FAST PHRASE SEARCH FOR ENCRYPTED CLOUD STORAGE
ABSTRACT
Cloud computing has generated much interest in the research community in recent years for its many advantages, but hasalso raise security and privacy concerns. The storage and access of confidential documents have been identified as one of the centralproblems in the area. In particular, many researchers investigated solutions to search over encrypted documents stored on remotecloud servers. While many schemes have been proposed to perform conjunctive keyword search, less attention has been noted onmore specialized searching techniques. In this paper, we present a phrase search technique based on Bloom filters that is significantlyfaster than existing solutions, with similar or better storage and communication cost. Our technique uses a series of n-gram filters tosupport the functionality. The scheme exhibits a trade-off between storage and false positive rate, and is adaptable to defend againstinclusion-relation attacks. A design approach based on an application’s target false positive rate is also described
EXISTING SYSTEM:
Boneh et al.’s work on an encrypted keyword searchscheme based on public key encryption was a the mostcited in the area. The author considered a scenario where auser wishes to have an email server verify messages associatedwith certain keywords without revealing the content ofthe emails. As sample application, the scheme would allowan urgent encrypted email to be flagged to the attention of auser while others sent to appropriate folders. The proposedsolution uses identity based encryption and a variant usingbilinear mapping. Another interesting application was proposedby regarding searching through encrypted auditlogs, where only relevant logs are retrieved. The scenarioinvolves an auditor which acts as a key escrow authorizinginvestigators to search audit records. The scheme uses anextension of Boneh’s scheme using identity based encryption.Song et al. also considered the scenario introducedby Boneh et al. and proposed a probabilistic search solutionbased on stream cipher.Many recent works have focused on conjunctive keywordsearch. Ding et al. [3] extended Boneh et al.’s schemeusing bilinear mapping to perform multiple keyword searchand described a solution that did not include expensivepairing operations in the encryption and trapdoor generationphase. Kerschbaum et al. considered the searchof unstructured text, where positions of keywords are unknown.The use of encrypted index for keyword searchwas examined in and a scheme secure against chosenkeyword attack was proposed. The ranking of searchresults was looked at by Wang et al. in. The authorsdescribed a solution based on the commonly used TFIDF(Term Frequency x Inverse Document Frequency) ruleand the use of order preserving symmetric encryption. Liuet al. considered the search for potentially erroneouskeywords termed fuzzy keyword search. The index-basedsolution makes use of fuzzy dictionaries containing variousmisspelling of keywords including wildcards.
PROPOSED SYSTEM:
In this paper, we present a phrase search scheme whichachieves a much faster response time than existing solutions.The scheme is also scalable, where documents caneasily be removed and added to the corpus.We also describemodifications to the scheme to lower storage cost at a smallcost in response time and to defend against cloud providerswith statistical knowledge on stored data. We begin bypresenting the communication framework in section 2 andvarious backgrounds including related works in section 3.Although phrase searches are processed independently usingour technique, they are typically a specialized functionin a keyword search scheme, where the primary function isto provide conjunctive keyword searches.Our framework differs from some of the earlier works, where keywords generally consist of meta-datarather than content of the files and where a trusted keyescrow authority is used due to the use of Identity basedencryption. When compared to recent works, our setupis equivalent to that of , where an organizationwishes to outsource computing resources to a cloud storageprovider and enable search for its employees, and similarto, where the aim is to return properly ranked files.Most other recent works related to search over encrypteddata have considered similar models such as where theclient acts as both data owner and user.Note that, depending on the application, the encrypteddocuments may or may not require retrieval once the queryis resolved. Should retrieval be required, further privacyissues may arise. These issues are considered in obliviousstorage and private information retrieval schemes .Our discussions will mainly restrict to the protocol leadingto the query resolution. Direct retrieval is assumed whereappropriate to better compare against existing solutions forphrase search.
CONCLUSION
In this paper, we presented a phrase search scheme basedon Bloom filter that is significantly faster than existingapproaches, requiring only a single round of communicationand Bloom filter verifications. The solution addresses thehigh computational cost noted in by reformulatingphrase search as n-gram verification rather than a locationsearch or a sequential chain verification. Unlike, our schemes consider only the existence of a phrase,omitting any information of its location. Unlike, ourschemes do not require sequential verification, is parallelizableand has a practical storage requirement. Our approachis also the first to effectively allow phrase search to run independentlywithout first performing a conjunctive keywordsearch to identify candidate documents. The technique ofconstructing a Bloom filter index introduced in section 4.2enables fast verification of Bloom filters in the same manneras indexing. According to our experiment, it also achievesa lower storage cost than all existing solutions except,where a higher computational cost was exchanged in favorof lower storage. While exhibiting similar communicationcost to leading existing solutions, the proposed solutioncan also be adjusted to achieve maximum speed or highspeed with a reasonable storage cost depending on theapplication. An approach is also described to adapt thescheme to defend against inclusion-relation attacks. Variousissues on security and efficiency, such as the effect of longphrases and precision rate, were also discussed to supportour design choices.
REFERENCES
[1] D. Boneh, G. D. Crescenzo, R. Ostrovsky, and G. Persiano, “Publickey encryption with keyword search,” in In proceedings of Eurocrypt,2004, pp. 506–522.
[2] B. Waters, D. Balfanz, G. Durfee, and D. K. Smetters, “Buildingan encrypted and searchable audit log,” in Network and DistributedSystem Security Symposium, 2004.
[3] M. Ding, F. Gao, Z. Jin, and H. Zhang, “An efficient public keyencryption with conjunctive keyword search scheme based onpairings,” in IEEE International Conference onNetwork Infrastructureand Digital Content, 2012, pp. 526–530.
[4] F. Kerschbaum, “Secure conjunctive keyword searches for unstructuredtext,” in International Conference on Network and SystemSecurity, 2011, pp. 285–289.[5] C. Hu and P. Liu, “Public key encryption with ranked multikeywordsearch,” in International Conference on Intelligent Networkingand Collaborative Systems, 2013, pp. 109–113.
[6] Z. Fu, X. Sun, N. Linge, and L. Zhou, “Achieving effective cloudsearch services: multi-keyword ranked search over encryptedcloud data supporting synonym query,” IEEE Transactions onConsumer Electronics, vol. 60, pp. 164–172, 2014.
[7] C. L. A. Clarke, G. V. Cormack, and E. A. Tudhope, “Relevanceranking for one to three term queries,” Information Processing andManagement: an International Journal, vol. 36, no. 2, pp. 291–311,Jan. 2000.
[8] H. Tuo and M. Wenping, “An effective fuzzy keyword searchscheme in cloud computing,” in International Conference on IntelligentNetworking and Collaborative Systems, 2013, pp. 786–789.
[9] M. Zheng and H. Zhou, “An efficient attack on a fuzzy keywordsearch scheme over encrypted data,” in International Conference onHigh Performance Computing and Communications and Embedded andUbiquitous Computing, 2013, pp. 1647–1651.
[10] S. Zittrower and C. C. Zou, “Encrypted phrase searching in thecloud,” in IEEE Global Communications Conference, 2012, pp. 764–770.
[11] Y. Tang, D. Gu, N. Ding, and H. Lu, “Phrase search over encrypteddata with symmetric encryption scheme,” in International Confer