“…The second approach is to restrict the set of possible generated shares from each original value in the domain of the sensitive attribute to be protected on each data server. Schemes based on this approach [13,15,16,18,19,21,22] are referred to as restriction-based SSDO schemes.…”
Section: Shamir's Secret Sharing Schemementioning
confidence: 99%
“…Inferences could be made from indices (in metadata-based SSDO schemes and encryption-based data outsourcing schemes with indexing), tuples of shares (usually in restriction-based SSDO schemes) or encrypted values (in encryption-based data outsourcing schemes and some SSDO schemes). In [13,16], the following five types of inferences from outsourced databases (using secret sharing or encryption) were introduced:…”
Section: Security Limitations Of Data Outsourcing Schemesmentioning
confidence: 99%
“…The scheme is resistant to all known attacks and several types of inferences but is still vulnerable to ordering and closeness inferences. A more detailed discussion of the SSDO schemes and their limitations can be found in [13,16].…”
Section: A Review Of Ssdo Schemesmentioning
confidence: 99%
“…Another way is to expose some information that can help adversaries to perform attacks on the shares of the cover attribute. Readers are referred to [13,16] to find what types of information can be exploited to perform attacks. 4.…”
Section: Recovery (By Adversaries After Successful Attack On )mentioning
confidence: 99%
“…As considered in [13], secret sharing is an efficient and practical approach to protecting the confidentiality of outsourced databases in contrast to data encryption. Several secret sharingbased data outsourcing schemes exist in the literature [13][14][15][16][17][18][19][20][21][22][23][24].…”
Data encryption‐based and secret sharing‐based data outsourcing schemes protect the confidentiality of sensitive attributes but not their secrecy. Ciphertexts/shares generated by a data encryption/secret sharing scheme can attract the attention of interceptors. Thus, it is desired to hide the existence of highly‐sensitive attributes (as secret attributes) in the outsourced relations in addition to protecting their contents. This paper proposes a novel scheme that integrates data hiding with secret sharing for relational databases to protect both the secrecy and confidentiality of secret attributes. It embeds one or multiple secret attributes in a relation into one or multiple cover attributes in the same relation. A set of share (and possibly index) columns are constructed such that they are pretended to be associated with only the cover attributes, while those share columns and some virtual share columns can be used to recover both the secret and cover attributes. What interceptors observe in each relation include the attributes stored in plaintext and the share (and possibly index) columns associated with the cover attributes but not any extra column. Thus, they find nothing suspicious. This is the first effective data hiding scheme for relational databases that protects the secrecy of secret attributes.
“…The second approach is to restrict the set of possible generated shares from each original value in the domain of the sensitive attribute to be protected on each data server. Schemes based on this approach [13,15,16,18,19,21,22] are referred to as restriction-based SSDO schemes.…”
Section: Shamir's Secret Sharing Schemementioning
confidence: 99%
“…Inferences could be made from indices (in metadata-based SSDO schemes and encryption-based data outsourcing schemes with indexing), tuples of shares (usually in restriction-based SSDO schemes) or encrypted values (in encryption-based data outsourcing schemes and some SSDO schemes). In [13,16], the following five types of inferences from outsourced databases (using secret sharing or encryption) were introduced:…”
Section: Security Limitations Of Data Outsourcing Schemesmentioning
confidence: 99%
“…The scheme is resistant to all known attacks and several types of inferences but is still vulnerable to ordering and closeness inferences. A more detailed discussion of the SSDO schemes and their limitations can be found in [13,16].…”
Section: A Review Of Ssdo Schemesmentioning
confidence: 99%
“…Another way is to expose some information that can help adversaries to perform attacks on the shares of the cover attribute. Readers are referred to [13,16] to find what types of information can be exploited to perform attacks. 4.…”
Section: Recovery (By Adversaries After Successful Attack On )mentioning
confidence: 99%
“…As considered in [13], secret sharing is an efficient and practical approach to protecting the confidentiality of outsourced databases in contrast to data encryption. Several secret sharingbased data outsourcing schemes exist in the literature [13][14][15][16][17][18][19][20][21][22][23][24].…”
Data encryption‐based and secret sharing‐based data outsourcing schemes protect the confidentiality of sensitive attributes but not their secrecy. Ciphertexts/shares generated by a data encryption/secret sharing scheme can attract the attention of interceptors. Thus, it is desired to hide the existence of highly‐sensitive attributes (as secret attributes) in the outsourced relations in addition to protecting their contents. This paper proposes a novel scheme that integrates data hiding with secret sharing for relational databases to protect both the secrecy and confidentiality of secret attributes. It embeds one or multiple secret attributes in a relation into one or multiple cover attributes in the same relation. A set of share (and possibly index) columns are constructed such that they are pretended to be associated with only the cover attributes, while those share columns and some virtual share columns can be used to recover both the secret and cover attributes. What interceptors observe in each relation include the attributes stored in plaintext and the share (and possibly index) columns associated with the cover attributes but not any extra column. Thus, they find nothing suspicious. This is the first effective data hiding scheme for relational databases that protects the secrecy of secret attributes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.