基于P2P泛洪DDoS攻击的防范研究  

Defending P2P from flooding-based DDoS

作　　者：耿技[1] 马新新[1] 

机构地区：[1]电子科技大学计算机科学与工程学院,成都610054

出　　处：《计算机应用研究》2009年第6期2116-2118,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(60473090)

摘　　要：提出了基于马尔可夫过程的信任和信誉模型,在节点间构建信任关系,利用节点间信任与信誉信息的交互对恶意节点进行识别,阻断对恶意消息的转发传播,从而增强抵御DDoS攻击的效能。仿真实验结果表明,提出的模型能有效地隔离恶意节点的消息数,提高网络抵御这种基于应用层的DDoS攻击的容忍度。This paper proposed a trust and reputation model based on Markov process, which constructed trust relationship among peers. In this model, well-behavior peers possessed ability to identify bad-behavior peers to intermit transmission of malicious message through the interaction of trust and reputation information established among peers, so that the networks could resist on the DDoS attack. Simulation result shows that the model of provides an efficient isolation on malicious message amount and promotion on tolerance of the DDoS attack resisting on the application layer.

关 键 词：对等网 GNUTELLA协议 分布式拒绝服务攻击 可信和信誉 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]