Neural Cryptography Key Generation via Blockchain Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing neural cryptography methods face security risks due to the need to transmit data protection parameters over networks and rely on static key generation or storage, which can be vulnerable to attacks and lack adaptability.
Innovation Solution
A method using artificial neural networks (ANNs) trained on a blockchain, where encryption keys, algorithms, and obfuscation methods are generated and updated at each network node without transmission, utilizing synchronized ANNs and blockchain data for continuous self-training and key generation, ensuring dynamic and secure encryption and decryption processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data protection parameters (encryption keys, algorithms) are transmitted over the network for encryption and decryption, then data security is compromised due to potential interception and attacks, but eliminating transmission requires a new approach to key distribution and synchronization
Solution Approach 1:
The patent extracts the data protection parameters (encryption keys, algorithms, obfuscation methods) from the transmission process entirely. Each node generates these parameters locally using its synchronized ANN, eliminating the need to transmit sensitive cryptographic materials over the network, thus resolving the security risk while maintaining system functionality
Solution Approach 2:
Each node autonomously generates its own encryption keys, selects encryption algorithms, and determines obfuscation methods through local ANN processing. The system serves itself by having nodes independently produce all necessary data protection elements without relying on external key distribution mechanisms, thereby eliminating transmission vulnerabilities
2Adaptability or versatility
If static key generation or storage methods are used in neural cryptography, then the system is simpler to implement, but security is vulnerable to attacks and lacks adaptability to new threats
Solution Approach 1:
The patent transforms static key generation into a dynamic process where encryption keys, algorithms, and obfuscation methods are generated on-demand by synchronized ANNs at each node. This dynamic generation adapts to each communication instance, providing resilience against attacks while managing complexity through the self-organizing capability of neural networks
Solution Approach 2:
The system changes the parameters of encryption dynamically by having each node independently select different encryption keys, algorithms, and obfuscation methods based on its local ANN state. This parameter variability ensures adaptability to new threats while the underlying ANN synchronization mechanism manages the complexity of coordinating these changes across the network
3Reliability
If neural networks are retrained continuously with blockchain updates for key generation, then security and adaptability are enhanced, but computational resources and processing time are consumed
Solution Approach 1:
The patent performs preliminary action by training the neural networks on blockchain data in advance, before actual data transmission occurs. This pre-training establishes the synchronized ANNs' capability to generate secure encryption parameters, reducing the computational burden during active communication while maintaining high security standards
Solution Approach 2:
The system implements periodic retraining of neural networks synchronized with blockchain updates. This periodic action balances security enhancement with resource management by retraining only when necessary (at blockchain update intervals) rather than continuously, reducing energy consumption while maintaining adaptability
Data Source
Figure 1~2
Figure 3
AI summary
A method of encryption and decryption of data over a network using an artificial neural network installed on each node of the network. The data protection elements-encryption keys, encryption algorithms, and encryption obfuscation-are generated or selected, respectively, at a new instance of communication across the network and no data protection elements are transmitted across the network. The artificial neural network is trained on a blockchain with the addition of each new block to the blockchain and is used to generate a finite set of encryption keys at each node simultaneously. Such encryption keys, encryption algorithms and encryption obfuscation are associated with the neural network on each node and are then used for decryption of the transmitted data.