摘要
为了解决由于P2P开放、匿名和高度动态的特性而容易受到攻击并被攻击者用来散布恶意信息的问题,需要建立P2P节点间的信任关系,提出一种TPP(trust in peer topeer)方案。该方案中每个节点通过计算被查询节点信任值的方式,使用信任向量建立本地信任表,并提交对另外节点的评价以建立全局可信表,最终建立一个信任网络。模拟结果表明,TPP比其他模型的交易成功率高,而通信和计算资源开销小,且能够很好地解决冒名、协同作弊以及"搭车行为"等安全问题。通过建立TPP模型,P2P网络有更强的健壮性和可扩展性,安全性提高,易于建立更加可信的网络。
Peer-to-Peer networks are open, anonymous, and highly dynamic systems. Hackers can easily attack such systems and use them to spread malicious information after invading the system. One approach to address these problems is to establish trust relations with peers in the P2P network. This paper presents a Trust-in-Peer-to-Peer model in which each peer calculates the trust value of queried peers to build a local trust table using a trust vector. Each peer can also submit opinions about other peers to construct a global credibility table which will lead to a trust network. Simulation results show that the Trust-in Peer-to-Peer model gives higher success ratios than other models with lower communication and computational overheads. The model also prevents the Sybil attack, malicious collectives attack, and free riders. This makes the P2P network more robust and extensible and increases security be establishing a more credible network.
出处
《清华大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第7期1224-1228,共5页
Journal of Tsinghua University(Science and Technology)
基金
国家自然科学基金资助项目(60473082)
中国工程物理研究院面上基金资助项目(20050658)
关键词
计算机网络
对等网络
全局可信表
信任向量
computer network
peer to peer （P2P）
global credible table
trust vector
