摘要
基于多径路由协议,采用RS编码技术,设计了一种将数据信息编码后分片沿不同路径传输的方法,在目的节点对恶意篡改行为进行检测,确定恶意节点的具体位置或将恶意节点锁定在恶意节点组内。在此基础上提出了一个完整的抵制篡改数据攻击的信誉机制,充分利用了从数据传输和恶意节点检测过程中获得的信息,采用简单贝叶斯模型方法,对节点进行信誉评价,引入了恶意节点组的概念。数据在正常传输过程中无需重复加密和签名。实验表明,该信誉机制能够有机地与多径路由协议相结合,快速准确地孤立篡改数据的恶意节点。
Based on the multi-routing protocols and the reed-solomon error-correcting coding scheme, a data transmission method was presented, in which, round-encryption and round-signature were not needed. Data packets were all encoded by RS codes and cut into pieces before sending in different paths. After receive all the pieces of a data packet, the destination node could judge whether a malicious modification had been occurred, if so, the malicious node would be found out or be located in a malicious-node-group by a detecting process. Then by fully using the information gotten from detecting process, a reputation mechanism was proposed to resist the modification attacks. When implement it, the Bayesian approach was used to evaluate a node's reputation. The simulation results show that it can react to the attacks more quickly and accurately.
出处
《通信学报》
EI
CSCD
北大核心
2008年第5期56-64,共9页
Journal on Communications
基金
国家自然科学基金资助项目(60403027)~~
关键词
AD
HOC网络
信誉机制
内部攻击
恶意节点组
ad hoc networks
reputation mechanism
internal attacks
malicious-node-group