摘要
针对传统flooding搜索算法面临的效率低下和网络流量过大等问题,提出了一种改进的基于兴趣和蚁群相结合的资源搜索算法(IASA)。该算法将TTL机制加以改进,并将兴趣相似度引入到蚁群算法的路径概率选择中,同时结合信息素的多样性和正反馈机制,积累历史搜索经验,获得路由指引信息,从而有效地指导查询请求消息的转发,将请求消息尽量发往资源可能存在的节点上。仿真实验表明:该算法能有效地指导资源搜索的方向,提高查询命中率,减少冗余消息包,其整体搜索效果较好。
In unstructured P2P network, the traditional flooding search algorithm suffers from some disadvantages of inefficiencies and ex- cessive network traffic. To address the problem,present a resource search algorithm based on the combination of interest and ant colony （ IASA ）. In this algorithm, the TTL mechanism was improved and interest similarity was introduced into the choice of the rooting path in ant colony algorithm, and combining with the diversity of the pheromones, positive feedback mechanism was conductive to accumulate history experience and get routing guiding information, so as to effectively guide query information＇s forwarding, so that the query infor- mation can be sent to the proper nodes with requested resources as much as possible. Simulation results indicated that the algorithm can ef- fectively guide the search direction and improve inquires and reduce redundancy information. On the whole, the search effect performs better.
出处
《计算机技术与发展》
2012年第7期67-70,74,共5页
Computer Technology and Development
基金
国家自然科学基金资助项目(90612003)
山东省自然科学基金资助项目(Y2007G11)
山东大学高校院所自主创新项目(2010040072)
关键词
P2P
兴趣相似度
蚁群算法
信息素
路径选择概率
P2P
interest similarity
ant colony algorithm
pheromone
path selection probability
