Defective Address Encoding for NAND Flash Memory Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for recording defective addresses in NAND-based flash memory arrays require significant storage space, leading to increased costs due to the large data size of the defective address list, which occupies valuable space in solid state drives.
Innovation Solution
The proposed solution employs Huffman encoding to reduce the data size of defective address information by dividing the memory array into segments and using segment codes that indicate the positions of fail bits, with each segment code comprising a start code and defective codes, thereby efficiently encoding the positions of fail bits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the positions of all fail bits are recorded using conventional addressing methods, then the complete defective address list can be maintained, but the storage space required increases significantly
Solution Approach 1:
The memory array is divided into multiple segments, and the defective address list is organized by segments. Each segment has its own segment code and associated fail bit positions. This segmentation allows the system to process and store defective address information in smaller, more manageable units, reducing the overall storage space required while maintaining complete tracking of all fail bits across the entire memory array.
Solution Approach 2:
The patent changes the parameter representation by introducing segment codes and using variable-length encoding for fail bit positions. Instead of using fixed-length absolute addresses for all fail bits, the system uses relative positioning within segments combined with Huffman encoding, which adjusts the number of bits required based on the frequency of position occurrences. This parameter transformation significantly reduces the storage space while preserving all necessary defective address information.
2Quantity of substance
If Huffman encoding is applied to defective address information, then the data size is reduced, but the encoding and decoding complexity increases
Solution Approach 1:
The Huffman encoding tables are pre-computed and stored during the defect mapping process. The segment codes and their corresponding Huffman encoded position mappings are established in advance based on the statistical distribution of fail bit positions. This preliminary action eliminates the need for complex real-time encoding calculations during memory operations, reducing the operational complexity while maintaining the data size reduction benefits.
Solution Approach 2:
The encoding complexity is reduced through segmentation by applying Huffman encoding separately to each segment's fail bit positions rather than encoding all fail bits across the entire memory array as a single large dataset. This division into smaller encoding units simplifies the encoding process, makes it more manageable, and allows for more efficient implementation while achieving the same overall compression ratio.
Data Source
AI summary
A defective address information encoding method for a memory array is provided. A page of the memory array is divided into plural segments. Each segment contains 2m bits. The defective address information encoding method includes the following steps. Firstly, positions of N1 fail bits in a first segment of the plural segments are acquired. Then, an (N1+1)-bit first segment start code is generated. Then, N1 m-bit defective codes are generated. The N1 m-bit defective codes follow the first segment start code to indicate the positions of the N1 fail bits in the first segment, wherein N1 is zero or a positive integer, and m is a positive integer.