Dynamic Data Decryption With ML-Driven Key Rotation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data exchanged between endpoints in a communication network is vulnerable to interception and manipulation by bad actors, leading to compromised communication and security operations, potentially causing large-scale losses and information exposure.
Innovation Solution
A system and method that dynamically encrypts and decrypts data using machine learning models and multiple keys, ensuring a constantly-changing encryption process, and verifies data authenticity by comparing hash combinations, reducing processor and memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional static encryption methods are used, then implementation is simple, but security is compromised because encryption keys remain constant and can be intercepted
Solution Approach 1:
The patent applies dynamics by transforming static encryption keys into dynamic, constantly changing keys through machine learning model transformations. The encryption key evolves over time based on ML model outputs, ensuring that even if one key is compromised, future keys remain secure. This resolves the contradiction by making the encryption process adaptive and resilient while maintaining manageable complexity through automated ML-based key generation.
Solution Approach 2:
The patent changes the parameter of encryption keys from fixed values to dynamically transformed values generated by machine learning models. Multiple keys are transformed through different ML models to produce varying encryption parameters, enhancing security without requiring complex manual management. This parameter transformation approach allows the system to achieve high reliability through mathematical transformations while keeping the implementation complexity manageable.
2Reliability
If machine learning models are used for dynamic encryption, then data security is enhanced through constantly-changing keys, but processor and memory usage increases
Solution Approach 1:
The patent implements partial action by applying machine learning transformations selectively to generate encryption keys rather than continuously processing all data through ML models. The ML models are used to generate transformed keys at specific intervals or triggers, reducing the computational burden while still achieving enhanced security through dynamic key changes. This approach balances security enhancement with acceptable processor and memory usage.
3Reliability
If multiple keys are transformed through machine learning models, then encryption security is strengthened, but the complexity of key management increases
Solution Approach 1:
The patent implements self-service by enabling the machine learning models to automatically generate and manage transformed encryption keys without requiring manual intervention. The system autonomously performs key transformations, generation, and rotation based on predefined triggers or time intervals, reducing the operational complexity of managing multiple keys while maintaining strong encryption security through automated processes.
Data Source
AI summary
An apparatus comprises a memory communicatively coupled to a processor. The processor is configured to decrypt encrypted shareable data based at least in part upon one or more keys upon executing one or more machine learning algorithms in accordance with one or more machine learning models. Further, the processor is configured to obtain shareable data, the shareable data being a decrypted version of the encrypted shareable data, generate a first hash of the encrypted shareable data, generate a second hash of the shareable data, combine the first hash and the second hash into a combined receiver hash, and perform one or more verification operations based at least in part upon the combined receiver hash and a decrypted combined sender hash. The decrypted combined sender hash is a decrypted version of the encrypted combined sender hash.