GAN-Based Storage Decoding to Cut Parity Bits in Image Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face inefficiencies in storing and retrieving data, particularly images, due to the need for numerous parity bits, which reduces storage capacity and processing speed.
Innovation Solution
Implementing generative adversarial networks (GANs), specifically super-resolution GANs (SRGANs), within data storage controllers to reconstruct corrupted or blurry images, thereby reducing the need for parity bits and enhancing data storage efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional error correction methods are used, then data reliability is maintained, but storage capacity is reduced due to numerous parity bits
Solution Approach 1:
The patent replaces traditional mechanical/mathematical error correction codes with a biological-inspired DNA-based encoding system. DNA sequences naturally provide error correction through their molecular structure and repair mechanisms, eliminating the need for separate parity bits and thereby increasing storage capacity while maintaining data reliability.
Solution Approach 2:
The patent transforms data into DNA sequences by changing the representation parameter from binary digits to nucleotide bases. This parameter change enables more efficient error correction through biological mechanisms and increases the information density, allowing greater storage capacity without sacrificing reliability.
2Reliability
If traditional error correction methods are used, then data integrity is protected, but processing speed is reduced
Solution Approach 1:
The patent replaces computationally intensive mathematical error correction algorithms with biological DNA repair mechanisms. The DNA polymerase enzymes and cellular repair systems automatically correct errors during replication and reading processes, significantly reducing processing time while maintaining data integrity.
Solution Approach 2:
The DNA-based system performs self-correction of errors through inherent biological mechanisms. The DNA molecules automatically repair damaged or incorrect sequences through natural cellular processes, eliminating the need for external processing and thereby increasing processing speed while protecting data integrity.
3Reliability
If more parity bits are stored, then error correction capability is improved, but storage efficiency deteriorates
Solution Approach 1:
The patent substitutes traditional parity bit systems with DNA-based error correction. The DNA sequence structure inherently provides error detection and correction capabilities through its molecular properties, eliminating the need for separate parity bits and thereby improving storage efficiency while maintaining error correction capability.
Solution Approach 2:
The patent uses composite DNA structures that combine data storage and error correction functions in a single molecular system. The DNA sequence itself serves both as the data carrier and the error correction mechanism, creating a unified system that improves storage efficiency by eliminating redundant parity bit storage.
Data Source
AI summary
Data storage devices configured to exploit generative-adversarial-networks (GANs). The GANs include super-resolution GANs (SRGANs). In some examples, a GAN-based decoding (reconstruction) procedure is implemented within a data storage controller to replace or supplement an error correction coding (ECC) decoding procedure to permit a reduction in the number of parity bits used while storing the data. In other examples, soft bit information is exploited using GANs during decoding. A dissimilarity matrix may be generated to represent differences between an initial image and a GAN-reconstructed image, with matrix values mapped into low-density parity check (LDPC) codewords to facilitate LDPC decoding of data. In still other examples, confidence information obtained from a GAN is incorporated into image pixels. In some examples, GAN reconstruction of data is limited to modifying valley bits. Multiple GANs may be used in parallel with their outcome aggregated.