Random Dissemination Algorithm Of Updating Packets For Distributed Virtual Environments Of P2P Architecture

With the network technology, especially the rapid development of wireless communicationtechnology, virtual reality technology has been more and more applied to the various fields of thesociety. Distributed virtual environment combines with the two above kinds of technology and ismore widely going into people’s lives. Distributed virtual environment refers that the dispersedgeographic locations of virtual reality terminal maintain a virtual environment through network. Inthe virtual world, the users who are in different locations participate in the same virtual environmentfor communication, interaction, collaborative work through its virtual character.Distributed virtual environment has many problems, such as scalability, view inconsistency.These problems have seriously affected the original intention which the distributed virtualenvironment provides interactive, immersive characteristics for people. The distributed virtualenvironments based on P2P can well solve the scalability of the system. But the inconsistencyproblem is difficult to solve. This problem is important to the design of the whole system.Considering system has good adaptability and self-organization in P2P distributed virtualenvironment, each of user node can communicate directly with interaction. So the factor of delaywhich node obtains status message has a direct impact on the view inconsistency problem.In this paper, the view inconsistency problem in the distributed virtual environment is describedand the present situations of the research of the problem are analyzed. We present the model designand analysis of the time-space inconsistency which measures the view inconsistency and make thefocus on the problem of the larger communication delay which nodes distribute the status messagesin the region of interest. We take a detailed analysis of the limited network bandwidth which affectsthe distribution of state update message in the area of interest. For getting a method of distributionrouting and at the same time to reduce the temporal inconsistency rate, the design of algorithm ofreducing the delay is given.First, by studying the existing research of view inconsistency problem, we determine to use thesystem model which is combined AOI (area of interest) and DR (dead reckoning). AOI partitiontechnique and DR prediction model are the technique of controlling the quantity and frequency ofdistribution of the update message. The article gives the space and time inconsistent models ofspecific introduction, which is a measure of view inconsistency. Based on the models, we define thedistribution problem of DR update messages. The problem is that find the distribution strategy tomake all the nodes in the region of interest to obtain a minimized delay of status message.Second, we analyze the distribution problem by using structured graph, and put forward many algorithms, which are the algorithm of constructing a communication sub graph with the minimizedmaximum delay and random broadcast algorithm based on random geometric graphs. But the firstalgorithm is not suitable for the situation where the user nodes have increased dynamically. Inaddition, the second algorithm also has certain disadvantages in the specific application. So, we usea greedy strategy for routing algorithm to optimize the random broadcast algorithm.Finally, the algorithms are implemented and simulation experiments are made. Theexperimental data shows that compared with several other algorithms, the random broadcastalgorithm can reduce the event rates of the time-space inconsistency and improve the systemperformance.