Long LDPC Code Layout for 7/8 Rate Without Puncturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The implementation of different LDPC codeword lengths for various code rates in wireless networks, such as in the IEEE 802.11ay standard, leads to inefficiencies in blocking and de-blocking processes, particularly for the 7/8 rate code, which results in increased redundancy and complexity during encoding and single carrier blocking.
Innovation Solution
A high rate, long LDPC code with a codeword length of 1344 bits is developed, maintaining consistency with lower rate codes, using a two-step lifting process and optimized lifting matrices to generate a 7/8 rate LDPC matrix that preserves the codeword length, thereby simplifying encoding and reducing redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If different LDPC codeword lengths are used for different code rates, then code rate flexibility is improved, but implementation complexity and redundancy increase
Solution Approach 1:
The patent applies universality by designing a unified LDPC coding scheme where the same codeword length (1344 bits) is used across multiple code rates (1/2, 5/8, 3/4, 7/8). This is achieved through a systematic construction method using base matrices and lifting operations that generate consistent codeword lengths regardless of the target code rate, thereby simplifying implementation while maintaining code rate flexibility
Solution Approach 2:
The patent employs parameter changes by modifying the lifting factor and base matrix structure to achieve different code rates from the same codeword length framework. By adjusting the lifting factor (e.g., from 672 to 1344) and selecting appropriate base matrices, the system maintains consistent codeword length while varying the code rate, thus resolving the contradiction between flexibility and complexity
2Adaptability or versatility
If different LDPC codeword lengths are used for different code rates, then code rate adaptation is improved, but spectrum efficiency deteriorates due to increased padding
Solution Approach 1:
The unified codeword length approach enables a single coding framework to serve multiple code rates without requiring different codeword lengths, thereby eliminating the need for extensive zero-padding and improving spectrum efficiency while maintaining code rate adaptation capability
Solution Approach 2:
Instead of using excessive zero-padding to achieve different code rates with varying codeword lengths, the patent employs a more efficient partial action approach where the same codeword length is used and code rate variation is achieved through systematic construction with different base matrices and lifting factors, reducing unnecessary padding and improving spectral efficiency
3Device complexity
If consistent codeword length is used across different code rates, then implementation complexity is reduced, but code rate flexibility may be limited
Solution Approach 1:
The patent resolves this contradiction by changing key parameters such as the lifting factor and base matrix selection to achieve different code rates within the unified codeword length framework. For example, by adjusting the lifting factor from 672 to 1344 and selecting appropriate base matrices, the system maintains consistent codeword length while providing flexibility across code rates 1/2, 5/8, 3/4, and 7/8
4Adaptability or versatility
If punctured solutions are used to achieve high code rates, then code rate flexibility is improved, but performance deteriorates
Solution Approach 1:
Instead of using puncturing (removing bits) to achieve high code rates, the patent inverts the approach by using systematic construction with carefully designed base matrices and lifting operations to directly generate the desired code rate with full codeword length preserved. This inversion avoids the performance degradation associated with puncturing while maintaining code rate flexibility
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Methods and devices are disclosed for encoding source words and decoding codewords with LDPC matrices.