Leakage Resilient Leveled FHE on Multiple Bit Message

Abstract

Fully Homomorphic Encryption (FHE) allows computing over encrypted data without decrypting the corresponding ciphertexts, and it constitutes a promising cryptographic primitive to preserve data privacy in the big data computing environments. In general, FHE schemes can be constructed by using the standard Learning with Errors (LWE) assumption, and the current crux lies in how to achieve efficient multi-bit FHE encryption while being leakage-resistant against attackers who may capture the information of cryptographic secret keys via side channel attacks. It aims to address this issue by giving a new structure of public key matrix with any number of LWE instances, thereby avoiding the use of a straightforward composition to achieve multi-bit FHE under standard LWE. FHE is a public key encryption scheme supporting algebraic operations on encrypted data. A leakage-resilient leveled FHE scheme (hereafter LRMGSW) via the methodology under the “bounded adaptive leakage resilient” model was constructed. The proposed multi-bit FHE is able to prevent the attacker from capturing the valid information of cryptography secret key via side channel attacks. The MGSW scheme achieves leakage-resilience (i.e., LRMGSW scheme) and tolerates more leaked bits as compared with the leakage resilient HAO scheme. Particularly, the proposed scheme attains provable security and the efficiency of FHE is improved.

Existing System

In this project it observes the Peikert-Vaikuntanathan-Waters (PVW) method of packing many plaintext elements in a single Regev-type ciphertext, can be used for performing SIMD homomorphic operations on packed ciphertext. This provides an alternative to the Smart-Vercauteren (SV) ciphertextpacking technique that relies on polynomial-CRT. While the SV technique is only applicable to schemes that rely on ring-LWE (or other hardness assumptions in ideal lattices). A good approximation of the runtime of such algorithms is therefore essential in the analysis of LWE’s security. In this cryptographic context however, it deals with lattice dimensions and computational costs that go far beyond the scope of experiments. The PVW method can be used also for cryptosystems whose security is based on standard LWE. By using the PVW method with LWE-based schemes leads to worse asymptotic efficiency than using the SV technique with ring-LWE schemes, the simplicity of this method may still offer some practical advantages. Also, the two techniques can be used in tandem with “general-LWE” schemes, suggesting yet another tradeoff that can be optimized for different settings.

Proposed System

The previous model of data leakage is overcome by the proposed scheme. The proposed model outsource the data in cloud. Big data technologies collect vast amounts of data from heterogeneous sources to gain greater insight into patterns and trend not generally discernible when analyzing smaller data sets. Cloud computing and big data are conjoined, and their relationship can be well expressed by “Big data provides users the ability to use commodity computing to process distributed queries across multiple datasets and return resultant sets in a timely manner. The secure key is transmitted to all the user. The data owner uses the secure FHE method for verification. The user who used to request the data need to provide a secure secret key generated at that moment. If the key was accurately provided by the user they can access the data. A common feature of these traditional FHE schemes is that they only allow computation on encrypted single bit and the efficiency is unsatisfactory. Most of FHE constructions focus on single bit encryption, though it is pointed out they can be amortized in the concatenation way (or a straightforward composition) to achieve multibit computation. Unfortunately, this straightforward construction for multi-bit FHE will not lead to the best performance. It there is any misidentification of key means the data will not response to the respected user. Thus this produce a two level security which the user want to give accurate key provided by the data owner. And at the same time the secure validity key generated by at that moment.

Architecture

ARCHITECTURE DIAGRAM