Recurrent Neural Network Encryption via Topological Permutation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing cryptographic encryption methods are vulnerable to cracking, particularly with the advancement of computing power, which compromises the security of communications and data storage.

Innovation Solution

The use of a recurrent artificial neural network to encrypt and decrypt information by identifying topological structures in patterns of activity and implementing a permutation of binary codewords, ensuring the encryption method is injective and surjective.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional cryptographic encryption methods are used, then encryption and decryption can be performed, but the security is vulnerable to cracking due to advancement of computing power

Engineering Contradiction:

Improveencryption securityVSAvoidvulnerability to cracking

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent replaces traditional mechanical cryptographic systems with a biological neural network-based encryption system. The neural network processes plaintext through its complex nonlinear transformations to generate ciphertext, leveraging biological computation rather than conventional mathematical algorithms to achieve enhanced security against cracking.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the fundamental parameters of encryption by using the dynamic states of neural network nodes and synapses as the encryption mechanism. The encryption process involves transforming plaintext into patterns of neural activity, where the complexity and variability of neural states provide the security foundation, replacing static mathematical keys with dynamic biological states.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If a recurrent artificial neural network is used for encryption, then security against cracking is improved, but the device complexity increases

Engineering Contradiction:

Improveencryption securityVSAvoidneural network complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The neural network serves multiple functions: it acts as the encryption mechanism, the decryption mechanism, and the key management system. The same network structure that processes plaintext for encryption can be configured to process ciphertext for decryption, eliminating the need for separate encryption and decryption devices and reducing overall system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The neural network performs self-organization and self-adjustment during the encryption and decryption processes. The network's inherent learning capabilities and plasticity allow it to adapt to different encryption tasks without external intervention, reducing the need for complex external control systems and simplifying the overall device architecture.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250070956A1Encrypting and decrypting information

Publication Date: 2025.02.27 INAIT SA

AI summary

Methods, systems, and devices for encrypting and decrypting data. In one implementation, an encryption method includes inputting plaintext into a recurrent artificial neural network, identifying topological structures in patterns of activity in the recurrent artificial neural network, wherein the patterns of activity are responsive to the input of the plaintext, representing the identified topological structures in a binary sequence of length L and implementing a permutation of the set of all binary codewords of length L. The implemented permutation is a function from the set of binary codewords of length L to itself that is injective and surjective.