FCAN: A Structured P2P System Based on Content Query

Abstract

The existing DHT-based file storage systems are suffering problems of being restricted to exact match and lacking semantic-based query. To solve these problems, a DHT-based file storage system named FCAN is proposed. Mapping a document vector from semantic space to P2P space by FastMap, FCAN can effectively reduce the dimensions of it to implement semantic-based query without affecting the semantic integrality of this document. By clustering the nearest neighbors based on centroid, FCAN is further improved to reduce the redundant messages and search time. Simulations show that FCAN has not only higher recall and precision, but also fewer search time and redundant messages