QC-LDPC Base Graph Structure for Flexible 5G Code Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LDPC codes struggle to support encoding and decoding of information bit sequences of various lengths and meet flexible code length and code rate requirements in communication systems.
Innovation Solution
The proposed solution involves using a low-density parity-check LDPC matrix based on a base graph with specific submatrices (A, B, C, D, and E) and orthogonal row structures, along with lifting factors Z to accommodate different code block lengths, and employing row/column permutations for encoding and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional LDPC codes are used, then decoding complexity is reduced, but the ability to support various information bit sequence lengths and flexible code rates is limited
Solution Approach 1:
The LDPC base matrix is divided into five submatrices (A, B, C, D, E) with specific structures. Submatrix A (mA×10) and submatrix B (mA×mA) form the core, while submatrices C, D, and E are added to support flexible code rates. This segmentation allows the system to support various information bit sequence lengths (40-640 bits) and code rates by selectively using different submatrices, while maintaining manageable decoder complexity through the structured approach.
Solution Approach 2:
The base graph design with five submatrices serves multiple functions: submatrix A and B provide the core LDPC structure, submatrix C (all-zero) enables rate matching, submatrix D provides additional parity bits for lower code rates, and submatrix E (identity matrix) supports systematic encoding. This multi-functional design allows a single base graph structure to support various information bit lengths and code rates without requiring multiple specialized decoders.
2Productivity
If LDPC base matrix with QC structure is used, then encoding efficiency is improved, but the flexibility to meet different code length and code rate requirements is reduced
Solution Approach 1:
The base graph structure allows dynamic configuration by adding different numbers of rows to submatrices C, D, and E based on the required code rate and information bit length. The parameters mA (4-7) and mD (0-38) can be adjusted to dynamically adapt the matrix dimensions, enabling the same base graph structure to efficiently encode various code lengths and rates while maintaining the QC structure's encoding efficiency.
Solution Approach 2:
The invention changes key parameters of the base matrix structure: submatrix A has mA rows where 4≤mA≤7, submatrix D has mD rows where 0≤mD≤38, and the total matrix has m rows where 4≤m≤42. By varying these parameters, the system can adjust the code rate and code length to meet different requirements while maintaining efficient QC-based encoding through the preserved base graph structure.
Data Source
Figure 1
Figure 2
Figure 3a
AI summary
This application concerns encoding decoding of LDPC codes.using a low density parity check LDPC matrix, wherein the LDPC matrix is obtained based on a base graph and lifting, and wherein the base graph includes submatrices A, B, C, D, and E, where the submatrix A is a matrix including mA rows and nA columns, mA and nA are positive integers, 4≤mA ≤7, and nA =10; the submatrix B is a matrix including mA rows and mA columns, and the submatrix B includes a column whose weight is 3 and a submatrix B' having a double diagonal structure; the submatrix D includes mD rows in a matrix F, the matrix F is a matrix including mF rows and (mA +nA ) columns, mD and mF are positive integers, 0≤mD ≤mF, and 35≤mF ≤38; the submatrix C is an all zero matrix including mA rows and mD columns; and the submatrix E is an identity matrix including mD rows and mD columns.