RIO Polar Code Fusion for Flash Memory Read Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RIO codes in flash memory devices do not account for noise, leading to amplified bit errors during read operations, degrading performance due to increased error bits beyond the capability of error correction codes.
Innovation Solution
Implementing an RIO code with a channel code for error correction, allowing for fewer than ⌊2k-1⌋ sensing operations in multilevel cells, and using a polar code for encoding and decoding to correct channel errors during read operations, thereby reducing error propagation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional RIO code is used without considering noise, then the coding structure is simple, but bit errors are amplified and error correction capability is insufficient
Solution Approach 1:
The patent merges the RIO code with a channel code (polar code) to form a combined coding structure. The RIO encoder encodes data using RIO code, then the channel code encoder further encodes the result using polar code. This merging allows the system to simultaneously achieve RIO code functionality and channel error correction, resolving the contradiction between reliability improvement and complexity increase.
Solution Approach 2:
The patent employs a composite coding structure combining RIO code and polar code. The RIO code handles the random input/output characteristics while the polar code provides channel error correction. This composite approach leverages the strengths of both codes to achieve both error correction and noise resilience without requiring excessive complexity.
2Measurement precision
If multiple sensing operations are performed to read data from multilevel cells, then data can be accurately read, but read performance is degraded due to increased sensing time
Solution Approach 1:
The patent performs preliminary encoding using RIO code and channel code before data is stored in multilevel cells. This preliminary action allows the data to be structured in a way that enables accurate retrieval through fewer sensing operations. The encoding structure prepares the data so that when read, it can be correctly identified and decoded with minimal sensing attempts, thus improving read performance without sacrificing accuracy.
Solution Approach 2:
The patent incorporates feedback mechanisms in the decoding process. The channel code decoder uses feedback from sensing operations to iteratively correct errors and refine the read data. This feedback loop allows the system to achieve high reading accuracy even with fewer sensing operations by continuously improving the data quality based on received feedback signals.
Data Source
AI summary
A method of decoding may include performing fewer than⌊2k-1k⌋number of sensing operations of multilevel cells within a nonvolatile memory device, decoding pages corresponding to each of the sensing operations while correcting a channel error using an RIO code, and extracting user data from the decoded pages.