云环境下基于动态聚类及相似树查询的无线体域网隐私数据检索算法研究

Research on Private Data Retrieval Algorithm in Cloud for WBAN Based on Dynamic Clustering and SS Tree

无线体域网收集的各项人体生物数据涉及隐私问题。大量的隐私数据存储在云服务器中,检索时要求高命中率和机密性。密文检索技术是解决云环境隐私安全问题的有效方法。针对此问题,论文提出了基于相似查询树的兄弟叶节点的查询结构——B-SS,以提升多关键字排序检索的结果命中率。在云存储的环境下提出改进的动态区间聚类算法MDB,在初始化文档集时,取文档集中最大和最小文档的向量差,等量的划分为k个槽,并对槽进行动态划分,聚类过程随文档量增加动态变化,且初始化时间复杂度为O(1),适用于无线体域网大数据环境下的密文检索。通过实验证明该方法随着文档的线性增加,消耗的时间呈线性变化,且变化幅度低,表明MDB算法在初始化效率上具有较大提升。

Information privacy for wireless body area network(WBAN)includes the user’s physiological parameters. A largeamount of privacy data is stored in the cloud server,which requires high hit rate and confidentiality. Ciphertext retrieval is an effec-tive method to solve the privacy problem of cloud environment. To solve this problem,this paper proposes a query structure calledbrotherhood similar query tree(B-SS)based on similar query tree to improve the hit ratio of multiple keyword sorting retrieval. Inthe environment of cloud storage,an improved dynamic interval clustering algorithm MDB is proposed,while documents clusteringinitialization,the differences between the maximum and minimum document vector are divided into k slots,and the size of each slotequals to hypersphere diameter. The clustering process increases the dynamic changes with the document,and the initialization timecomplexity is O(1),suitable for wireless body area network of large data environment ciphertext retrieval. Experiments show thatthe method is linear with the increase of document,and the time consumed is linear,and the range of change is low. It shows thatMDB algorithm is greatly improved in initialization efficiency.