具有小世界特性的语义覆盖网络模型  

Semantic Overlay Network Model with Small-world Features

作　　者：刘浩[1,2] 贺文华[2] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083 [2]湖南人文科技学院计算机科学技术系,湖南娄底417000

出　　处：《计算机工程》2012年第13期79-82,88,共5页Computer Engineering

基　　金：湖南省自然科学基金资助项目(11JJ3074);中南大学博士后科研基金资助项目;湖南人文科技学院高层次人才科研启动基金资助项目(2011RC03)

摘　　要：目前大多数的结构化覆盖网络模型忽略节点的聚类性和对称性,并且只能提供精确的资源查询机制。为此,给出一种具有小世界特性的语义覆盖网络模型。该模型采用显式分区机制、语义搜索链和组扩散机制。分析和实验结果表明,该模型比现有的覆盖网络模型Chord和CAN具有更好的网络性能,搜索效率更高,且具有较好的多匹配结果查询能力。The topology structure of P2P overlay network decides its system performance. However, the existing P2P overlay network models do not take clustering and symmetry of nodes into account, and support only precise resource locating scheme. A Semantic Overlay Network（SON） model with small-world features is proposed in the paper. It adopts the mechanism of distinctly grouping, semantic search edge and Group Flooding（GF）. Results of analysis and experiment show that this model provides better network performance and higher search efficiency than the existing overlay network models such as Chord and CAN. Furthermore, it has the querying capability of multi-matching results.

关 键 词：语义覆盖网络 CAYLEY图 聚类 对称性 语义搜索 分区 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]