模糊化疑似网络入侵中行为确定的方法与设计

The Method and Design for Detecting and Judging Fuzzy Suspected Internet Intrusions

针对网络入侵判断过程出现模糊化、检测结果不理想的问题,提出一种引入多层模糊逻辑模型的模糊化疑似网络入侵行为的检测方法.先要保证每一个数据项的取值处在同等区间内,设计一种多层模糊入侵判断逻辑规则,以自适应和容错性作为约束条件,通过求解多层模型的反馈解,完成对疑似网络入侵行为的确定与检测.实验结果表明,采用改进算法进行模糊化疑似网络入侵行为的检测,能够提高检测的准确性,减小误报率,满足实际的检测需求.

This paper concerns wi th the issues of fuz z y judgement node and inaccurate detect ion results in the process of internet int rusion judgement. It proposes a method that introduces a mult i layered fuzzy logic model for detect ing suspected internet int rusions. Wi th this method, in order to complete the detect ion and conf irmation of suspected internation int rusions it is necessary to first ensure that the value of each data item of suspected int rusions fal ls into the same range; to design a type of rule based on mult i layered fuzzy judge-ment logic, constrained by the adaptabi l i ty and faul t tolerance of the judgement process; to rule out the problem of fuzzy threshold through solvinig the feedback solut ion of the mult i layered model. As demonst ra-ted by the exper iment resul ts, the detect ion that adopts this part icular algori thm would increase the accuracy of detect ion, reduce the rate of false positive, and fulfill the demand of pract ical test ing.：