Convolutional LDPC Decoding with Iterative LLR Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional convolutional LDPC decoding methods in high-speed optical transmission systems suffer from inaccuracies, leading to incorrect decoding results and reduced transmission performance.
Innovation Solution
The method involves adjusting the log-likelihood ratio (LLR) based on all check node information and symbol values, using a symbol caching unit and re-demapping units to improve decoding accuracy, by determining extrinsic information and performing iterative calculations to refine the decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional convolutional LDPC decoding methods are used in high-speed optical transmission systems, then the decoding process can be implemented, but the decoding accuracy is insufficient leading to incorrect decoding results
Solution Approach 1:
The patent implements an iterative decoding process where the LLR values are repeatedly updated based on feedback from check node information. The decoding unit uses the initial LLR values, performs iterative calculations with check node information to generate updated LLR values, and continues this process until a stopping condition is met, thereby improving decoding accuracy through feedback-driven refinement
Solution Approach 2:
The patent performs preliminary processing of received signal values to generate initial LLR (log-likelihood ratio) values before the main decoding process. This preliminary action includes calculating initial LLR values from the received signals and storing them, which prepares the data in an optimized format for subsequent iterative decoding operations
2Measurement precision
If iterative calculations with check node information are performed, then decoding accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent segments the decoding process into distinct functional units: a receiving unit for initial LLR calculation, a decoding unit for iterative processing, and a determining unit for final result extraction. The check node information is also segmented into multiple pieces that are processed iteratively. This segmentation allows the complex decoding task to be divided into manageable stages, reducing overall computational complexity
3Measurement precision
If all check node information is utilized in LLR adjustment, then decoding precision is enhanced, but the processing time increases
Solution Approach 1:
The patent employs periodic iterative updates of LLR values using check node information. Instead of processing all check node information in a single pass, the decoding unit performs multiple iterative rounds where LLR values are updated periodically based on subsets of check node information. This periodic action allows thorough utilization of all check node information while managing processing time through structured iteration
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present invention provide a convolutional LDPC decoding method and apparatus, a decoder, and a system that relate to the field of communications technologies, so as to improve accuracy of a decoding process. The method according to the embodiments of the present invention includes: receiving first check node information including all check node information that is corresponding to a target bit and output by a previous-level decoding unit; reading, from a symbol caching unit, a symbol value corresponding to the target bit; determining an LLR and a first APP based on the first check node information and the symbol value, where the first APP is used to indicate a decoding result of the target bit at a first moment; and sending the first check node information and the first APP to a next-level decoding unit, so that the next-level decoding unit obtains a second APP and second check node information based on the first check node information and the first APP, where the second APP is used to indicate a decoding result of the target bit at a second moment, and the second check node information includes all check node information that is corresponding to the target bit and output by the next-level decoding unit.