LDPC Parity Check Matrix Layout for High-Throughput 5G Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional LDPC encoding methods struggle to achieve high throughput in the transmission of large blocks with relatively long lengths, particularly in 5G communication systems where diverse code rates are required to support wide coverage.
Innovation Solution
A method for generating a parity check matrix based on a 46 × 68 base matrix, where each element corresponds to a shift index value determined by a modulo operation between the base matrix and a lifting value, associated with the length of the input data, is used for encoding in user equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional LDPC encoding methods are used, then encoding can be performed with standard complexity, but throughput for large block transmission is insufficient
Solution Approach 1:
The parity check matrix is divided into sub-matrices, where each sub-matrix corresponds to a specific block of input data. This segmentation allows the encoding process to be divided into multiple independent stages, enabling parallel processing and improving throughput for large block lengths while maintaining manageable complexity at each stage.
Solution Approach 2:
The patent transforms the traditional single-matrix encoding approach into a multi-dimensional structure by organizing the parity check matrix into sub-matrices arranged in a specific pattern. This dimensional reorganization enables more efficient memory access patterns and computational workflows, thereby increasing throughput without proportionally increasing complexity.
2Adaptability or versatility
If a fixed parity check matrix structure is used, then implementation is simplified, but adaptability to diverse code rates is limited
Solution Approach 1:
The parity check matrix structure is designed to be dynamic rather than fixed. Sub-matrices can be selectively activated or deactivated based on the required code rate, allowing the same base structure to adapt to diverse transmission requirements. This dynamic configuration enables support for multiple code rates without requiring completely different matrix structures for each rate.
Solution Approach 2:
The patent employs parameter changes in the parity check matrix construction, where certain parameters such as sub-matrix dimensions, positions, and connectivity patterns are adjusted according to the target code rate. This allows a single base matrix structure to generate multiple effective parity check matrices for different code rates by simply changing structural parameters rather than the entire matrix.
3Productivity
If encoding is optimized for high throughput, then data transmission efficiency improves, but implementation complexity in user equipment increases
Solution Approach 1:
By segmenting the encoding process into sub-matrix operations, the patent reduces the computational burden on individual processing units while maintaining high overall throughput. Each sub-matrix operation is simpler and more suitable for implementation in user equipment with limited resources, yet the collective effect achieves high data transmission efficiency.
Solution Approach 2:
The segmented sub-matrix structure enables self-service processing where each sub-matrix operation can be independently executed and combined. This modular approach allows user equipment to implement encoding using available resources without requiring a monolithic complex encoder, thereby improving ease of implementation while maintaining throughput.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for performing encoding on the basis of a parity check matrix of a LDPC code, according to one embodiment of the present invention, comprises the steps of: generating, by a terminal, a parity check matrix, wherein the parity check matrix corresponds to a characteristic matrix, each element of the characteristic matrix corresponds to a shift index value determined by a modulo operation between a corresponding element in a base matrix and a lifting value, and the base matrix is a 46 x 68 matrix; and performing, by the terminal, encoding of input data by using the parity check matrix, wherein the lifting value is associated with the length of the input data.