摘要
文章在分布式环境中,按照系统和网络的异常使用模式的不同特征和环境差异,分别引入了分布处理、分层管理、自治代理和安全通讯,仿效人体对于外界刺激响应模式的思想,提出了分布式入侵检测系统框架和自治代理组织单元,给出了基于自治代理的分布入侵检测系统体系结构与详细设计。并引入了贝叶斯推理反馈信息的统计方法,对系统检测结果进行动态优化,提高了检测效率和准确性。
In this article,the approaches of distributed processing,hierarchical management ,autonomous agent and security communication are adopted,according to the distinct features and environment diversity of the abnormal usage of system and network,in the distributing type environment ;The framework of distributed intrusion detection system and autonomous agent constructor cell are put forward,emulating how the body answer the provoke coming from the outside world,and the architecture and the detail of the IDS are gave.Moreover the Bayersian reasoning feedback information statistics means are adopted to carry on the dynamic optimization to results of the system detection,thus examination effectiveness and accuracy are lifted.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第24期30-32,共3页
Computer Engineering and Applications
基金
国家863高技术研究发展计划基金资助(编号:863-317-01-10-99)
全国教育科学"十五"规划重点课题资助(编号:AYA010034)
湖北省重点自然科学基金资助(编号:2001ABA001)
关键词
入侵检测系统
自治代理
分布式
贝叶斯推理
Intrusion detection system,Agent ,Distributing type,Bayersian reasoning