Lattice-Based Functional Encryption for Next-generation Privacy-Preserving Computation

Abstract: With the rise of cloud-based data processing and multi�party computations, the demand for secure, sturdy, and effective
privacy-preserving tasks has risen. Homomorphic Encryption
(HE) and Bilinear Functional Encryption (FE) users face
computational and properties constraints in the form of high
computation overheads, high ciphertext expansions, and lower
functionality accuracy reducing usability. This research wishes to
avoid such limitations by developing Lattice-Based Functional
Encryption (FE) framework, which takes into consideration the
neighborhood of security, efficiency and accuracy. Constructing a
series of structured lattice based schemes for key generation and
ciphertext formatting, allows for encrypted data evaluation
without any need for decrypting, with worst-case hardness
security properties. Experimental evaluations based on
structured datasets, time of encryption, time of queries
evaluation, ciphertext length, and accuracy of functions were
observed. Lattice-Based FE achieved a total time of 7.5 ms/1 KB
encryption time, 14.3 ms/ query evaluation time, 2.1 KB/ 1 KB
ciphertext size and resulted in 99.0% functional accuracy as
compared to HE and Bilinear FE methods. The findings show
that the lattice-based FE was able to deliver scalable and effective
privacy-preserving computations. This framework moreover will
provide a promising approach for secure data processing tasks
based on the IoT, finance and cloud interest, ushering an
expansion into more cryptographic methods for years to come.
Keywords: Lattice-Based Functional Encryption, Homomorphic
Encryption, Privacy-Preserving Computation, Ciphertext
Efficiency, Encrypted Data Evaluation, Secure Computation