Advanced Materials Research Vols. 756-759

Abstract: With the growing popularity of Web applications and Web attacks increasing; Web applications protection and data security have become the core focus of the information security system of enterprises and institutions. The paper briefly presents the harm of the common several kinds of Web attacks, and analyzes their principle, gives their defense measures. On this basis, the defense system of both software and hardware design is proposed. It achieves software defense design based on Web software development and Web application management, and hardware defense design based on H3C SecBlade IPS, to effectively guarantee the security of Web applications.

Abstract: This paper analyzed the basic object-oriented concepts from the perspective of testing. The effects of the characteristics of object-oriented software on the software testing were discussed. The class testing method of object-oriented software was put forward. This method includes tests based on the state transition diagram and data flow testing on class. A integration testing of object-oriented software was put forward based on the event-driven characteristics of object-oriented software, and a data-generating method of software test based on genetic algorithm was provided. The test case generating technology of object-oriented software was discussed, which utilized an intercalation method of branch function and regarded the genetic algorithm as the core search algorithm.

Abstract: many functions are possessed by the neural network such as parallel processing, self-learning and self-adapting. It could approximate any nonlinear function with any precision. A very effective way is provided by the neural network to deal with complex control problems, such as nonlinear, multivariable and uncertain ones etc. Therefore, the neural network is widely used in many aspects: pattern recognition, system identification and control fields and so on.It is developed in the paper about the application of neural networks pattern recognition and system identification. With MATLAB 6.1 and Visual Basic 6.0 design platform and developing tool, for some application instances, implement modeling, simulation and systematic test tasks of the neural networks pattern recognition and system identification. The above research and instances indicate that the neural networks pattern recognition and system identification based on MATLAB have better application prospects.

Abstract: We study the problem of traffic lights scheduling in a dynamic three-dimension intersection. In this paper, an intelligent traffic lights schedule approach is proposed to optimize the traffic lights sequence according to traffic data detected and predicted in real-time, e.g. traffic volume, delays and traffic density. Evaluations have been conducted to compare the performance in terms of traffic throughput, and the result depicts that our intelligent approach can obtain better performance.

Abstract: In Cognitive Radio Networks (CRN), an important issue associated with MAC-layer sensing is how often the availability of licensed channels should be sensed. To solve this issue in multi-channel environment, we address how to maximize the discovery of spectrum opportunities by setting a set of reasonable sensing-periods. Based on alternating renewal theory we develop an channel-usage pattern estimation method, and derived the states transition matrix by discrete-time Markov process. At last, we proposed an objective function of spectrum sensing efficiency, and solved the multi-objective optimization problem by improved Genetic Algorithms (GA). Our simulation results demonstrate the efficiency of the objective function and the significant performance of the improved GA. The sensing-period vector we derived discovers more than 80 percent of the analytical maximum of discoverable spectrum-opportunities in six-channel environment, which is shown to discover up to 30% more than the method that sensing each channel with the same period.

Abstract: As a tool for energy metering, electric meter is the function integration base of smart grid. Traditional mechanical meters and ordinary electronic meters havent met the needs of the modern smart grid. In this paper, a novel smart meter system based on 3G is proposed. This 3G smart meter system is integrated of some new technologies, such as effective energy metering, embedded MCU, 3G wireless communication & Database management. The constitution of system and the hardware and software design have been introduced in detail in this paper. The experiment shows that some real-time energy measurement parameters have been provided to power users by the 3G smart meter system, and the two-way wireless data communication between power users and power corporations has been accomplished effectively. Furthermore the management level of power corporations has been promoted.

Abstract: Bayesian network method for system reliability evaluation which is based on a Bayesian network that transformed from a fault tree has gotten much attention these years. After a brief introduction to the method how to transform a fault tree into a Bayesian network, the paper elaborates the Bayesian network inference algorithms. The paper focuses on the way how the inference algorithms can be applied to the practice of system reliability evaluation and designs a systematic flow chart used to evaluate system reliability in a Bayesian network way. The experiment demonstrates the feasibility of the systematic flow chart.

Abstract: In recent years, increasing attention has been paid to disciplinary variation of chunk use, but so far little has been done on chunks in the field of applied linguistics. This paper aims to explore the structures and functions of 4-word chunks in 1, 032,497 word tokens corpus of applied linguistic research articles. The analysis reveals that applied linguists tend to use more prepositional phrases with of fragments and noun phrases with of fragments. Moreover, research-oriented chunks are the most prevalent, text-oriented the next, participant-oriented the fewest. Lastly, pedagogical suggestions are put forward that students awareness should be aroused of unique features of chunk use in applied linguistics in terms of structures and functions.

Abstract: We consider a finite butter single server queue with batch arrival, where server serves a limited number of customer before going for vacation (s).The inter arrival times of batches are assumed to be independent and geometrically distribute. The service times and the vacation times of the server are generally distributed and their durations are integral multiples of slots duration. We obtain queue length distributions at service completion, vacation termination and arbitrary epochs.

Abstract: The key of the network teaching system is how to provide appropriate teaching resources and adopt different teaching strategies according to different users. In this paper, data mining is used in network teaching system, and builds a personalized network resources database. According to the information of the user, find the relevant rules and provide the learning resources for each learner to adapt to their needs. The results demonstrate that this proposed approach for improving the level of personalized service in network teaching system is effective.