IMPROVED PARALLEL HILL CIPHER ALGORITHM USING MAPREDUCE

Abstract

In this thesis, a hill cipher algorithm using parallel block matrix multiplication on MapReduce is put forward to reduce the time for encryption process and provide additional security against internal and external attacks. The data being stored in the cloud is usually very large, the demand to decrease the time for encryption, as well as to increase the security of the overall cloud, becomes necessary. To overcome these drawbacks, a parallel algorithm on the modified hill cipher is implemented on a MapReduce framework, with parallelism implemented for easier simplification of the crux of hill cipher. The key matrix generated for the algorithm is intended to be self-invertible, to decrease the decryption time of the encrypted text. The experimental analysis on the input data demonstrates the effectiveness of the proposed technique.