Advanced Materials Research Vols. 989-994

Abstract: Large misalignment is unavoidable for subsystems which could be deployed randomly on the carriers such as shipborne aircrafts, AUV. Ordinary linear filtering algorithms don’t converge fast and accurately in non-linear conditions. It's critical for the accuracy of the transfer alignment. In this paper, a new misalignment and gyroscope bias online estimation method based on angular velocity processing is presented. Sensor measurements of M-SINS and S-SINS will be recorded for a certain period. Misalignment and the gyroscope bias will be calculated from these measurements directly with non-linear least square algorithm. Trust region method with pre-conditioning, subspace and conjugate gradient are applied for faster converge and better accuracy. Simulation results demonstrate the effectiveness of the algorithm.

Abstract: The singularly perturbed systems for the nonlinear three-species food-chain reaction systems with time delay are considered. By using the method of the stretched variable, the formal asymptotic solution is obtained under suitable conditions.

Abstract: An algorithm of DOA estimation based on linear canonical transform is proposed. In this method, the observed signals are first transformed to linear canonical domain, through selecting appropriate transformation parameters, characteristics of the observed signal appear to be energy-concentrated in the corresponding linear canonical domain, and by that, the linear canonical domain steering vector is defined, then root-MUSIC is used to estimate DOA of impinging signals. Simulation examples are provided showing the effectiveness of the presented technique.

Abstract: Aim at the multi-faceted nature and universality of the EMC problems in high-speed PCB, based on a three-line coupling model and considering the influence of via holes caused, board-level EMC problems of PCB were researched on point for point. In this paper, the research and simulation were launched by a kind of software named Hyperlynx, and a 6-layer PCB model was built and used. Our Research was focused on the EMC problems existed in hierarchy and layer medium. On the basis of the simulation results, and combined with previous experience in the PCB design, some effective measures for improving EMC in high-speed PCB design were summarized, which could be taken as a clearly reference in design of high-speed PCB.

Abstract: In the visualized simulation, the application of texture mapping technology can significantly improve image fidelity and superimpose the rich texture details to the surface, making the scene more realistic with real-time and fast invocation. This paper introduced the principle of texture mapping, the mapping process and implementation in software, and described in detail the implementation process of multi-layered texture mapping by example.

Abstract: The data mining can help to extract the information and knowledge with potential application value from a huge amount of data. In order to do the date mining scientifically and efficiently, this article provides an improved algorithm based on the classical association rules Aprior algorithm, taking advantage of which, we can engage in the association rules mining of the database to obtain the useful information. The improved algorithm avoids pattern matching and reduces the number of times of visiting the database, thus improves the speed of date mining to some extend.

Abstract: In the information management process, it is necessary to classify all information optimization to provide the support for quickly query the search target providing this, a classification optimization method based on improved ant colony algorithm is proposed in this paper. That is, extract the characteristics of information attribute, and then establish the optimization classification model based on the characteristics to achieve the classification of information. Experimental results show that using the improved algorithm for classification information, the classification accuracy can be improved to meet the actual demand data management.

Abstract: Herd mentality is a universal, objective social psychological phenomenon in humans describing how people are influenced by their peers to adopt certain behaviors even against their wishes. Herd mentality has a dual nature. This paper analyzes the performance and causes of herd mentality, and makes appropriate adjustment strategy, which has great significance to both the individual and national development.

Abstract: In aerobics, the motion state randomness of key feature points is great, so it is difficult to establish an accurate dynamic model for sports' shape base. Traditional 3D reconstruction algorithms use fixes shape base which hardly expresses the change parameters of complex movement and motion law of large-scale dynamic features, thereby leading to non-realistic reconstruction results. The paper proposes a new reconstruction algorithm for aerobics 3D motion images that corrects the neighborhood system of feature points by motion parameters until the parameter is stable to ensure accuracy and the stability of correction. The simulation results show that, the proposed algorithm avoids drawbacks of sports reconstruction results caused by the great randomness of aerobics' motion state, thereby complete 3D reconstruction for aerobics' motion images.

Abstract: The data classification is an important issue in massive data classification. This paper proposes an inter-cell classification algorithm based on phase recombination neighbor points convergence which analyzes the convergence value weights of inter-cell characteristic points and filter the interferences of the minority local optimal characteristic points. The proposed algorithm can promote the convergence of the inter-cell classification data neighbor points. The simulation experiments testify the models by three types of actually collected data sets which illustrate the models have better classification performance.