非结构化P2P网络引导型进化博弈算法

Guided evolutionary game algorithm of unstructured P2P network

为促进动态开放性对等网络中节点间的合作,在SLACER(selfish link-based adaptation for cooperation excluding rewiring,基于自私连接排除重构的自适应合作)算法的基础上引入标兵节点,提出了引导型进化博弈算法G-SLACER(guided-SLACER)。通过初始化,网络节点总数的30%为标兵节点;拓扑重构过程中,新增一条到最具优势节点的引导型连接;为鼓励节点相互学习,加大网络整体收益。实验结果表明,G-SLACER算法针对不同规模的网络均具有良好的通用性,网络中CCP(cooperative connected path,合作连接路径)的稳定性增强。与其他进化博弈算法相比,G-SLACER算法形成的P2P网络的合作状态出现得更早、更平稳。

In order to promote the cooperation among the nodes which exist in dynamic and open peer-to-peer network, G-SLACER algorithm was provided by introducing pacesetter nodes. 30% of network nodes were initialized to pacesetter nodes. In the process of topology reconstruction, a guided link to the most advantage node was added. To encourage studies between nodes, the payoff of the whole network was increased. The experimental results show that the G-SLACER algorithm has good generality for different sizes of networks, and it enhances the stability of CCP. Compared with other evolutionary game algorithms, cooperation state of P2P network formed by G-SLACER algorithm appears earlier and more stable.