Double Fold Encryption Algorithm Using Adjacency Matrix and Linear Feedback Shift Register (LFSR) as Diffusion

Abstract

Sensitive data transfer and sharing between two or more legitimate entities have become a difficult task due to indiscriminate and ever-increasing usage of internet in technology and communications. The sole method for defending data against cyber-attacks is to mask the important information in transit. For a very long time, graph theory plays a crucial role in encryption. The adjacency matrices of undirected graphs, in particular, are used in the current research to develop a block cypher. The data is encrypted in blocks, with each block having three rounds and each round having two different phases: the first phase uses an adjacency matrix, and the second phase applies a logical XOR operation in a particular pattern to each component of the message. The adjacency matrix in this case is known to everyone, and the entities use a linear feedback shit register (LFSR) to create new matrices for encrypting each block. The diffusion layer in this case is the LFSR. The LFSR polynomial is a secret key that allows the entities to interact with one another or with a reliable third party.