloud of office desktop solution. It is illustrated application of virtual technology by example. It can realize the unified management and deployment of hardware resources and the application of the data center and provide applications with high reliability, high availability, and service of mobile office environment. The IT platform can effectively gather or carrying spare computing capacity, corresponding the IT resource and service priority, improve IT management level.

In order to deliver centralized visibility for login activity, reduce identity proliferation and confusion, increase user adoption and security, reduce administrative costs, and support for entire identity management lifecycle in cloud, a new single sign-on solution is proposed in this paper. By introducing OAuth protocol combined with identity federation mechanism and identity mapping, the new single sign-on model can give the cloud user that has succeed through an identity authentication the permission to access other cloud services in a reasonable time period without entering the username and password repeatedly. Empirical results show that the solution will be used as an impactful measure in scenarios where frequent interactions among different cloud services and clouds that result significant impact across multiple security domains. The OAuth-based single sign-on solution can effectively solve the problems of complexity of identity management and cross-domain authentication in cloud environment and thus increased the security and improved the user’s efficiency.

With consideration of the invalid registration flows and load balance problems in hierarchical mobile IPv6 (HMIPv6) networks, a collaborative load control scheme (COLC) for HMIPv6 networks is proposed to reduce registration flows and balance load. In COLC, mobile anchor point (MAP) is allowed to transfer part of its packet delivery load to its neighboring MAPs with lower load, by which the invalid registration flows decrease, and more mobile nodes (MNs) register with their favorite MAPs without capacity expansion. The validities of the scheme in reducing registration flows of HMIPv6 and performing better load balance are examined in the simulations.

Data transmitted by satellite communication system should be encrypted in order to provide confidentiality. A selective content encryption method suitable for satellite communication system is presented in this chapter, the key content information in the compressed stream is extracted and encrypted, and the variable modulus encryption method is proposed to solve the problem of variable length code encryption; thereby, the encrypted stream can be format compliant. This method can improve the efficiency of encryption and achieve fast, secure, and high efficient encryption of satellite communication system. The experimental results prove the effectiveness of our method.

A parking guidance system can help a driver quickly find an available parking space. Most currently available parking guidance systems require wire deployment in installation, thus entailing high installation costs. In this chapter, we discuss the network design of a low-power parking guidance system. We developed a tiered communication architecture including Wireless Sensor Network (WSN), General Packet Radio Service (GPRS) network and Internet to realize wireless parking space availability data transmission, and thus installation complexity can be greatly reduced. In order to reduce the battery replacement frequency of the WSN, we designed a power-minimized Medium Access Control (MAC) protocol. The proposed MAC protocol divides one network working cycle into four dedicated intervals to realize robust network organization and energy-efficient data delivery. Experimental results showed that the proposed MAC protocol can extend the battery lifetime of the WSN to more than ten years. Based on the collected parking space availability data, we built a portable parking guidance terminal to let drivers locate available parking spaces conveniently.

Under the current cloud computing environment, a reasonable and practicable access control strategy is needed, which is a guarantee to protect cloud computing suppliers to provide services and many cloud users access to services. In this paper, based on analysis of many cloud computing safety features, trust management is introduced into the cloud computing service access control, within the domain of a trust-based access control strategy, in domain, presents a trust-based access control policy. Credible value will be given through the comprehensive treatment of the entity, and then AAC (authentication and authorization center) authorizes the appropriate access rights to achieve the control of the monomer in the domain. Combined with the characteristics of the existing cloud computing environment, in multiple management domains, this paper proposes a role mapping, with the role mapping relationship between the domain, which can make the inter-domain access to resources and security shared access between different domains, in order to avoid the problem of permission penetration and privilege escalation, this paper presents the mirror role based on role mapping, ultimately solves the problem.

In this paper, in order to detect the unhealthy status in the cloud system, a Basic Detection Strategy and a Threshold Strategy based on mathematic theory and statistical knowledge is proposed to solve this problem. By introducing unhealthy status percentage parameter

α

, both Basic Detection Strategy and Threshold Strategy are combined to detect and monitor the unhealthy cloud system status. For illustration, an eBay company example is utilized to show the feasibility of Basic Detection Strategy and Threshold Strategy. Empirical results show that Basic Detection Strategy with setting a suitable value to

α

can pinpoint most of unhealthy status in the cloud system, however, for some special unhealthy status, it must adopt the Threshold Strategy to pinpoint. The combination of Basic Detection Strategy and Threshold Strategy can effectively detect and pinpoint the unhealthy status in the cloud system and help staff to improve the performance of cloud system.

In order to reduce the cost of operation training and improve efficiency of examination, the development of simulation training platform has achieved very good results. An intelligent scoring system based on expert system plays the role of the teacher and gives the student a just assessment. It uses the professional theory and practical experience as the evaluation criteria and analyzes the operator’s operation process to realize the automatic scoring through the program algorithm. The application of scoring system evaluates the operation level of students and gives students guiding opinions and error analysis.

Although various performance characteristics of distributed file system have been documented, the potential performance efficiency of distributed file system on virtualized cloud computing infrastructure is not clear. This chapter focuses on the performance of Hadoop Distributed File System (HDFS) on virtualized Hadoop. We construct a virtualized Hadoop platform and perform a series of experiments to investigate the performance of HDFS on the virtualized Hadoop cluster. Experimental results verify the efficiency of distributed file system on virtualized Hadoop to process the mass-intensive application.

In order to locate and remove the faults in the critical systems where the faults occur, this paper proposes a three-layer decision support system for fault diagnosis, in which both static information and dynamic information of the system are used to find out suspicious components. In the process of locating the faults, a bipartite graph is applied to describe the relation between the symptom and the components, on the basis of which a method is proposed to calculate the value of fault evidence of a component. Then, the components whose values are larger are chosen as the result. Meanwhile, the decision support system adjusts the data of the bipartite graph according to the actual situation in order to improve the effectiveness of the diagnosis. The experiment shows that the fault diagnosis process in the decision support system can locate the fault more effectively.

The design and implementation of an SD card controller circuit architecture for multiple-target interface, suitable for communication function extension of existing electronic device for UBICOMP, are presented in this paper. The SD to multiple targets bridge includes an SD memory controller, a ping-pong FIFO, and a target selectable interface, such as UART, SPI, parallel, and NAND Flash IO. The bridge follows SD memory card v2.0 specification so that it is fully flexible in terms of portable device without any special drivers. The ping-pong FIFO increases the throughput of this system, and the availability of UART, SPI, parallel, and NAND flash interfaces provides flexibility for implementation of applications that requires the conversion of data to feed the SD bus. A tidy NAND flash is also implemented in the multiple-target interface for FTL of NAND flash. The new design has been verified and implemented in FPGA. It has also been synthesized and will be taped out through a 0.18 μm CMOS technology. Experiment reveals that the proposed architecture presents superior performance in platform-independent, interface-scalability and integrality compared with existing works.

In order to improve storage efficiency of small files in the cloud storage systems based on HDFS (Hadoop Distributed File System), this paper proposed a merging process approach based on a two-dimensional packing algorithm, called TDPHDFS (two-dimensional packing for HDFS). In it the correlations between file size and arrival time are comprehensively considered to assist the small files to be merged into large ones. The simulation results demonstrate that the storage efficiency of small files is improved, while the stability remains the same, yet less resource is consumed. The TDPHDFS algorithm can effectively reduce the performance penalty in both storage space and memory consuming while managing massive small files.

The influence diagram (ID) is introduced into advertising media selection and delivery strategy making by reducing uncertainty in the process of decision-making. This paper conducts a survey and selects relevant variables including product category, advertising budget, target audience, media selection, authority, and coverage. The topology layer of the ID model is constructed by distinguishing the causal relationship among variables, and the parameter layer is defined through the judgment of conditional probability. Empirical results show that scientific assessments of the various expected utility values in the decision-making program are put by probabilistic reasoning. Based on it, the larger profit can be obtained under a smaller cost with the principle of expected maximization. Therefore, the model does an effective job and provides reference for decision-makers.

For the lack of safety and reliability of the information in the cloud computing environment, in order to create a more flexible and adaptable security mechanism, the combination of cloud computing and credible concept is a major research direction in today’s security. Based on the view mentioned above, this paper strengthens the research of trust computing technology to solve the security issues in cloud and cloud-based trust transfer, on the basis of the practical work of the experts and scholars on the trust transfer technology, and expands the theoretical model of the trust chain. This paper uses the stochastic process algebra and Petri nets as a modeling tool to build two trust chain models, demonstrates the credibility of certain behavioral characteristics of the chain, analyzes several constraints of credible chain, and provides a valuable reference for engineering practice of the credible chain.

Based on the process of corporate value formation and performance system, this chapter extracts key factors indicating E-commerce application level in enterprises and has established a set of E-commerce measurement indicator system. In addition, this chapter uses Delphi method and Analytic hierarchy process to identify the coefficients of various factors. By applying this model to measure the E-commerce application level in 23 heterogeneous enterprises in Chinese domestic market, this study proves that the proposed model can yield a relative accurate measurement of E-commerce application level in enterprises. The results also indicate that there are strong individual differences among different enterprises in China. The E-commerce application level in individual enterprises is affected by corporate strategy, informatization level, E-commerce application performance. and human resources. The nature and the size of enterprises have significant correlation with E-commerce application level. The study also finds that the big-sized enterprises will become stagnant when they develop to a certain level, which is known as a “trap”.

This chapter aims to solve the problem of how to construct and manage digital educational resources effectively. It puts forward a trinity mode based on system architecture, workflow, and technology system. The trinity model consists of “Pre-Stage, Mid-Stage, Post-stage,” “Theory, Practice, Regulation” and “Approach, Tool, Rule.” By combining the trinity mode with case studies, the issues concerning construction and management of digital educational resources are to be analyzed, including topic selection, relationship between quantity and quality, implementation. Over the past year, results have showed that the trinity model could shorten more than 50 % of the development cycle of the project. The model could greatly help improve the construction and management of digital educational resources.

Cloud computing has became a very popular vocabulary in recent years. The combination of cloud computing and GIS (geographic information system) can improve the performance of GIS. By analyzing the technology of cloud computing, this paper introduces the concept of GIS based on cloud computing; based on the current major GIS application development trends, key technologies of cloud GIS are proposed; finally four application modes of cloud GIS are presented. Cloud GIS can improve stability and efficiency services to end users by optimized network resource allocation of underlying data and services.

For space application to improve the reliability of the memory operation, quadruple modular redundant (QMR) architecture is used in all registers of the memory controller. The QMR architecture in this paper can correct one-bit faults, detect two-bit faults, and also tolerate single event transient (SET). By modifying finite state machine (FSM) of the memory controller, when one uncorrectable fault is checked, the memory operation can be terminated in time and return the error information. Compared with triple modular redundancy (TMR), although the area overhead is increased by 47,530.59297 μm

2

, the single event upset (SEU) failure rate is lower by 6 orders of magnitude. Experimental results show that when 1 bit-flip or 2 bit-flips are injected in QMR registers, they can be corrected or detected in time, respectively. Memory controller using QMR architecture increases the area overhead, but the advantage is the higher reliability valuable for safety system.

In order to reduce the unnecessary power waste of computer system, the working principle of earlier Windows power management program and the new face recognition function of Windows 8 are analyzed in this paper. And the conflict between the convenience of use and the effects of energy conservation and environmental protection is given attention to. We put forward a new method based on the detection of frontal face in front of the monitor instead of the events of keyboard or mouse. Experimental results show that the method is a fast and effective one. Particularly, when user is leaving for a moment, this method is better than the work of Windows power management program. The results tell us that this method can save electrical energy about 4.28 % than windows power management program.

Laser sheet bending is a new metal forming process realized by thermal stresses resulted from the irradiation of laser beam scanning. Laser forming is a new type of sheet metal forming process. The sheet metal is formed by asymmetrical thermal stresses. The three-dimensional elastoplastic thermomechanical coupled finite element model of laser bending for Ti-6Al-4V plates was established with nonlinear finite element analysis software ANSYS. The bending properties of sheet metal with different processing parameters were simulated. The results show that the twist rotation deformation of sheet metal can be influenced by laser power, spot diameter, scanning velocity, scanning path curvature, and the distance between scanning path and free end.

The chapter studies voltage transient stability when the wind generator changes its control modes. The chapter studies the influence caused by connection with wind farms based on simulation and makes comparison between different control modes, then gives the conclusion. The chapter takes the real grid model and the result of the study has some means to the relative study.

Interturn short circuit is a common kind of fault in generator. The chapter takes multi-loop theory to analyze the theory of fault on generator stator and puts the math model of the generator. Changes of main parameters are analyzed separately when fault happens. According to the result, the chapter analyzes the influence on the main parameters caused by the fault of generator stator and summarizes the factors of parameter changes.

The traditional edge marking algorithm cannot fill the elongated polygon and a polygon with local points correctly. After doing a lot of research and analysis about polygon fill algorithms, this paper presents a new improved algorithm, which is suitable for hardware implementation, to meet the need for high-quality graphic display in the embedded system. The new algorithm makes full use of the characteristic that the local point or elongated point is accessed repeatedly when it meets local points and elongated points. We can define a measurement variable named FLAG, which is used to mark the boundary point of the polygon. The flag of the present point will add one when it is accessed. This method can conveniently and simply distinguish singular points and elongated points from ordinary points. What’s more, the improved algorithm solves the previously mentioned problems effectively. In the new algorithm, we only use the addition operation so it is easy to be implemented by the hardware.

Cross-platform video transmission is of vital importance in industrial applications. In this paper, we introduce a method for transmitting video from the computer with Windows system to the ARM11 board with embedded Linux system using the Ethernet based on the TCP/IP protocol. The ARM11 board is used as the server to receive video information using its Qt GUI, while the computer on the bank is used as the client that receives video information from the remote-operated underwater vehicle showing with its MFC (Microsoft Foundation Classes) interface and then sends the video information to the handheld controller. The image gained from the computer MFC is JPG format, and after coding, the images are transmitted to the server on the handheld controller continuously. Then the Qt GUI receives the data and decodes the JPG images before displaying them on the screen. The transmission is based on TCP/IP protocol and an image parsing protocol made by us. After testing, the video image can successfully conduct real-time transmission and can meet the industry application.

Recently, energy consumption has emerged as a critical factor in designing storage system. In order to test the energy consumption of InfiniBand switch (IB switch), we establish an energy consumption model for IB switch and formulate the test cases. Using the method, you can obtain the energy consumption of the IB switch scientifically and efficiently. Empirical results illustrate the correctness of the energy consumption and reflect the distribution laws of the energy consumption of the IB switch clearly. The scheme can solve the problem of testing and analyzing the energy consumption of the IB switch efficiently. It has positive practice significance to reduce the cost of storage system.

In order to implement the real-time filtering and tracking of robot signals with high efficiency, a novel real-time filtering method based on neuron filtering mechanism is developed in this paper. By considering the ubiquity of resonance in mammal and combining the mechanism of neural information processing, the derived details and the feasible parameter criterion under minimum error variance condition are given. For illustration, the application on quadruped robot is discussed. The quadruped robot feet speed signals are processed by developed real-time filtering method and Kalman filtering algorithm, respectively, and the computation time of both methods is tested. Experiment results show that the performance of developed real-time filtering method is better than that of Kalman filtering algorithm, not only in filtering and tracking performance but also in filtering speed. The novel real-time filtering method based on neuron filtering mechanism can effectively implement the real-time filtering and tracking with regard to robot signals.

It is a challenging task to integrate multi-view representations, each of which is of high dimension to improve the clustering performance. In this paper, we aim to improve the clustering performance of spectral clustering method when the manifold for high-dimensional data is not well defined in the multiple-view setting. We abstract the discriminative information on each view by spectral embedded clustering which performs well on high-dimensional data without a clear low-dimensional manifold structure. We bootstrap the clusterings of different views using discriminative information from one another. We derive a co-training algorithm to obtain a most informative clustering by iteratively modifying the affinity graph used for one view using the discriminative information from the other views. The approach is based on the assumption that the clustering from one view should agree with the clustering from another view. Comprehensive experiments on four real-world multiple-view high-dimensional datasets are presented to demonstrate the effectiveness of the proposed approach.

In this paper, we examine the merits of hardware/software co-design of a feedback dynamic voltage scaling algorithm and a new processor are capable of executing instructions in the frequency and voltage conversion. We study several energy-aware feedback schemes based on earliest-deadline-first scheduling, dynamic adjustment of the behavior of the system, for different workload characteristics. An infrastructure for investigating several hard real-time dynamic voltage scaling schemes, including our feedback dynamic voltage scaling algorithm, is implemented on an NEC 530 embedded board. System structure and algorithm overhead is evaluated for different dynamic voltage scaling schemes. Feedback dynamic voltage scaling algorithm saves at least more energy frequently than the previous dynamic voltage scaling algorithm, with an additional 18 % energy reduction peak savings.

Based on the analysis of existing problems of interface software development process of industrial control, the importance of building the general interface system based on OPC (Object Linking and Embedding for Process Control) was proposed. The data model was given with database technology. On the basis of the data model, the configurable general interface system based on OPC was implemented. Versatility and configurability is the most important feature of the interface system. By simple modification of configuration information, the interface system will meet the needs of different projects. The application results show that the interface system greatly reduces the development cycle of the related software, improves the reliability and stability of the application system, and reduces costs of system operation and maintenance.

In order to solve intercommunication problem between IPv4 network and IPv6 network more flexibly, this paper has proposed an improved IPv4-IPv6 translator based on embedded system. By using an optimized address mapping regulation, it can support both stateful and stateless translation method. In addition, a lightweight SIP-ALG and Modbus-ALG have been designed to assist the translator to process the datagram, which may take address and domain information at the seven layers of OSI model. The results show the translator can work well between sensor network and Internet, and the mixed use of stateful and stateless method has much less memory usage than stateful method and nearly the same process delay as stateless method.

Traditional collaborative filtering is widely used in social media and e-business, but data sparsity and noise problems have not been solved effectively yet. In this chapter, we propose a novel approach of collaborative filtering based on field authorities, which achieves genre tendency of items by mapping tags to genres and simulates a fine-grained word-of-mouth recommendation mode. We select the nearest neighbors from sets of experienced users as field authorities in different genres and assign weights to genres according to genre tendency. Our method can solve sparsity and noise problems efficiently and has much higher prediction accuracy. Experiments on MovieLens datasets show that the accuracy of our approach is significantly higher than traditional user-based kNN CF approach in both MAE and precision tests.

Impedance analysis is very important for antenna design. In this chapter, the internal impedance of the plasma antenna is analyzed by building the model of high-frequency electromagnetic waves acting with plasma. At the same time, a model of surface current for plasma antenna is developed in accordance with the eigenvalue equation of guided mode, and the radiation resistance of plasma antenna is analyzed according to the method of Poynting vector. From the results, we find that the internal impedance and the radiation resistance of the plasma antenna are affected distinctly by the plasma density and electron-neutral collision frequency. The internal resistance could be reduced, and the radiation resistance would be added efficiently by increasing the plasma density and decreasing the collision frequency.

A 5.8-GHz transceiver front-end applied in dedicated short-range communication (DSRC) systems which is developed in public traffic transportation to improve the safety is fabricated on a chip using TSMC 0.18-μm CMOS process. The proposed prototype includes an asymmetric T/R switch, a current-reused LNA, and a class A power amplifier (PA) on the low-voltage operation in order to minimize the power consumption. Measured results achieve the power gain of 11 dB, the NF of 4.9 dB, the third-order intercept point (IIP3) of −5.4 dBm, and the power consumption of 3.9 mW in the receiving (Rx) mode. On the other hand, the power gain of 12.4 dB, the output 1 dB compression point (OP

−1dB

) of 11.4 dBm, the PAE of 14.7 % at P

−1dB

, the IMD3 of −15.8 dBc at 1 dB compression level, the output power of 2.6 dBm with a 50

Ω

load, and power consumption of 116.3 mW are obtained in the transmitting (Tx) mode. The overall chip area is 1.5 (1.32 × 1.14) mm

2

. This RF CMOS transceiver front-end includes all matching circuits and biasing circuits, and no external components are required.

A 5.8-GHz frequency synthesizer is implemented in TSMC 0.18-μm CMOS process. This paper proposes a dynamic current-matching charge pump linearization technique and uses a current-switching differential Colpitts VCO to lower the phase noise and an averaged varactor circuit to increase the linearity of the VCO tuning range. At the supply voltage of 1.8 V, measured results achieve the locked tuning frequency from 5.55 to 5.94 GHz, corresponding to 6.8 % and the phase noise of −105.83 dBc/Hz at 1 MHz offset frequency from 5.8 GHz. The overall power consumption is 21.6 mW. Including pads, the chip area is 0.729 (0.961 × 0.761) mm

2

.

This chapter studies magnetic resonance based on wireless power transmission (WPT) system for electronic vehicle (EV). In this system, the two resonant coils mounted on the bottom of the vehicle and on the ground were simultaneously analyzed by the method of moments (MoM), an accurate and efficient full-wave electromagnetic analysis method. Then, compared with traditional WPT in ideal circumstance, the different performance of WPT in wireless charging system of EV is studied. Finally, a new design of the WPT integrated with circumstance is proposed, which achieves 90 % energy transmission efficiency at the resonant frequency of 13.56 MHz with the distance between two resonant coils varying within 15–25 cm.

Reconfigurable computing is being widely used in Computation-intensive applications. With the rapid development of applications, we have higher requirements for the computational efficiency of reconfigurable computing. In order to improve the computational efficiency, the array size gradually increased for applications that are more complex. With the upgrade of the array size, the hardware overhead of traditional interconnection structure used for reconfigurable processing unit (RPU) increases significantly. This paper proposed a new interconnection structure called hierarchical local interconnection for RPU. Comparing to traditional full-mesh structure used in MorphoSys, the hierarchical local interconnection greatly enhanced the area efficiency while retaining the flexibility of interconnection. When the array scale is 8 × 8, hardware overhead of new structure is 28.6 % of the traditional structure.

A methodology is presented to detect and locate high impedance faults (HIFs) in radial distribution system by means of nonlinear frequency analysis. The proposed technique is based on the analysis of the feeder responses to power line carrier signals, which are periodically injected at the outlet of transformer. The effectiveness of the method has been verified through simulation studies. The results demonstrated that the proposed method has the potential to be applied in practice to resolve HIF real-time monitoring problem.

A methodology is presented to detect incipient faults in distribution networks by means of DWT and energy detection algorithm. The proposed technique is to extract the characteristic of incipient fault by DWT method, that is, to extract the d5 coefficient of wavelet decomposition of residual current and residual voltage. Compare energy value with normal situation using an energy detection algorithm; incipient faults are detected. The proposed technique has been investigated by ATP/EMTP. Simulation results show that this technique is effective and robust, and the proposed method has the potential to be applied in practice to resolve incipient fault real-time monitoring problem.

The mathematical expression and physical implementation are analyzed for a D/A converter and illuminated the T-type network framework of a binary digital-to-analog transform by dividing current means in this paper. Based on it, slice of half-dividing current is suggested by way of the symmetry of the drain and the source terminals in CMOS transistor. The paper puts forward a novel CMOS D/A converter based on T-type weighted current network with R-2R ladder. It has the merits of low power consumption and easy making of integration. Simulation result reveals a monotonic characteristic of the D/A converter.

The rapid development of the smartphone brings immense convenience to people. Recently more and more developers publish their own applications (or apps) on the android markets to make profits. The so-called repackaged apps emerge by embedding malicious codes or injecting ads into the existing apps and then republishing them. In this paper, focusing on the shortcomings of existing detection system, we propose an efficient repackaged apps detection scheme based on context-triggered piecewise hash (CTPH). We also optimize the similarity calculation method (edit distance) and filter unnecessary matching process to make the matching more efficient. Experimental results show that there are about 5 % repackaged apps in pre-collected data. The proposed scheme improves the detection accuracy of the repackaged apps and has positive significance to the ecosystem of android markets.

In order to solve wireless local area network (WLAN) security problem due to the open-wide nature of wireless radio and the improvement of computing power, a design of WLAN security program based on near field communication (NFC) is presented in this paper. In this paper, the importance of having access to handshake for WPA2 brute force is explained. The proposed design protects the four-way handshake by taking advantage of NFC short-range character to eliminate the risk of intercept. For implementation, Android system is selected as a mobile device development platform. The design is compatible with the IEEE 802.11i which ensures the massive expansion in the future. Furthermore, the design simplifies operations to improve users’ experience without much extra hardware cost and offers an option to the owner of WLAN to control the access physically, which benefits commercialization of NFC. From one perspective, this design can solve the wireless network security problem effectively.

One of the fundamental challenges in research related to cyber-physical system is accurate modeling and representation of these systems. The main difficulty lies in developing an integrated model that represents both cyber and physical aspects with high fidelity. Among existing techniques, an approach to integrate Modelica with AADL is a suitable choice, as it can encapsulate diverse attributes of cyber-physical systems. AADL modeling language provides a comprehensive set of diagrams and constructs for modeling many common aspects of systems engineering problems, such as system requirements, architectures, components, and behaviors. Complementing these AADL constructs, the Modelica language has emerged as a standard for modeling the continuous dynamics of cyber-physical systems in terms of hybrid discrete event and differential algebraic equation systems. Integrating the descriptive power of AADL models with the analytic and computational power of Modelica models provides a capability that is significantly greater than provided by AADL or Modelica individually. A transformation of AADL into Modelica is developed that will support implementations to transfer efficiently the modeling information between AADL and Modelica models without ambiguity. This chapter proposes an approach to transform the models of AADL into the models of Modelica, to clarify the transformation principles, and to illustrate the important synergies resulting from the integration between these two languages.

Cyber-physical systems have varying quality-of-service (QoS) requirements driven by the dynamics of the physical environment in which they operate. Developing cyber-physical systems is hard because of their end-to-end QoS requirements. Aspect-oriented development method can decrease the complexity of models by separating their different concerns. We can model QoS as a crosscutting concern of cyber-physical systems to reduce the complexity of cyber-physical system development. In this paper, we propose an aspect-oriented QoS modeling method based on AADL. We present our current effort to extend AADL to include new features for separation of concerns, and we make an AADL extension for QoS by aspect-oriented method. Finally, we illustrate QoS aspect-oriented modeling via an example of transportation cyber-physical system.

Cryptographic hash functions play a fundamental role in today’s security applications. In general terms, the principal applications of a cryptographic hash function are to verify the integrity of the data, which refers to data authentication or data integrity. The one-way hash chain is an important topic in key management and is also an important cryptographic primitive in many security applications. As one-way chains are very efficient to verify, they are also the primitives to design security protocols for ultra-low-power devices. In this chapter, an RC4-based hash function RC4-BHF is introduced and how to use RC4-BHF to construct efficient one-way hash chains is proposed. The proposed construction for one-way hash chains is efficient and is designed for ultra-low-power devices.

Tag prediction is proposed to reduce the leakage power consumption of instruction cache and the power consumption of branch prediction that represent a sizeable fraction of the total power consumption of embedded processors in this chapter. By extending the architectural control mechanism of the drowsy cache to predict the cache line read in the next access, the tag prediction wakes up the necessary cache line in advance, while the rest of cache line is in the drowsy mode. Empirical results show that the tag prediction reduces the 77 % power consumption compared to the policy adopting branch prediction, and the accuracy of tag prediction is roughly same with the accuracy of BTB prediction. By removing the BTB and adopting the technique of drowsy cache, the tag prediction effectively reduces the power consumption without significant impact on performance of processors.

The essence of cyber counterwork mainly reflects as the counterwork process between people, in order to solve effectively the cyber counterwork test problem about the cognitive level and decision level. In this paper, firstly, the dynamic adaptive cyber attack and defense “observation, orient, decision, act” (OODA) loop process models are established, which are based on the traditional military command and control operational process model. Secondly, establishing the cyber attacker and defender role models during the cyber counterwork process, which are mainly from role’s identities, role’s function, role’s capability, and role’s relationship utilizing the multi-attribute group description method. At last, establishing capability evaluation index system for each role and evaluating the capability through Delphi method. The human role model can provide theoretical basis for configuring attacker and defender role during the cyber cognitive and decision level test process.

IEEE 802.11p vehicular ad hoc network (VANET) applies multiple channels, including one control channel (CCH) and six service channels (SCHs); the enhanced distributor channel access (EDCA) mechanism is used to support wireless channel assignment and QoS requirements. But the method of SCH assignment is not proposed in IEEE 802.11p standard. We present a scheme to perform SCH assigning, previous transmission indicators of service channels are detected dynamically by service channel assignment controller set in medium access control (MAC) layer, service packets would be delivered into suitable SCH and EDCA access category (AC) queue according to SCH reservation probability and estimated transmission delay. Saturated throughput of our scheme in SCH is analyzed by theoretical model in different conditions; the results show that it can ensure higher SCH utilization and is an efficient way to improve performance of intelligent transportation system.

According to the problems of exception handling for service-oriented software, this paper presents a framework for Web service exception handling (EHF-S) based on policy driven. The EHF-S processes the response message of invoking Web service and produces a response message which is added exception information and exception handling message. We introduce the realization principle, the component, and the key technology for EHF-S. This framework can support the development and integration of exception handling logic for Web service process, improve the exception handling capability, and simplify the exception handling process for Web service.

In order to reduce the operation blocking rate of static resource allocation in SDH Mesh network effectively, balance network traffic, optimize the allocation of network resources, enhance the success rate of multiline information routes, and improve the overall performance of the network. This chapter introduced resource congestion avoidance algorithm (RCAA) based on the adjustment, which can effectively solve the resource congestion in the static resource allocation. In order to prove the feasibility of this RCAA, three simulation examples of resource allocation theory were adopted. Through analysis validation of these three examples, this article proved that RCAA based on the adjustment proposed in this paper can effectively reduce the blocking rate and improve the overall performance of SDH network. RCAA based on adjustment is more superior to ANM. RCAA can avoid resource congestion problems caused by the allocation of resources effectively.

In this paper we face the challenge of designing self-tuning systems governing the working parameters of base stations on a mobile network system to optimize the quality of service and the economic benefit of the operator. In order to accomplish this double objective, we propose the combined use of fuzzy logic and reinforcement learning to implement a self-tuning system using a novel approach based on a two-agent system. Different combinations of reinforcement learning techniques, on both agents, have been tested to deduce the optimal approach. The best results have been obtained applying the Q-learning technique on both agents, clearly outperforming the alternative of using non-learning algorithms.

Virtual machines have been widely adopted as servers nowadays. They have essential difference with physical machine. We can utilize the feature of virtual machine to let them be safer and resist an attack from Trojan and hackers. This paper introduces a kind of network access security scheme, which deploys the execution of security strategy outside virtual machine and monitors virtual machine’s access to security-sensitive device. The measurements above can transfer the control for key hardware from upper Guest OS to host a platform. Even if Guest OS is affected by virus or Trojan, host can still effectively monitor the network communication of upper virtual machine. In this project, software running in Host OS is programmed to realize the scheme introduced above, it monitors the network communication of virtual machine according to the rules written in XML format. The software can prevent Guest OS or an application running on the virtual machine from communicating with designated domain or IP address successfully, which verifies the effectiveness of the proposed security scheme.

Aiming at some special applications, such as monitoring of high-speed rail and monitoring of large farm field, a wireless sensor network based on chain structure is proposed. Considering of simplicity and energy saving, two light protocols, which are based on time slot and competition, respectively, are applied in the above network. Finally, the two light protocols are compared with IEEE802.15.4 protocol by OPNET simulation, and the results show that the proposed light protocols have good reliability and low energy consumption.

The importance to build relevant peripheral environment in the testing process for complex embedded software is becoming higher. This paper discussed the current design method of simulation test environment for the embedded software and then presented a modelling method which is used to build peripheral simulation environment for the SUT (system under test) through ICD (interface control data) documents and the software requirement specification. Using this method, the peripheral environment simulation tool which consisted of relevant database and simulation model was set up with Ruby program language. This tool could provide necessary control commands and data support just like in a real running environment for the SUT. Furthermore, a DSL (domain-specific languages) design method for this domain was researched on the basis of the model. The experiment result has demonstrated that it’s feasible to set up a peripheral environment for embedded system with our simulation tool.

In order to analyze the regulation of information dissemination on the complex network, SIR probability model has been built to represent the peoples’ interaction during information dissemination on complex networks. By introducing and computing the state transiting probabilities of the net nodes, we can effectively analyze and update the nodes’ states at each step in information dissemination. Accordingly, the evolution algorithm of information dissemination is designed and realized by simulation. Simulation experiments of information dissemination on ER network and BA network with different parameters reveal that the density of final awareness will not be affected by the total of nodes, but increase progressively following the increase of average degree until a certain value. Different degree distributions can also be effect on the density of final awareness. SIR probability model can accurately reflect the process of information dissemination on complex networks. It can be used for the description and analysis of information dissemination on complex networks.

The UML is a semiformal modeling language which only has syntax and static semantics precisely defined. The dynamic semantics for the UML is specified neither formally nor algorithmically. When using UML at the design phrase, there does not exist a systematic way that allows the model designer to specify its formal semantics and automatically verify correctness properties of the described model. The UML sequence diagrams are widely used to describe the behaviors of software. Reasoning about properties of sequence diagrams at the analysis and design phrase may reveal software faults before software implementation. We propose to use the theorem proof assistant—Coq to verify syntax and semantics constrains of sequence diagrams. The verification and proof process are useful for improving the correctness of sequence diagrams and hence increases the software quality.

In order to verify the reliability of the bounded retransmission protocol, probabilistic model checking technology is used in this paper. The integer semantics approach is introduced, which allows working directly at the level of the original probabilistic timed automaton (PTA). In such a method, clocks are viewed as counters storing nonnegative integer values, which increase as time passes. The PTA modeling the system can then be seen as a discrete-time Markov chain. Based on this fact, the protocol is modeled directly with DTMC. Properties are described in probabilistic computation tree logic. By making an analysis of the quantitative properties of the protocol, a threshold is obtained. Experimental result shows that no matter how many chunks to be transmitted, if the maximum retransmitted time is greater than or equal to 3, the protocol can be considered reliable. Method in this paper can not only verify the correctness of a system but also make analysis of nonfunctional indices of a system such as reliability or performance.

As using traditional MDS-MAP algorithm to locate nodes’ position in irregular WSN leads to low positional accuracy, based on this fact, this chapter presents an improved algorithm named MDS-MAP(C, RF). The algorithm can effectively divide a WSN into several clusters, and each cluster locates all nodes’ position in it and forms a local position map. After all clusters get local position maps, the algorithm merges all the local position maps together using the information of inter-cluster nodes. Simulations demonstrate the proposed algorithm yields smaller accuracy error in irregular WSN.

In order to discover a variety of network resources of structured P2P, resource information organization methods are required, which should have scalability and robustness. However, structured P2P has bad performance because of churn, so it cannot be widely used currently. To solve the problem, a resource information organization method based on node encoding is provided in this chapter. A node group-based resource information organization and resource distribution-based node encoding algorithm are presented. Redundancy tables are established based on the overlay of the node. The proposed algorithm can decrease the burst of transmission and reduce the traffic load of transited information. The experiment results show that the presented method is tolerant to churn.

In order to solve the EPC codec problems based on Internet of Things (IOT) applications for trade enterprises, in this chapter an EPC codec system is designed for enterprise applications. According to the “Tag Data Standards,” we design the encoding and decoding algorithm/schema of SGTIN-96. On the basis of the algorithm/schema, the system has improved its coding and decoding algorithm, making the coding algorithm more simple and practical and improving the efficiency. Besides, it also has realized the transformation between SGTIN-96 and GTIN-14, which makes the final printed electronic tag contain both a bar code and the EPC code, and realizes the compatibility of bar code and EPC code. The coding and decoding system can code and decode well for the products of the trade enterprise. Through the system, SGTIN-96 labels can be generated and printed. And the content of SGTIN-96 labels can be decoded. Finally, we test the codec system to prove that it can achieve our established requirements.

Many natural large-sized complex networks exhibit a scale-free, power-law distribution of vertex degree. To better understand the formation mechanism of power law in the real network, we analyze the general nature in scale-free network based on the vertex-degree sequence. We show that when the power exponent of scale-free network is greater than 1, the number of degree-

k

1

vertices, when nonzero, is divisible by the least common multiple of 1,

k

2

γ

/

k

1

γ

, …,

k

i

γ

/

k

1

γ

, and the length of vertex-degree sequence

l

is of order log

N

, where 1 ≤

k

1

<

k

2

< … <

k

l

is the vertex-degree sequence of the network and

N

is the size of the network. We verify the conclusion by the coauthorship network DBLP and many other real networks in diverse domains.

In order to analyze the problem of packet collision in the asynchronous mode of persistent carrier sense multiple access (p-CSMA), in which there is no time slot different from synchronous mode and propagation delay have a heavy effect on the probability of packet collision, a preemptive asynchronous p-CSMA probability model is established for the first time in the chapter. From sub-cycle conditional probability, the model gives the expectations of an idle and busy period. On the basis, performance targets, e.g., throughput/delay/success rate and channel efficiency, are gotten. For illustration, VDL2 (a typical asynchronous p-CSMA network) simulation model is set up on OPNET platform and experiments are also carried out to verify the correctness of this model in diverse scenarios. Through simulation, the results of fixed position distribution have the good consistency with the preemptive probability model. Finally, the conclusion is achieved that packet collisions will aggravate with the stations distribution becoming more uneven.

In order to display the attack scene in the description of the multistep process-oriented attack—advanced persistent threat, a specific model on advanced persistent threat behavior analysis—EPNAM is proposed, which is based on the Petri net and combined with the characteristics of APT. Firstly we carry out hierarchical analysis on the attack scene with AHP method to build the APT architecture and extract scene factors, then associate the attack scene with Petri net to construct extended Petri net, and finally, traverse the extended Petri net to generate the formal expression. The proposed model can achieve the combination of the attack scene, attack process, and state space, and its feasibility is proved by the application on actual case analysis of the RSA SecurID theft attack.

This chapter mainly discusses the problem of wireless sensor network routing protocols. Based on analysis of the disadvantages of information implosion and overlapping caused from implementation mechanisms of the flooding protocol, energy-efficient routing protocol based on probability of wireless sensor network (ERPBP) is proposed and evaluated. It uses the node distance and residual energy as the weights to calculate the forwarding probability of neighbor nodes and chooses some of maximum forwarding probability nodes as router. It saves the energy by avoiding redundancy packet copies produced and improves the disadvantage of flooding routing protocol. Performance analysis and simulation experiment show that the new protocol effectively reduces the data redundancy, reduces the energy consumption, and prolongs the network lifetime.

Many sensor query processing systems have been developed to acquire, process, and aggregate data from wireless sensor networks. The energy consumption of query processing is significantly impacted by routing protocol. In this chapter, we propose a dynamic routing protocols switching scheme for query processing. The scheme supports multiple kinds of routing protocols coexisting in a single sensor node, and these protocols can be switched according to query tasks. Simulation results show that the dynamic scheme is more energy efficient than single routing protocol.

It is difficult for conventional relaying algorithms to detect incipient faults, such as insulator current leakage, electrical faults due to tree limbs, and transient or intermittent earth faults, which are frequent in distribution networks. With the time, they may lead to a catastrophic failure. In order to avoid this situation, S-transform technique is proposed to extract the suitable features of incipient fault in this chapter. A least square support vector machine (LS-SVM) classifier is developed utilizing the features so that incipient fault is distinguished from the normal disturbances. Then the polarity of magnitude difference of residual current is used to determine the fault section of distribution network. The proposed technique has been investigated by ATP/EMTP simulation software. Simulation results show that this technique is effective and robust.

This paper is to introduce the new concept of coalition Nash equilibrium of a strategic game. A coalition Nash equilibrium for a strategic game consists of (1) a subset

S

of players, (2) independent mixed strategies for each member of

S

, and (3) the conjecture of the actions for the other players not in

S

with the condition that each member of

S

maximizes his/her expected payoff according to the product of all mixed strategies for

S

and the other players’ conjecture. Let us consider that each player communicates privately not only his/her belief about the others’ actions but also his/her rationality as messages according to a protocol and then the recipient updates their private information and revises her/his prediction. Then we show that the conjectures of the players in a coalition

S

regarding the future beliefs converge in the long run communication, which lead to a coalition Nash equilibrium for the strategic game.

In order to find out the optimal layout of bike rental stations for B+R mode, a bi-level programming model combined of genetic algorithm and the joint model with mode split and traffic assignment model is built. The optimal layout plan can minimize the total travelling cost and facility cost. A case is used to test and verify the practicability of the model. The result shows that the model can effectively solve the layout problem of bike rental stations for B+R mode and can offer suggestions for related planning.

Train control systems must guarantee a very high level of safety because their incorrect functioning may have very serious consequences such as loss of human life, large-scale environmental damages, or considerable economical penalties. The software reliability is related to several factors, such as completeness, consistency, and lack of ambiguity. Formal methods are widely recognized as fault avoidance techniques that can increase dependability by removing errors during the specification of requirements and during the design stages of development. In this chapter, a brief overview of existing results on formal specification of train control systems is first presented. Then we propose an integrated formal approach to specify train control systems; this integrated approach combines CSP and Object-Z with Clock theory to specify the Railway Control System concerning both the linear track and crossing area, especially the time delay between any two aspects of the railway system.

In cyber-physical systems, the elapse of time becomes the most important property of system behavior, and time is central to predicting, measuring, and controlling properties of the physical world. A cyber-physical system is composed of two interacting subsystems: a cyber system and a physical system. The behavior of the cyber system is controlled by the execution of programs on a distributed digital computer system, while the laws of physics control the behavior of the physical system. The different models of time—continuous physical time in the physical system versus discrete execution time in the cyber system and the impossibility of perfect synchronization of the physical clocks of the nodes of a distributed computer system, lead to interesting phenomena concerning the joint behavior of these two subsystems. The chapter describes the case studies in applying clock theory to the production cell. The clock theory described is very simple, in that it models clocks as potentially infinite lists of reals. Xeno’s paradox and similar problems are avoided by specifying limits on clock rates, which effectively means that the model sits somewhere between a discrete synchronous model and a fully dense continuous-time model as assumed by some other formalisms. The case study of the specification of the production cell shows that using clock theory to specify cyber-physical systems can give a more detailed description of the every subsystem and give a much more considerate observation of the time line and sequence of every event.

This chapter proposes a novel worm signature that is appropriate for the polymorphic worm detection. Most of the recent worm signatures are constructed based on worm bytes themselves or relationships between worm bytes. In this case, most of these signatures cannot detect the polymorphic worms successfully. Our worm signature takes the worm bytes themselves and the relationships between worm bytes into consideration. So, it is called position-relation signature (PRS). The new signature is capable of handling certain polymorphic worms. The experiments show that the algorithm could be used as a basis to implement a worm detection system.

Since many predicting methods, such as CS and PP are not very precise, Wavelet-ANFIS with high estimation precision is always used to model the decomposed series recently. This chapter uses wavelet analysis to decompose water level series and then uses ANFIS to model the decomposed series; in the end, it combined these series and predicted Lingxi Reservoir’s runoff. The runoff forecast of reservoir is essential for its flood control safety. The forecast result shows that the prediction accuracy of Wavelet-ANFIS is very high and the model is quite fit to use in daily runoff and water level prediction.

Hybrid layout is a common visualization technique for clustered network graphs. Since most previous hybrid layout methods do not consider a reasonable balance between screen utilization and layout aesthetics of the network graphs, the inappropriate partition of the display region may result in unpleasant display effect of network graphs. This chapter proposes to address this problem with nonlinear constrained optimization techniques. This chapter analyzes why the circular algorithm would fail in region partition. To ensure that every subgroup of network nodes can be assigned to a rectangular region, the maximal utilization of the display area is taken as an objective function and the rectangular ratio is taken as constraints. The constrained optimization layout model leads to efficient balance between regional utilization and layout aesthetic. Experimental results show that the constrained optimal partition layout generates more balanced relation network graphs with better visual effects.

In order to decrease the power waste of relay node, the paper presents a novel ultra-wideband cooperative communication method that uses two relay nodes to transmit reference impulses and data impulses separately. A transmitted cooperative-reference UWB cooperative communication model is developed in this paper. Based on the model and sampling expansion approach, a closed-form SER expression was deduced for delay-hopped transmitted-reference UWB systems which use cooperation strategy of decode and forward relaying and equal-gain combining. Simulation results show that the transmitted cooperative-reference method can obtain multi-order diversity gains.