Advanced Materials Research Vols. 268-270

Abstract: The negative high-voltage power supply of Electron Cyclotron Resonance Heating (ECRH) is a nonlinear system with serve sensitivity and it is not well for traditional controller to meet restrict demand on stability and quick response. Based on the concept of credit a novel CMAC is designed to accelerate the convergence of traditional CMAC and also is used as an intelligent controller for the power of ECRH based on the idea on direct inverse control. Experiment results show that ICA-CMAC can control the power of ECRH well with shorter settling time and less CPU consumption thus the validity of ICA-CMAC is determined.

Abstract: In this paper, a novel improved Credit Assigned CMAC (ICA-CMAC) is designed based on the concept of credit to enhance the performance of the traditional CMAC. Then ICA-CMAC is used to predict the human hepatic clearance according to in-vitro data of drugs. The experiment results show that the prediction by ICA-CMAC is faster and more accurate and can be thought as a new and effective way for drug metabolism prediction.

Abstract: CMAC neural network has two advantages: the local generalization and no local maximum value. Currently, ICA-CMAC and FCMAC models are used extensively. However, the two models cannot reasonably characterize the direction and magnitude of network weight in the weight correction algorithm. To solve the problem, an improved CMAC learning algorithm is proposed. It takes iterative errors, iteration number and a window function as the performance. Based on information fusion strategy, it introduces global information into the calculation to optimize the network weight. Through a simulation test, it can be found that the model has significant improvement in terms of convergence speed and prediction control.

Abstract: In the digital image stabilization system, Kalman filter is the most commonly used filter for motion correction. When the wanted movements have large assumptions deviation with the movement model, the result of motion correction will cause divergence and even error. For this problem, a novel motion correction method with adaptive Karlman filter is proposed. The back and forth characteristic of the unwanted motion and the smoothness characteristic of the wanted motion is used to adjust the system noise and the observation error adaptively. Experiment results show that the proposed method can effectively distinguish the wanted and the unwanted movement. Compared with the method with fixed parameters, the proposed method takes into account the smoothness and delay of wanted motion at the same time and it is more adaptively.

Abstract: It has been shown that learning to rank approaches are capable of learning highly effective ranking functions. However, these approaches have mostly ignored the important issue of efficiency. Given that both efficiency and effectiveness are important for real search engines, models that are optimized for effectiveness may not meet the strict efficiency requirements necessary to deploy in a production environment. In this work, we present a unified framework for jointly optimizing effectiveness and efficiency. We propose new metrics that capture the tradeoff between these two competing forces and devise a strategy for automatically learning models that directly optimize the tradeoff metrics. Experiments indicate that models learned in this way provide a good balance between retrieval effectiveness and efficiency. With specific loss functions, learned models converge to familiar existing ones, which demonstrate the generality of our framework. Finally, we show that our approach naturally leads to a reduction in the variance of query execution times, which is important for query load balancing and user satisfaction.

Abstract: In the current, there are less fractal researches for the financial market prediction. In this research, we try to find the rule in the random. After found out the rule by the fractal, we can use the rule to analysis and forecast. We used fractals theory to analysis the feasibility and the accuracy for prediction of weighting stock index in Taiwan. The analysis methods are two parts in the stock market. One is Fundamental Analysis. The analysis is performed on historical and present data, but the goal to make financial projections. The analysis is performed on historical and present data, but the goal to make financial projections. We forecast the weighting stock index is 9057.42 in Taiwan on 2008/04/29, and the actual stock weighting stock index is 8891.74. The error value is 0.0186. It represents the fractal theory to apply to the financial stock market prediction is feasible.

Abstract: The research of pedestrian detection ahead of vehicle is the front direction in the field of vehicle safety assistant driving at present. The method of SVM pedestrian detection based on HOG features is studied in this paper. Firstly, the histograms of oriented gradient features between pedestrian and non-pedestrian samples are extracted. Then the features are used as an input vector of SVM algorithm, getting pedestrian classifier with a higher recognition by training. Finally the trained classifier is loaded into the online pedestrian detection system to detect the transport road image. The experimental results show that the algorithm can effectively identify the different scales and attitude pedestrian in complex background.

Abstract: In the era of computer networks, digital, network, information, with the rapid development of high technology has also brought many problems. Widespread use of network, the network increasing proportion of the population, and uneven quality of the network, network become a new type of criminal tools, criminal place. Crime prevention network, has become the computer field, the legal profession must face one of the topics. And how to resist the various modus operandi, how to do in the case of the detection process did not miss any clues, it became a breakthrough in fight against computer crime areas. This paper[1] firstly outlines the theory of electronic evidence, feature analysis, which describes computer crime / network crime cases in the exploration of electronic evidence collection, extraction and analysis. [1] This work is partially supported by the Opening Project of Key Lab of Information Networks Security of Ministry of Public Security（C09608）

Abstract: IPv6, the address has to aggregation, thus greatly reducing the length of the routing equipment routing table to improve the efficiency of routing and security, but then there is any possibility of network intrusion attack. This paper used to implement IPv6 Snort intrusion detection software, intrusion detection system is proposed as long as the server itself TCP / IP stack on the handling of data packets are different, the packet will bypass the intrusion detection system from the ground to produce a TCP fragment attack.

Abstract: Resource-Constrained Project Scheduling Problem (RCPSP) is a well-known NP hard problem and more intelligent optimization algorithms are developed to solve it. In this paper, genetic algorithm(GA) is employed to deal with RCPSP. A priority value encoding scheme is designed to in the algorithm. The numerical results indicate that our methods is slightly better as far as solution quality is concerned and requires smaller solution time than the GA where an activity list encoding with schedule mode is used.