面向泛在电力物联网云端数据的轻型动态完整性审计方案

Service provider is Honest-But-Curious in a cloud data storage mode. Therefore, data owners usually perform static or dynamic integrity auditing on encrypted data for guaranteeing the integrity of the outsourced data. However, most of the proposed methods are inapplicable to the integrity auditing of cloud data in ubiquitous power internet of things(UPIoT). In the case of high data acquisition frequency and fast update speed, there are certain limitations in data retrieval efficiency and dynamic update. In this paper, to solve above problems, we present an authentication data structure based on lightweight hierarchical MHT, which combines an existing secure encryption method and locality sensitive hash(LSH) technology. First of all, the hierarchical MHT is composed of 16 smart meters and divided into four MHT subtrees. Their root nodes as leaf nodes constitute a new MHT tree. And each node in the tree is a data information storage container, which stores the relevant nodes information. Besides, nodes in the tree are given six definitions, namely Definition 1(node information), Definition 2(smart meter identifier), Definition 3(local root node identifier), Definition 4(dynamic operation identifier), Definition 5(node hash) and Definition 6(authentication certificate). These innovations on nodes improve node utilization and shorten the authentication path length. Secondly, the data is encrypted before uploaded to the cloud. Because the smart meter in UPIoT collects electricity consumption data every 15 minutes, a time stamp is added into the smart meter data. This encryption method ensures the privacy of the smart meter data in the cloud. Then, the data is stored to the cloud by LSH technology. The cloud constructs I hash buckets, and I=i(where i is the number of MHT subtrees). According to the smart meter identifier SM_ij, LSH maps the data that belongs to the same region into the same hash bucket. LSH technology improves data update efficiently and retrieves data quickly. Finally, the cloud data is given to three algorithms for dynamic operation, namely Algorithm 1(data modification algorithm), Algorithm 2(data deletion algorithm) and Algorithm 3(data insertion algorithm). In the experimental part, we compared our scheme with MHT scheme(2010), SL scheme(2015), and PMHT scheme(2019). And we performed a series of experiments based on 126M real smart meter data sets. Under the condition that the fixed data size was unchanged and the number of smart meters was changed, we conducted comparisons of communication overhead, storage overhead, calculation overhead, and update efficiency. In addition, under the condition that the number of fixed smart meters was unchanged and the data size was changed, we conducted comparisons of auditing efficiency and calculation overhead. The experimental results further demonstrate that our work in this paper can effectively lessen the calculation and communication overhead, and support efficient dynamic data update. Our scheme realizes a lightweight dynamic integrity auditing of cloud data in UPIoT.