论文快速发表网

社科类论文科技类论文医学类论文管理类论文教育类论文农林类论文新闻类论文建筑类论文文艺类论文法学类论文

EI Compendex Source List（2022年1月） EI Compendex Source List（2020年1月） EI Compendex Source List（2019年5月） EI Compendex Source List（2018年9月） EI Compendex Source List（2018年5月） EI Compendex Source List（2018年1月）中国科学引文数据库来源期刊列 CSSCI(2017-2018)及扩展期刊目录 2017年4月7日EI检索目录（最新） 2017年3月EI检索目录最新公布北大中文核心期刊目录 SCI期刊（含影响因子）

本站服务项目浅谈青年教师如何开展教科研大学青年教师如何处理教学和科

论文范文

Secure Deduplication Based on Rabin Fingerprinting over Wireless Sensing Data in Cloud Computing 时间:2018-09-08 21:48 来源:未知作者:admin 点击: 次 Abstract:The rapid advancements in the Internet of Things (IoT) and cloud computing technologies have significantly promoted the collection and sharing of various data. In order to reduce the communication cost and the storage overhead, it is necessary to exploit data deduplication mechanisms. However, existing data deduplication technologies still suffer security and efficiency drawbacks. In this paper, we propose two secure data deduplication schemes based on Rabin fingerprinting over wireless sensing data in cloud computing. The first scheme is based on deterministic tags and the other one adopts random tags. The proposed schemes realize data deduplication before the data is outsourced to the cloud storage server, and hence both the communication cost and the computation cost are reduced. In particular, variable-size block-level deduplication is enabled based on the technique of Rabin fingerprinting which generates data blocks based on the content of the data. Before outsourcing data to the cloud, users encrypt the data based on convergent encryption technologies, which protects the data from being accessed by unauthorized users. Our security analysis shows that the proposed schemes are secure against offline brute-force dictionary attacks. In addition, the random tag makes the second scheme more reliable. Extensive experimental results indicate that the proposed data deduplication schemes are efficient in terms of the deduplication rate, the system operation time, and the tag generation time.
1. Introduction
The wireless sensor network (WSN) is an ad hoc network composed of a large number of sensors, and the sensors communicate with each other over a wireless channel in a multihop manner [1–5]. Sensors are usually a low-cost, simple device with limited computing power and working batteries, which have the ability to collect, process, and transfer data. With the rapid development of Internet of Things (IoT) and cloud computing technologies, WSN has found many promising applications. As an extension to the cloud computing paradigm, fog computing makes it possible to execute the IoT applications in the network of edge. Xu et al. [6] proposed a dynamic resource allocation method for load balancing in fog environment. Cloud computing [7, 8] supports distributed data storage and parallel processing and its data processing framework handles huge amounts of data in a local computer rather than requiring to transmit these data remotely [9–11]. We know that cloud storage technology is the most common and most popular cloud computing service today. The extensive application of cloud storage motivates enterprises and organizations to outsource data storage to third-party cloud providers [12–16]. Zhang et al. [17] proposed a fine-grained access control system suitable for resource-constrained users in cloud computing. It is reported that the average size of backup data for a medium size enterprise is 285 TB and faces an annual growth rate of about 24-27%. According to the analysis report of IDC, personal user data has reached terabytes in 2006. From 2006 to 2010, global data volume continues to grow at a rate of 57% annually. In 2011, the global data volume has entered the era of ZB, and the total amount of data used globally exceeds 1.8 ZB. It is expected that the global data volume will reach 40 ZB by 2020 [18].
Data deduplication has been widely accepted as an effective technique to reduce workload and overhead of the cloud storage system [19–23]. Today’s commercial cloud storage services, such as Dropbox, Google Drive, Bitcasa, Mozy, and Memopal, have been applied deduplication to save maintenance cost. However, the extensive application of data deduplication makes its security problems increasingly prominent [24, 25]. Compared with traditional information security, cloud storage security [26–28] mainly has two characteristics: users do not enjoy physical control over the data they upload to the cloud storage system and the same kind of physical resources is shared by multiple users. The confidentiality and integrity of data will be threatened. It is noted that cloud storage security has drawn many attentions [29, 30]. Xu et al. [31] proposed a cost and energy aware data placement method for privacy-aware applications over big data in hybrid cloud. Harnik et al. [32] pointed out that there were security vulnerabilities in the deduplication technology used by the provider. Douceur et al. [33] introduced convergent encryption (CE) that uses the hash value of the data itself as a secret key to solve the problem of contradiction between deduplication and confidentiality. Bellare et al. [34] defined a cryptographic primitive called message-locked encryption. Li et al. [35] implemented Dekey using the Ramp secret sharing scheme to manage the CE keys. Literature [21] pointed out that, in the data deduplication, simply using the hash value of the file represents the entire file, making the data deduplication process vulnerable to hacking, and the hash value is not confidential, and the attacker can obtain the entire file content by obtaining the hash value. Abadi et al. [36] proposed two schemes, including a completely random scheme and a deterministic scheme, which support the randomization of tags to ensure the security of the data deduplication system. In the schemes, CE directly uses the data fingerprint as the key derivation function and hence only achieves security for unpredictable data. In fact, offline brute-force dictionary attacks can be easily launched because of the determination of CE keys [37]. Moreover, current deduplication schemes [35, 37] directly deduplicate the encrypted data, which increases the computational overhead. In the future, it is possible to realize decentralized data deduplication schemes via blockchain technologies, which have been used to realize decentralized outsourcing computation [38, 39] and searchable encryption with two-side verifiability [40] in cloud computing.