Multi-Layer Data Decoding With Identity-Matrix Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data decoding systems require large storage circuits and consume significant power, especially when handling large codewords with non-binary decoding, which exacerbates the demand for efficient data processing solutions.
Innovation Solution
The implementation of a data processing system that includes a data encoder circuit applying a multi-layer code structure with a last row conforming to an identity matrix, utilizing a compression-based decoding circuit that compresses and decompresses data to reduce memory requirements, and employs a data decoder circuit that adds and subtracts decoded outputs to align with the identity matrix for efficient decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional data decoding circuits store entire codewords including parity bits, then decoding accuracy is maintained, but storage circuit area and power consumption increase significantly
Solution Approach 1:
The patent extracts and removes the identity matrix portion (last row) from the code structure matrix during decoding operations. By taking out this redundant information that is already known, the system reduces storage requirements while maintaining decoding accuracy, as the extracted identity portion does not need to be stored in memory circuits
Solution Approach 2:
The patent segments the code structure matrix into two parts: the stored codeword data and the identity matrix portion. This segmentation allows the system to process and store only the necessary codeword data while handling the identity matrix portion through computational operations, thereby reducing storage circuit area requirements
2Measurement precision
If traditional data decoding circuits store entire codewords including parity bits, then decoding accuracy is maintained, but power consumption increases
Solution Approach 1:
By extracting and removing the identity matrix portion from storage requirements, the system reduces the amount of data that needs to be maintained in memory circuits. This extraction eliminates unnecessary power consumption associated with storing and retrieving redundant identity matrix information while preserving decoding accuracy
Solution Approach 2:
The segmentation of code structure into stored codeword data and computationally processed identity matrix reduces the active storage burden, thereby lowering the power consumption of storage circuits while maintaining the ability to perform accurate decoding operations
3Measurement precision
If non-binary decoding is performed with multiple probability values per symbol, then decoding precision is improved, but storage and power requirements are exacerbated
Solution Approach 1:
The patent applies the extraction principle specifically to non-binary decoding by removing the identity matrix portion from the code structure matrix before processing. This reduces the storage burden for multi-valued symbols and their associated probability values, thereby lowering power consumption while maintaining the enhanced decoding precision that non-binary methods provide
Solution Approach 2:
The patent modifies the parameter representation by changing how the code structure matrix is handled - specifically by extracting and removing the identity matrix portion. This parameter change reduces the storage requirements for non-binary symbols and their probability values, thereby reducing power consumption while preserving decoding precision
Data Source
AI summary
Various embodiments of the present invention provide systems and methods for data processing. For example, a data processing system is disclosed that includes a data encoder circuit. The data encoder circuit is operable to apply an encoding algorithm to an input data set in accordance with a multi-layer code structure including a first row and a last row to yield an encoded data set. The last row of the multi-layer code structure represented in the encoded data set conforms to an identity matrix.