Pipelined LDPC Decoding Stages for Lower Storage Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The LDPC code requires a large number of decoding iterations and occupies considerable storage capacity, making the implementation of optical fiber communication systems extremely complex.
Innovation Solution
A decoding processing method involving iterative decoding by a plurality of decoders connected in successive stages, with a pipelined approach to achieve successive iteration and stage-by-stage decoding, reducing implementation complexity without requiring large storage capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LDPC code is used for error correction in optical fiber communication, then error correction performance is improved, but system complexity increases due to large number of decoding iterations and storage capacity requirements
Solution Approach 1:
The patent divides the LDPC decoding process into multiple stages, where each stage performs a portion of the required decoding iterations. The input data is processed through several decoding stages in sequence, with each stage handling a subset of the total iterations. This segmentation reduces the storage capacity required at any single stage while maintaining the overall error correction performance of the complete decoding process.
2Reliability
If LDPC code with soft decision decoding is implemented, then bit error rate is reduced, but storage capacity requirements increase considerably
Solution Approach 1:
The patent segments the decoding iterations across multiple stages, where each stage maintains only the storage necessary for its specific iteration range. Instead of requiring all storage for all iterations simultaneously, each decoding stage holds data only temporarily during its processing window, significantly reducing peak storage requirements while preserving soft decision decoding capabilities.
Solution Approach 2:
The patent performs preliminary processing by dividing the input data into multiple groups before decoding begins. Each group is prepared for processing by a specific decoding stage, with data being passed sequentially from one stage to the next. This preliminary organization allows each stage to operate with minimal storage while ensuring all data receives the complete decoding treatment.
3Measurement precision
If multiple decoding iterations are performed, then decoding accuracy is improved, but processing time increases
Solution Approach 1:
The patent divides the multiple decoding iterations into parallel processing streams, where different stages process different portions of the data simultaneously. Each decoding stage performs its iterations in parallel with other stages, reducing the overall processing time while maintaining the total number of iterations required for accurate decoding.
Data Source
AI summary
A decoding processing method a storage medium and an electronic device are disclosed. The method may include: performing an iterative decoding on input data by a plurality of decoders connected in successive stage to obtain a plurality of decoding results of the plurality of decoders; determining a target decoding result from the plurality of decoding results of the plurality of decoders; and outputting the target decoding result.