论文部分内容阅读
针对传统的判定算法一直存在判定结果不准确,效率低的问题,提出基于宏块模式分布的多防区光纤网络入侵的模糊判定方法。通过投影寻踪的方法,确定所需方差投影指标、Luis O.Jinenez投影指标、离散度投影指标三种指标作为模糊判断基础,提取多防区光纤网络入侵数据特征,并以此为基础,采用宏块模式分布法,确定多防区光纤网络入侵数据阈值,实现多防区光纤网络入侵的快速模糊判定算法的改进。仿真实验证明,采用模糊的判定算法,相比传统的判定方法,其判定比降低了约45.5%,判定算法计算时间缩短了约11.08s,判定准确度提高了约26.9%,具有一定的优势。
Aiming at the problem of the inaccurate decision and inefficiency of the traditional decision algorithm, a fuzzy decision method of multi-zone optical network intrusion based on the macro-block mode distribution is proposed. By projection pursuit method, three indicators of variance projection, Luis O. Jinenez projection and dispersion projection are selected as the basis of fuzzy judgment to extract the intrusion data characteristics of multi-zone optical network. Based on this, Block pattern distribution method to determine the threshold of intrusion data in multi-zone optical network and improve the fast fuzzy decision algorithm of multi-zone optical network intrusion. Simulation results show that compared with the traditional method, the judgment ratio is reduced by 45.5% compared with the traditional method. The calculation time is reduced by 11.08s and the judgment accuracy is improved by 26.9%, which has some advantages.