A HYBRID ELGAMAL – PAILLIER APPROACH FOR ACHIEVING PRACTICAL FULLY HOMOMORPHIC ENCRYPTION IN CLOUD COMPUTING

Abstract

Data in the processing stage becomes vulnerable and unsecure as one cannot fully trust the individual responsible for the operation. This research paper aims to offer privacy for data in processing using a method known as Homomorphic Encryption. We have examined all significant and influential cryptographic algorithms, including RSA, Paillier, Elgamal, and the Goldwasser-Micali algorithm, which can be applied to data that is currently in use. A code is written in python and is executed to conduct a brief analysis of functioning and efficiency of each algorithm. A comparison chart has been created based on the time taken by each algorithm in seconds, along with the respective encryption time, homomorphic encryption time, and decryption time. Additionally, a line graph has been generated to evaluate the performance of each algorithm. The major part of our research is an attempt to combine two algorithms i.e. Paillier cryptography algorithm and Elgamal cryptography algorithm and have created a hybrid approach which will not only provide the fully homomorphic capability but also enhance the security features for the computation of in use data. This research could have a significant impact as this technology is increasingly required in the IT industry, given the rise in data, necessitating enhanced security features.