基于兴趣域的P2P气象资料搜索机制研究

Research on Searching Meteorological Data Mechanism Based on Interest in P2P Network

P2P技术有助于解决目前气象资料传输效率不高的问题,如何高效地搜索资源是P2P网络中最为关键的问题。文中为了在P2P环境中实现气象资料的更快更精确搜索,结合气象资料特点,提出了一种基于兴趣域的P2P气象资料搜索机制。通过该方法建立气象元数据向量空间模型,计算向量空间模型相似度来划分不同兴趣域,从而提高搜索成功率。实验结果表明该方法只将查询消息路由到相关兴趣的节点上,缩短了平均搜索路径,提高了查询效率,高于一般洪泛算法。

P2P technology is helpful to solve the problem of low transport efficiency. How to search the resources efficiently is the key problem in peer-to-peer networks. In order to search meteorological data faster and more accurate in P2P environment, according to the character of meteorological data, propose an efficient scheme for searching meteorological data based on interest-domain. Firstly, this method builds vector space model for meteorological metadata. After that, vector space proximity is computed to build different interest domain, thus improves the search success rate. Experimental results show that in this method information search is routed to peers of rele- vant interests, reduce average routing distance, rise query success rate and the efficiency is higher than the typical search mechanism.