Catálogo de publicaciones - libros

Real-time rendering of vegetation is currently a problem in need of a solution. The large number of polygons that form this kind of objects means that current hardware cannot achieve interactive rendering of outdoor scenes. This paper deals with the problem and it presents a multiresolution scheme that allows us to represent the whole geometry of the trees using both uniform and variable levels of detail. The method presented here models the trees using two multiresolution models. This is due to the different characteristics of the geometry that forms them. The trunk is modelled by LodStrips , a model oriented towards representing continuous meshes, and the foliage is modelled by the multiresolution model Level of Detail Foliage , presented in a previous work. In this paper, it has been efficiently implemented and extended to allow us to change the level of detail in a variable way, by adapting the resolution of this part of the tree to certain criteria determined by the application. Both of them have been designed to be hardware-oriented. They take advantage of the graphics hardware by adapting the data structures and the rendering algorithms to make the visualisation time efficient. Finally, the multiresolution scheme presented in this paper is compared with the only work that has appeared up to now that uses the same technique.

This paper proposes a multi-resolution technique for real-time animation of large crowds, which is useful for massively multi-player online games. Skinning is used to animate non-player characters (NPCs) at close proximity as well as player characters, vertex keyframe animation is used for middle-distance NPCs, and impostors are used for rendering long-distance NPCs. Both of skinning and vertex keyframe animation are accelerated by GPU, and each character is animated independently of the others. The experiments show that hundreds of thousands of characters can be animated at real-time.

In this paper, we present our work in designing, implementing, and evaluating a set of 3D interactive spatial measurement tools in the context of Cultural Heritage Toolbox (CH Toolbox), a framework for computer-aided cultural heritage research. Our application utilizes a bi-manual, spaceball and mouse driven user interface to help the user manage visualized 3D models digitized from real artifacts. We have developed a virtual radius estimator, useful for analyzing incomplete pieces of radial artifacts, and a virtual tape measure, useful in measurement of geodesic distances between two points on the surface of an artifact. We tested the tools on the special case of pottery analysis.

A tree structure has been widely used in constructing application-layer overlays. It is known that the instability of a tree will significantly reduce the performance of the overlay. In this paper, we propose a novel stochastic model that captures the (in)stability characteristics of an application-layer multicast tree. Our model has considered various important factors related to the tree stability, and we have derived closed-form solutions to a class of typical multicast trees. Our results offer a better understanding on the (in)stability of application-layer multicast trees, and also suggest three effective enhancements to improve their stability.

This paper considers the global communication environment of B3G wireless networks when integrating security mechanism of two wireless networks, WLAN and 3G cellular networks. As a result, the existing integration paradigms are extended to a more realistic network model for B3G based on dividing the cellular network into the visited network and the home network. Also, we presents a security analysis based on a formal model of the cryptographic mechanisms.

We consider the problem of working paths optimization in survivable MPLS network. Both restoration methods: global and local repair are addressed. We focus on an existing facility network, in which only network flows can be optimized to provide network survivability. The lost flow due to a single failure of a network link is applied as the performance metric. Since joint optimization of working and recovery paths is very complex, we suggest to partition the problem into two simpler subproblems and first optimize working routes. We introduce a new objective function for optimization of working routes and propose a heuristic algorithm to solve the considered problem. Next, simulation performance of various methods for working paths assignment is examined.

Mobile Ad-hoc networks (MANETs), providing infrastructure-free wireless instant communication, play important role in tactical military networks. However, providing security in tactical military MANETs, having very large and dynamic structure without infrastructure support in hostile environments, is a very difficult task. In order to address security problems in tactical military MANETs, we propose a new HIerarchical MUlti-Tier adaptive ad-hoc network security protocol based on SIgncryption type key exchange Schemes: HIMUTSIS. Our protocol makes contribution to the military MANETs in three major points: Architectural design, cryptographic methods used in military MANETs and key management techniques. Novel architectural design of HIMUTSIS facilitates certification and key management procedures, provides flexibility and reduces cryptographic workload of the military MANETs. In HIMUTSIS, as a novelty, we offer to use DKEUTS (Direct Key Exchange Using TimeStamp) protocol providing security and performance advantages when compared to some traditional cryptographic methods. Also, multi-security level approach provides adaptive solutions for each layer of the HIMUTSIS. As a key management technique, HIMUTSIS uses hybrid key management approach which reduces rekeying workload of the networks significantly while minimizing single point of failure risk of the military MANET.

This paper introduces a novel dual-level traffic smoothing technique for hard real-time communication over Fast/Gigabit switched Ethernet without any modification to existing Ethernet hardware. First, the necessary restricted conditions for hard real-time communication over switched Ethernet are analyzed using the network calculus theory. Then the dual-level traffic smoothing scheme, which combines the node-level traffic smoothing with the global-level traffic smoothing, is introduced. Finally, we make thorough measurements and the results show that the dual-level traffic smoothing mechanism can provide very good hard real-time communication performance with low system overheads and has more potential for most hard real-time and embedded control systems than other existing approaches.

The decision and ability to aggregate packets can have significant impact on the performance of a communication system. The impact can be even more substantial when the system operates under a heavy load. In this paper, we present a queuing model which describes a packet encapsulation and aggregation process. Using this model, we provide analysis of the end-to-end delay of a packet transmitted by the system. The analytical model is verified by a simulation model of the system. We calculate the maximum number of packets in a single frame for which packet aggregation minimizes the average total delay of a packet. It is numerically shown that when the load is high, the higher the variability of the packet service time, the higher the maximum allowed number of packets in the frame should be to achieve the minimum average total packet delay. However, the impact of the variability of the packet service time is insignificant when the system load is moderate or low.

Long-range dependency and self-similarity are the major characteristics of the Internet traffic. The degree of self-similarity is measured by the Hurst parameter ( H ). Various methods have been proposed to estimate H . One of the recent methods is an eigen domain estimator which is based on Principle Component Analysis (PCA); a popular signal processing tool. The PCA-based Method (PCAbM) uses the progression of the eigenvalues which are extracted from the autocorrelation matrix. For a self-similar process, this progression obeys a power-law relationship from which H can be estimated. In this paper, we compare PCAbM with some of the well-known estimation methods, namely; periodogram-based, wavelet-based estimation methods and show that PCAbM is reliable only when the process is long-range dependent (LRD), i.e., H is greater than 0.5. We also apply PCAbM and the other estimators to real network traces.