Data Protection Mechanism with Soft Information for Computing Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems lack effective data protection mechanisms to ensure improved data reliability and recovery, especially in modern consumer and industrial electronics, where data pervasiveness and commercial pressures necessitate enhanced data protection solutions.
Innovation Solution
A computing system that processes data through both failed and remaining channels, calculates an aggregated output from hard decisions, determines a selected magnitude with error detection, generates extrinsic soft information, and decodes the failed channel using a scaled soft metric based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is processed through multiple channels including failed channels, then data reliability is improved, but system complexity increases due to the need for aggregated output calculation and extrinsic soft information generation
Solution Approach 1:
The system segments the data processing into distinct functional components: channel processing units that handle individual channels separately, an aggregation module that combines hard decisions from remaining channels, and an extrinsic information generation unit. This segmentation allows complex multi-channel processing to be managed through modular, independent units that can be implemented and maintained separately.
Solution Approach 2:
The patent introduces extrinsic soft information as an intermediary element that mediates between the hard decisions from remaining channels and the decoding of failed channels. This intermediary representation (aggregated output with selected magnitude) simplifies the interaction between channels by providing a standardized form of information that can be used to reconstruct failed channel data without requiring direct complex interactions between all channel components.
2Reliability
If soft information and hard decisions are combined for decoding, then error detection and correction capability is improved, but computational requirements increase
Solution Approach 1:
The system applies partial action by selectively processing only the necessary components from each channel. Instead of fully processing all channel information, the system extracts hard decisions from remaining channels and generates only the essential extrinsic soft information (aggregated output and selected magnitude) needed for failed channel decoding. This partial processing approach reduces computational overhead while maintaining sufficient error detection and correction capability.
Solution Approach 2:
The patent transforms channel information into different parameter representations: converting channel outputs into hard decisions (binary values), aggregating these into an aggregated output with selected magnitude, and using these transformed parameters for decoding. This parameter transformation allows the system to work with simplified representations that reduce computational complexity while preserving the essential error detection and correction information.
3Reliability
If iterative decoding is implemented for failed channels, then data recovery is improved, but processing time increases
Solution Approach 1:
The system performs preliminary action by pre-calculating and storing extrinsic soft information (aggregated output and selected magnitude) from remaining channels before the actual decoding of failed channels is required. This preliminary preparation of recovery information allows the iterative decoding process to proceed more efficiently, as the necessary extrinsic information is already available rather than needing to be computed during the iterative process itself.
Data Source
AI summary
A computing system includes: a data block including a data; a storage engine, coupled to the data block, configured to process data, as hard information or soft information, through channels including a failed channel and a remaining channel, calculate an aggregated output from a hard decision from the remaining channel, calculate a selected magnitude from a magnitude from the remaining channel with an error detected, calculate an extrinsic soft information based on the aggregated output and the selected magnitude, and decode the failed channel with a scaled soft metric based on the extrinsic soft information.