IoT lightweight cipher security investigation by machine learning techniques
No Thumbnail Available
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Oum El Bouaghi
Abstract
The standard Use of Internet of Things (IoT) technology is increasing dramatically and is present in every field of our modern lives. In most IoT systems, the information encryption and decryption of sensitive data are completely implemented by the used terminals which merely limit its specific functionality in the secure transmission of sensitive information. Where the development of lightweight ciphers adequately addresses
the limitations of the modest size, small storage memory, low consumption of energy, and weaker computing power. Cryptanalysis work published on IoT encryption may be impractical or convincingly demonstrate apparent limitations to generalized. Because they frequently require a considerable amount of critical time, known plain texts, and large storage memory, they are generally performed without the restriction of
keyspace, or only the small round variants are attacked. This work proposes deep learning (DL) model-based approach for a successful attack that discovers the plain text from the ciphertext one, the proposal DL-based cryptanalysis is shown to represent a promising step towards and an automated test to verify the security of emerging IoT ciphers. the results are given and communicated to precisely demonstrate the effective performance of the attack.
Description
Keywords
Tensorflow ; Deep learning ; Neural networks ; Cryptanalysis , Lightweight cipher , Attack , Internet of Things