LDPC Code Shortening Patterns for Variable-Length Modulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LDPC codes used in communication systems, such as DVB-S2, are limited to only two codeword lengths, which restricts their extendibility and flexibility, especially in high-order modulation schemes like 16-QAM, 64-QAM, and 256-QAM, where bit reliabilities differ, necessitating a method to generate LDPC codes with various codeword lengths without requiring new parity-check matrices.
Innovation Solution
A method and apparatus for generating LDPC codewords with different lengths using shortening or puncturing, optimized for high-order modulation schemes, by determining specific shortening and puncturing patterns based on the modulation scheme, allowing for efficient support of various codeword lengths using a single parity-check matrix.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If LDPC codes are limited to two fixed codeword lengths, then the system has simpler design and implementation, but the extendibility and flexibility are restricted
Solution Approach 1:
The patent applies parameter changes by modifying the codeword length parameter through shortening and puncturing operations on the original LDPC code. By changing the length parameter N to N' < N while maintaining the same parity-check matrix structure, the system achieves multiple codeword lengths without designing new codes for each length, thus improving extendibility without proportionally increasing design complexity
Solution Approach 2:
The patent makes a single LDPC code design serve multiple functions by enabling it to generate various codeword lengths through systematic shortening and puncturing. The same base parity-check matrix H is used universally across different codeword lengths, allowing one code design to fulfill multiple communication requirements with different data rates and lengths
2Reliability
If new parity-check matrices are designed for each codeword length, then optimal performance for each length is achieved, but storage requirements and system complexity increase
Solution Approach 1:
The patent creates virtual copies of the original LDPC code by applying shortening and puncturing operations. Instead of storing multiple distinct parity-check matrices, the system generates derived codes from the original matrix H through systematic modifications, reducing storage requirements while maintaining decoding performance through the preserved structural properties of the original code
3Reliability
If shortening and puncturing patterns are not optimized for high-order modulation, then the implementation is simpler, but the reliability and performance deteriorate
Solution Approach 1:
The patent applies local quality by optimizing shortening and puncturing patterns specifically for high-order modulation schemes where bit reliabilities differ. The shortening pattern selects which bits to keep or remove based on their reliability characteristics in the modulation scheme, and puncturing patterns are similarly optimized, ensuring that more reliable bits are preserved and less reliable bits are removed, thereby improving communication reliability for high-order modulations
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method and apparatus for encoding and decoding a channel in a communication system using a Low-Density Parity-Check (LDPC) code. The encoding method includes determining a modulation scheme for transmitting a symbol; determining a shortening pattern in consideration of the determined modulation scheme; grouping columns corresponding to an information word in a parity-check matrix of the LDPC code into a plurality of column groups; ordering the column groups; determining a range of a resulting information word desired to be obtained by shortening the information word; based on the range of the resulting information word, performing column group-by-column group shortening on the ordered column groups of the information word, according to the determined shortening pattern; and LDPC-encoding the shortened information word.