摘要
针对当前信息存储方法存在耗时长和安全性差的问题,提出多形态模式的信息存储非线性重构方法。依据具有IP可跟踪性的安全检测方法,实时记录多形态模式信息结构,相隔额定周期开展一次安全检测。利用EWMA预测模型预测各周期信息值,实现信息安全检测。采用主成分分析法将原始信息由高维变量空间转换至低维特征空间,实现多形态模式信息降维。引入非线性动力机制,对多形态模式信息时间序列进行相空间重构,并依据密集度刻画不同信息在相空间中存在的差异性,基于信息安全检测与降维,结合支持向量机分类器构建信息存储模型,实现不断更新的多形态模式信息分类存储。实验结果表明,该方法信息存储效率高,且具有安全性,与当前相关方法比较性能更强。
Aiming at the problems of long time consuming and poor security in the current information storage methods, a multi-morphological information storage nonlinear reconstruction method is proposed. According to the security detection method with IP traceability, the multi-mode information structure is recorded in real tim e, and the security detection is carried out once every rated period. The EWMA prediction model is used to predict the information values of each cycle, and the information security detection is realized. The principal component analysis ( PCA) is used to transform the original information from the high-dimensional variable space to the low-dimensional feature space to reduce the dimensionality of the multi-morphological pattern information. A nonlinear dynamic mechanism is introduced to reconstruct the phase space of the time series of multi-morphological information, and to depict it according to the intensity. Based on the information security detection and dimensionality reduction, the information storage model is constructed by combining support vector machine classifier to realize the updated multi-morphological information classification storage. Simulation results show that this method has high information storage efficiency and security, and better performance compared with the current related methods.
作者
仲崇丽
ZHONG Chong-li(Eurasia University, School of Finance, Shanxi Xi'an 710065 , China)
出处
《计算机仿真》
北大核心
2019年第8期463-466,471,共5页
Computer Simulation
关键词
多形态模式
信息存储
非线性重构
Polymorphic pattern
Information storage
Nonlinear reconstruction
