Font Size: a A A

Research On Key Algorithms Of P2P VoD System

Posted on:2009-12-25Degree:MasterType:Thesis
Country:ChinaCandidate:H ZhaoFull Text:PDF
GTID:2178360242487782Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the explosive growth of P2P media streaming applications over the Internet, especially in the P2P VoD (Video on Demand) system, many approaches have been proposed to improve the QoS (Quality of Service) of streaming received by peers. The typical methods are discussed mainly on three aspects: incentive mechanism, peer selection and data assignment.Based on a systematical summary of relevant works on media delivery technology, this dissertation focuses on related technologies of P2P media delivery and gains several achievements on some sub-topics. The major contributions of this dissertation are as below:(1) This dissertation proposes a novel incentive mechanism based on Game Theory. Peers in P2P network belong to different organizations and individuals so that their own value orientations affect their behaviors in a certain degree. Thus, besides computer technology, research on incentive mechanism in P2P network includes sociology, behavior and psychology as well. This is novel and interdisciplinary.(2) This dissertation presents a novel mechanism of incentive consistent peer selection. Since most of the recent peer selection mechanisms used in streaming system hasn't taken peer's historical behavior into account, this paper introduces a novel incentive consistent peer selection mechanism. A peer calculates a value according to Game Theory which is the price that requesting peer has to pay and reflects its service ability. Experimental results suggest that the new mechanism can provide steady streaming quality.(3) This dissertation presents a data assignment schema based on genetic algorithm. Data assignment schema is in essence a concurrent schedule issue in P2P VoD systems. Belonging to NP issue, it is unable to find a polynomial time optimal algorithm. In order to find a global optimization algorithm, this dissertation presents a data assignment schema based on genetic algorithm. Not only does the new algorithm design effective encoding, fitness-evaluating methods and punishment function of constricted fitness-evaluating, but it also adopts the patching schema to alleviate the load of source server. The experiment result shows the algorithm's efficiency and is propitious to improve the stabilization and expansibility of the system.
Keywords/Search Tags:P2P Media Stream, Game Theory, Peer Selection, "Genetic Algorithm, Data Assignment
PDF Full Text Request
Related items