Research on resource discovery method for networks with unstructured peer model

Abstract

Unstructured peer model is widely used in many fields, but the common blind search method, such as flooding, generates lots of redundant messages. The limited bandwidth can easily cause network congestion. For the existing problems in traditional discovery method, a resource discovery method based on peer interest similarity is put forward. Based on grouping, similar peers are divided into domain on the basis of peer interest similarity. Query message is forwarded in the same group of the present domain at first, and network topology is adjusted dynamically during searching. Then, route hops are reduced and the amount of redundant messages is decreased effectively. Simulation experiment analyzes and confirms the searching efficiency of the method.