A novel extreme learning machine‐based cryptography system

Abstract

Extreme Learning Machine (ELM) is a well‐known algorithm for the training of neural networks for two modes of functionality: regression and classification. This paper presents a novel model using ELM in ciphering. The study begins with an investigation of the Real‐time Recurrent Neural Network (RRNN) derived from the gradient‐based learning for symmetric cipher. The weakness of this cipher is that the error converges to zero, and that the RRNN will not change regardless of how the plaintext is changed. Given the nature of the ELM, a technique with an ELM‐based cipher is proposed to provide the capability of performing the training independently from the input and the error gradient. Different simulation scenarios were used to evaluate and validate the effectiveness of the proposed cipher. Results revealed that the ELM‐based cipher performed better than RRNNs, especially in terms of security. Moreover, the ELM‐based cipher demonstrated significantly competing performance for a wide range of evaluation measures. Using an ELM‐based cipher instead of an AES or other type of ciphers has the added advantage of providing an addition level of security—by allowing the user to change the algorithmic core of the cipher by simply changing the weights of the neural network. This allows hardware programmed ciphers to be more secure while costing less compared to other ciphers. Copyright © 2016 John Wiley & Sons, Ltd.

A Review on Machine Learning Approaches for Cryptography

“… It is observed that most of the literature uses neural networks, genetic algorithms, and deep learning in cryptography for generating secret keys and/or to perform encryption and decryption.  Neural network is more secure and robust to generate and exchange the key [5,7,8,10].  The cryptosystem produced by machine learning approaches achieves significant security.…”

“… Modern cryptographic techniques can be integrated with a neural network if a lossless neural network is achieved [9].  The key length of 64-bit provides high security against brute force attacks [8].…”

“…The suggested approach may be vulnerable to geometric and genetic attacks. Atee et al [8] developed a neural network-based sub-key generation scheme for a secure cryptosystem. It is based on an extreme learning machine for one hidden neural network layer with 100 neurons.…”

“…Table 1 summarises the literature reviewed in cryptography and machine learning regarding machine learning approaches, types of cryptosystems, evaluation metrics, and future scope. [5,6,7,8] and tools or programming language is not mentioned in [10,11,12].…”