A NOVEL EFFICIENT REMOTE DATA POSSESSIONCHECKING PROTOCOL IN CLOUD STORAGE

 

ABSTRACT

As an important application in cloud computing,cloud storage offers user scalable, flexible and high quality datastorage and computation services. A growing number of dataowners choose to outsource data files to the cloud. Because cloudstorage servers are not fully trustworthy, data owners needdependable means to check the possession for their filesoutsourced to remote cloud servers. To address this crucialproblem, some remote data possession checking (RDPC) protocolshave been presented. But many existing schemes havevulnerabilities in efficiency or data dynamics. In this paper, weprovide a new efficient RDPC protocol based on homomorphichash function. The new scheme is provably secure against forgeryattack, replace attack and replay attack based on a typicalsecurity model. To support data dynamics, an operation recordtable (ORT) is introduced to track operations on file blocks. Wefurther give a new optimized implementation for the ORT whichmakes the cost of accessing ORT nearly constant. Moreover, wemake the comprehensive performance analysis which shows thatour scheme has advantages in computation and communicationcosts. Prototype implementation and experiments exhibit that thescheme is feasible for real applications.

EXISTING SYSTEM:

The first RDPC was proposed by Deswarte et al. basedon RSA hash function. The drawback of this scheme is that itneeds to access the entire file blocks for each challenge. In 2007,the provable data possession (PDP) model was presented byAteniese et al. , which used the probabilistic prooftechnique for remote data integrity checking without accessingthe whole file. In addition, they supplied two concrete schemes(S-PDP, E-PDP) based on RSA. Although these two protocolshad good performance, it’s a pity they didn’t support dynamicoperations. To overcome this shortcoming, in 2008, theypresented a dynamic PDP scheme by using symmetricencryption . Nonetheless, this scheme still did not supportblock insert operation. At the same time, lots of research works devoted to construct fully dynamic PDP protocols. Forinstance, Sebé et al.  provided a RDPC protocol for criticalinformation infrastructures based on the problem to factor largeintegers, which is easily adapted to support data dynamics.Erway et al. first presented a fully dynamic PDP scheme(DPDP) by using authenticated skip list, which allowed dataowner to append, delete, insert and update file blocks atanytime. Wang et al. used Merkle hash tree (MHT) topropose another dynamic method for remote data checking, inwhich each block was hashed to be a leaf node of MHT. Bysorting all leaf nodes from left to right, the MHT implicitlyidentified the block position which is essential for dynamicoperations. However, using MHT caused heavy computationcost. In 2013

PROPOSED SYSTEM:

It is essential for data owners to verify the integrity for thedata stored on CSS before using it. For example, a biginternational trading company stores all the imports and exportsrecord files on CSS. According to these files, the company canget the key information such as the logistics quantity, the tradevolume etc. If any record file is discarded or tampered, thecompany will suffer from a big loss which may cause badinfluence on its business and development. To avoid this kindof circumstances, it is mandatory to check the integrity foroutsourced data files. Furthermore, since these files may referto business secret, any information exposure is unacceptable. Ifthe company competitor can execute the file integrity checking,by frequently checking the files they may obtain some usefulinformation such as when the file changes, the growth rate ofthe file etc, by which they can guess the development of thecompany. Thus, to avoid this situation, we consider the privateverification type in our scheme, that is, the data owner is theunique verifier. In fact, the current research direction of RDPCfocuses on the public verification, in which anyone can performthe task of file integrity checking with the system public key.Although RDPC with public verification seems better than thatwith private verification, but it is unsuitable to the scenariomentioned above.Motivated by the above application scenarios, we present anovel efficient RDPC scheme by using homomorphic hashfunction , which has been used to construct RDPC schemes. Unfortunately, these schemes are either insecure or notefficient enough. To overcome these drawbacks, we refer to theidea of  and introduce a private key for each tag generationin our RDPC scheme. Simultaneously, a new construction ofORT is presented for data dynamic which can improve theefficiency of the protocol greatly. Compared with the previousones, our scheme has better performance in term ofcomputation and communication. Our contributions aresummarized as follows:We present a novel efficient RDPC scheme with datafunction technique, in which the hash value of the sum for twoblocks is equal to the product for two hash values of thecorresponding blocks. We introduce a linear table called ORTto record data operations for supporting data dynamics such asblock modification, block insertion and block deletion. Toimprove the efficiency for accessing ORT, we make use ofdoubly linked list and array to present an optimizedimplementation of ORT which reduces the cost to nearlyconstant level. We prove the presented scheme is secure againstforgery attack, replay attack and replace attack based on atypical security model. At last we implement our scheme andmake thorough comparison with previous schemes. Experimentresults show that the new scheme has better performance and ispractical for real applications.

CONCLUSION:

In this paper, we study the issue for integrity checking of datafiles outsourced to remote server and propose an efficientsecure RDPC protocol with data dynamic. Our scheme employsa homomorphic hash function to verify the integrity for the filesstored on remote server, and reduces the storage costs andcomputation costs of the data owner. We design a newlightweight hybrid data structure to support dynamic operationson blocks which incurs minimum computation costs bydecreasing the number of node shifting. Using our new datastructure, the data owner can perform insert, modify or deleteoperation on file blocks with high efficiency. The presentedscheme is proved secure in existing security model. Weevaluate the performance in term of community cost,computation cost and storage cost. The experiments resultsindicate that our scheme is practical in cloud storage.

REFERENCES

[1] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, and I. Brandic,“Cloudcomputing and emerging IT platforms: Vision, hype, and reality fordelivering computing as the 5th utility,” Future Gener. Comp. Sy., vol.25, no. 6, pp. 599 – 616, 2009.

[2] H. Qian, J. Li, Y. Zhang and J. Han, “Privacy preserving personal healthrecord using multi-authority attribute-based encryption withrevocation,” Int. J. Inf. Secur., vol. 14, no. 6, pp. 487-497, 2015.

[3] J. Li, W. Yao, Y. Zhang, H. Qian and J. Han, “Flexible and fine-grainedattribute-based data storage in cloud computing,” IEEE Trans. ServiceComput., DOI: 10.1109/TSC.2016.2520932.

[4] J. Li, X. Lin, Y. Zhang and J. Han, “KSF-OABE: outsourcedattribute-based encryption with keyword search function for cloudstorage,” IEEE Trans. Service Comput., DOI: 10.1109/TSC.2016.2542813.

[5] J. Li, Y. Shi and Y. Zhang, “Searchable ciphertext-policyattribute-based encryption with revocation in cloud storage,” Int. J.Commun. Syst., DOI: 10.1002/dac.2942.

[6] J.G. Han, W. Susilo, Y. Mu and J. Yan, “Privacy-Preserving DecentralizedDKiesytr-iPboultiecdy SAystttreimbust,e v-Bola. s2e3d, nEon.1c1ry, pptpio. n2,1”5 0IE-2E1E62 ,T 2r0an1s2a ctions on Parallel and

[7] Z. J. Fu, X. M. Sun, Q. Liu, L. Zhou, and J. G. Shu, “Achieving efficientcloud search services: multi-keyword ranked search over encryptedcloud data supporting parallel computing,” IEICE Transactions onCommunications, vol. E98-B, no. 1, pp.190-200, 2015.

[8] Z. J. Fu, K. Ren, J. G. Shu, X. M. Sun, and F. X. Huang, “Enablingpersonalized search over encrypted outsourced data with efficiencyimprovement,” IEEE Transactions on Parallel and Distributed Systems,DOI: 10.1109/TPDS.2015.2506573, 2015.

[9] Z. H. Xia, X. H. Wang, X. M. Sun, and Q. Wang, “A secure anddynamic multi-keyword ranked search scheme over encrypted clouddata,” IEEE Transactions on Parallel and Distributed Systems, vol. 27,no. 2, pp. 340-352, 2015.

[10] Y. J. Ren, J. Shen, J. Wang, J. Han and S. Y. Lee, “Mutual verifiableprovable data auditing in public cloud storage,” Journal of InternetTechnology, vol. 16, no. 2, pp. 317-323, 2015.