面向云数据安全存储的分段融合模糊聚类算法

Segmented Fusion Fuzzy Clustering Algorithm for Cloud Data Security Storage

为了提高云数据的安全存储性能,需要对数据进行优化属性聚类归集。针对传统方法采用模糊C均值聚类进行云数据存储归类设计具有对初始聚类中心敏感、容易陷入局部收敛的问题,提出一种基于分段融合模糊聚类的云数据安全存储模型构建方法。建立云数据安全存储的网格分布结构模型并进行数据结构分析,进行云数据属性集的向量量化特征分解,对海量的云存储数据流采用分段匹配检测方法进行特征压缩,实现冗余数据自适应归集合并,挖掘云数据信息流的高阶谱特征。在模糊C均值聚类算法的基础上采用分段数据融合进行数据分簇模糊聚类,提高数据存储的安全性,同时降低云数据存储的负荷。仿真结果表明,采用该方法进行云数据聚类和优化存储设计,能降低数据聚类的误分率,提高云数据存储的吞吐量,确保云数据的安全存储。

In order to improve the safety performance of cloud data storage, the collection of attribute clustering data need to be optimized. Since the traditional method which uses fuzzy C means clustering classification of cloud data stora- ge design is sensitive to initial clustering center and is easy to fall into the local convergence, a method of constructing the cloud data security storage model was proposed based on segmentation fusion and fuzzy clustering. The data struc- ture analysis of distribution grid structure model is given to build cloud data security storage and decomposition of vec- tor quantization characteristics of cloud data attributes, cloud storage data on mass flow uses piecewise matching feature detection method to realize adaptive compression, redundant data collection and mining are realized, high order spectrum of cloud data stream is mined. Based on the fuzzy C means clustering algorithm, the data clustering fuzzy clustering is used to improve the security of data storage and reduce the load of cloud data storage. The simulation results show that the proposed method can reduce the error rate of data clustering and improve the throughput of data storage, and ensure the security of data storage.