中文摘要：

结构化对等（P2P）网络是基于文件标识符的分布式系统．它能够很好地支持对文件标识符的等值查询，但对于其他的复杂查询，则需要进行进一步扩展．现有的一些技术虽然在结构化对等网络实现了关键词搜索，但是仍然存在诸如负载不均衡，结构复杂，不易于部署和效率低等问题．对此提出了一种基于结构化对等网络的关键词搜索新方法，将构建在一般分布式哈希表上的倒排索引与top-k查询算法结合起来，以较少的代价实现关键词搜索，同时有效地克服了数据分布不均匀带来的负载不均衡问题．

英文摘要：

Structured Peer-to-Peer networks are based on hash functions, which can support equality queries very well but can hardly to support complex queries, particularly keyword search. Existing techniques for keyword search in structured P2P networks suffer from several problems, such as load imbalance, complicate structure, unease of deployment and inefficiency. To address this issue, in this paper, we combine DHT-based inverted indexes with a top-k query algorithm and propose a framework, which can achieve efficient query processing performance as well as load balance issues.