HPAC-sbox- a novel implementation of predictive learning classifier and adaptive chaotic s-box for counterfeiting sidechannel attacks in an IOT networks

Today, embedded systems are augmented with the Internet of things and more with the artificial intelligence to make world even connected with aliens. With an IoT networks are getting its insight since it deals with large number of data information, security has considered to be more important and needs to be a diagnosis for every minute. To enhance the security in the network, a mathematically secure algorithms were formulated and runs on the cryptographic embedded chips to counterfeit the risks which are caused by the different attacks such as side channel attacks (SCA) on the networks. Even though many cryptographic encryption algorithms such as AES, DES, RC4 algorithms were gaining its importance, fixed encryption keys, non-intelligent detection of attacks, cognitive countermeasures are some of the real-time challenges in an existing system of encryption. Following the limitations of existing systems, this research article focuses on design of new AES with HPAC-SBOX (Hybrid Prediction and Adaptive Chaos) which integrates powerful predictive learning algorithms and adaptive chaotic logistic S-Box. The following contributions of this research articles are: a) Preparation of Data Sets from the Power consumption traces captured from Multi Core Embedded boards while running the Advanced Encryption Systems(AES) on it b) Implementation of High Speed and High Accurate Prediction learning machines for the prediction of side-channel attacks c) Design of Adaptive Chaotic S-Box using 3-Dlogistic Hyperbolic maps for attacked bits. To evaluate the proposed architecture, experimentation in carried out in an IoT networks and various performance parameters were calculated and analyzed. The results show that the proposed architecture outperforms the other existing algorithms in terms of prediction and performance.