“…Thanks to a packed representation of the biometric templates, the protocol is able to compute the Hamming distance with only three products. Tests performed on a 3.07-GHz processor show that only 18.10 milliseconds are necessary for distance computation, which is not only faster than the SHEbased implementation of [21], but also faster than the Hamming distance computational time of SCiFI (310 milliseconds) [26] and [30] (150 milliseconds). In both [21] and [39], only the distance is computed by means of SHE operating on integers.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 97%
“…An efficient OT implementation of both Hamming and squared Euclidean Distance has been proposed in [20]. It is also possible to implement such distances through SHE [21], while, given the limited number of operations required in both cases, resorting to FHE is not necessary.…”
“…In [21], the first noninteractive biometric authentication protocol, based on an integer extension of the SHE scheme described in [38], is presented. All the computation is moved on the server's side, leaving only the encryption of the inputs and the decryption of the result to the client.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 99%
“…With respect to an equivalent implementation based on the Pailler cryptosystem, the computational complexity is considerably reduced (59 seconds for Troncoso et al implementation versus the 420 seconds of an equivalent Paillier implementation), with the additional advantage of avoiding the interaction between the parties. On the other hand, due the larger expansion factor of a lattice-based cryptosystem like [38], the communication complexity is larger than the Paillier-based version: 393 megabytes in [21] and 16.4 megabytes for the Paillier-based version. Another authentication protocol based on SHE has been proposed in [39].…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 99%
“…Such schemes permit only the computation of polynomial functions of the inputs, and they cannot be used for comparisons. For this reason, in [21] and [39] the final comparison is carried out in plain domain by the client. For completeness, we highlight that beyond papers strictly focusing on biometric recognition, other interesting privacy-preserving applications that can be also applied to biometric protocols have been developed.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
“…Thanks to a packed representation of the biometric templates, the protocol is able to compute the Hamming distance with only three products. Tests performed on a 3.07-GHz processor show that only 18.10 milliseconds are necessary for distance computation, which is not only faster than the SHEbased implementation of [21], but also faster than the Hamming distance computational time of SCiFI (310 milliseconds) [26] and [30] (150 milliseconds). In both [21] and [39], only the distance is computed by means of SHE operating on integers.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 97%
“…An efficient OT implementation of both Hamming and squared Euclidean Distance has been proposed in [20]. It is also possible to implement such distances through SHE [21], while, given the limited number of operations required in both cases, resorting to FHE is not necessary.…”
“…In [21], the first noninteractive biometric authentication protocol, based on an integer extension of the SHE scheme described in [38], is presented. All the computation is moved on the server's side, leaving only the encryption of the inputs and the decryption of the result to the client.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 99%
“…With respect to an equivalent implementation based on the Pailler cryptosystem, the computational complexity is considerably reduced (59 seconds for Troncoso et al implementation versus the 420 seconds of an equivalent Paillier implementation), with the additional advantage of avoiding the interaction between the parties. On the other hand, due the larger expansion factor of a lattice-based cryptosystem like [38], the communication complexity is larger than the Paillier-based version: 393 megabytes in [21] and 16.4 megabytes for the Paillier-based version. Another authentication protocol based on SHE has been proposed in [39].…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
confidence: 99%
“…Such schemes permit only the computation of polynomial functions of the inputs, and they cannot be used for comparisons. For this reason, in [21] and [39] the final comparison is carried out in plain domain by the client. For completeness, we highlight that beyond papers strictly focusing on biometric recognition, other interesting privacy-preserving applications that can be also applied to biometric protocols have been developed.…”
Section: Optimization Of Sped Protocols Through Cryptographic Primitimentioning
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.