학술논문

Design of chaotic neural network based method for cryptographic substitution box
Document Type
Conference
Source
2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT) Electrical, Electronics, and Optimization Techniques (ICEEOT), International Conference on. :864-868 Mar, 2016
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Engineered Materials, Dielectrics and Plasmas
Engineering Profession
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Cryptography
Biological neural networks
Neurons
Chaos
Generators
Transfer functions
Boolean functions
Substitution box
neural network
chaotic maps
block cryptosystem
security
Language
Abstract
The cryptographic substitution boxes are the substantive constituent of most modern day block cryptosystems. Here, we proposed a novel method to generate cryptographically potent S-boxes by exploring the blended strength of chaos and neural network in its design. The designed chaos-based neural network, engaged to yield S-boxes, consists of four layers each of which have eight, four, two and one neuron(s), respectively. The excogitation and cognitive operation of chaotic neural network is couched to sample the random elements which eventually render infrangible configuration of S-box. By utilizing the features of chaos and neural network efficiently, we explicate cryptographically strong S-boxes that have the desired potentiality and practicability. The statistical scrutiny of proposed method against widely accepted performance measures suggest that the method is amicable to contrive dynamical S-boxes for strong block cryptosystem with respectable cryptographic features.