Search

Internet has infiltrated into every aspect of our lives. Accurate topological modeling is essential to improve Internet performance in many research areas, such as protocols design, routing and searching algorithm, virus spreading protection, and node failure effect. Based on the power-law characteristic of AS-level Internet topology, a time-varying evolving Internet topology model is proposed by applying dynamic rates in node growing, different level link connecting and link disconnecting in modeling process. Node degree distribution, average clustering coefficient and average node distance of the model match Internet topology corespondent parameters. An Internet topology with any given node number within span can be easily built by this model.

To present the shape and ore grade distribution of orebody in three dimensional space, and apply visualization to ore resources evaluation and analysis, an interactive visualization system for orebody modeling and grade estimation were developed. A group of sophisticated algorithms were utilized to integrate the orebody surface model and block model to a new scheme of hybrid model. Orebody surface model was reconstructed from planar contours. To convert the complex surface model to block model, an octree was used as data structure, OBB trees and Feito-Torres algorithm were employed respectively to implement intersection test and inclusion test of a block and surface model. Ore grade of blocks within block model were estimated by geostatistical techniques. An example of a copper mine has been applied to the visualization system is presented. This system can be a basic requirement for orebody modeling and facilitates the understanding of ore resource under the earth. Users of this visualization system can find that its strengths lie in simplicity of complex orebody modeling and increased availability of important information in 3D space.

Algorithms for Boolean operations using BSP trees and other algorithms have problems with efficiency and robustness. In order to improve the speed and correctness of operations, a new algorithm for Boolean operations on polygonal meshes was presented. In this algorithm, a fast collision detection algorithm based on oriented bounding box (OBB) was employed, which speed up the intersection test between every two triangles of two different polygonal meshes. After getting the intersection segments, the polygons can be obtained from intersected triangles and intersection segments, then establish the intersection region. Finally, the ultimate result was obtained by acceptance or rejection for the other meshes based on intersection region and the neighborhood of these meshes. In this algorithm, orientability, topology and other geometric characteristics are made full use of. The result shows that the algorithm is efficient and robust. Furthermore, it can apply to both closed meshes and open meshes.