Memory Error Protection Encoding With Precomputed Parity Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As memory devices become denser, they are more prone to errors due to factors like storage charge loss, random telegraph signal effects, and cosmic rays, especially in multi-level architecture where signal levels are close, leading to increased error probability and decreased read/write margins.
Innovation Solution
The implementation of a method to compute parity data using pre-computed coefficient data representing canonical coefficients, which allows for efficient error protection by generating parity data without iterative calculations, thereby reducing computational resources and latency, and enabling the use of combinational logic circuitry with small size and low power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If memory devices use multi-level architecture to increase storage capacity, then storage density is improved, but error probability increases due to decreased read/write margins
Solution Approach 1:
The patent applies preliminary action by pre-computing coefficient data representing canonical coefficients before actual encoding operations. This pre-computation stores essential encoding parameters in advance, allowing the encoder to quickly generate parity data using simple combinational logic without iterative calculations during actual data encoding, thus maintaining high storage capacity while improving encoding speed and reliability
Solution Approach 2:
The patent changes the parameter representation from iterative calculation parameters to pre-computed canonical coefficient parameters. By transforming the encoding approach to use pre-determined coefficient data, the system achieves faster encoding with reduced computational complexity, thereby improving reliability without sacrificing storage density
2Reliability
If iterative calculations are used to generate parity data for error protection, then error protection capability is improved, but computational resources and latency increase
Solution Approach 1:
The patent eliminates iterative calculations by pre-computing and storing canonical coefficient data. During actual encoding, the system directly uses these pre-computed coefficients to generate parity data through simple combinational logic operations, significantly reducing computational latency while maintaining full error protection capability
Solution Approach 2:
The patent creates a copy of the essential encoding information in the form of pre-computed coefficient data. Instead of performing iterative calculations each time parity is needed, the system uses this pre-computed copy of the encoding parameters, achieving the same error protection result with much lower computational overhead and latency
3Reliability
If complex encoding algorithms are used to reduce error rates, then reliability is improved, but logic complexity and power consumption increase
Solution Approach 1:
The patent reduces logic complexity by performing the complex computational work in advance when generating canonical coefficient data. The actual encoding process then uses only simple combinational logic operations with the pre-computed coefficients, achieving strong error protection with minimal real-time logic complexity and power consumption
Solution Approach 2:
The patent replaces iterative mechanical calculation processes with a static lookup-based system using pre-computed coefficient data. This substitution eliminates the need for complex iterative logic circuits during encoding, reducing device complexity and power consumption while maintaining reliable error protection through the use of predetermined canonical coefficients
Data Source
AI summary
Subject matter disclosed herein relates to error protection of data stored in and/or read from a memory device. Coefficient data representing canonical coefficients can be pre-computed by an apparatus before the apparatus is provided with program data, for example. For example, coefficient data may be pre-computed external to the apparatus and stored before program data is provided to an apparatus.