Parallel SISO Decoder Layout for Faster LDPC Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LDPC error correction systems face inefficiencies in memory usage and processing speed due to high memory requirements and complex permutation handling during iterative decoding processes, particularly in fast architectures.
Innovation Solution
A parallel SISO decoder structure that updates Log-likelihood-ratios in a single operation, utilizing a difference array and current array with low and high resolution memory respectively, and efficiently handling permutations by storing variable nodes in their last accessed permutation, reducing the need for multiple permutation blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If fast LDPC decoder architecture is used to increase processing speed, then decoding speed is improved, but memory requirements and complexity increase
Solution Approach 1:
The decoder is divided into multiple parallel SISO (Soft-In Soft-Out) units that process different portions of the codeword simultaneously. Each SISO unit handles a specific set of parity check equations, allowing the decoding process to be segmented into independent parallel operations that reduce overall memory requirements while maintaining high processing speed
Solution Approach 2:
The patent introduces a new dimension of parallelism by processing multiple parity equations simultaneously across multiple SISO units. This dimensional expansion of the decoding architecture allows throughput to be increased without proportionally increasing memory resources, as the parallel units share common memory structures through time-multiplexed access patterns
2Reliability
If traditional two-pass Tanner Graph decoding is used, then decoding accuracy is maintained, but processing time increases
Solution Approach 1:
Multiple SISO units are merged into a single iterative decoding framework where they collectively perform the functions of both traditional passes. The parallel SISO units process different parity check equations simultaneously and exchange messages through shared memory structures, combining their computational power to achieve accurate decoding in fewer iterative cycles than traditional sequential approaches
Solution Approach 2:
The parallel SISO architecture enables continuous processing of multiple codewords and parity equations without idle cycles. While one SISO unit is computing, others are accessing memory or preparing data, ensuring that all computational resources are continuously utilized. This continuous action maintains decoding accuracy while reducing total processing time through efficient resource utilization
3Measurement precision
If high resolution memory is used for all data storage, then measurement precision is improved, but memory capacity requirements increase
Solution Approach 1:
Different memory structures are assigned different resolution requirements based on their specific functions. The extrinsic information memory requires high resolution to maintain decoding accuracy, while the check message memory and other auxiliary structures use lower resolution. This local differentiation of quality allows the system to maintain necessary precision where critical while reducing overall memory capacity requirements in non-critical areas
Data Source
AI summary
The current invention involves a forward error detection system, especially for use with Low Density Parity Check codes. A parallel SISO structure allows the decoder to process multiple parity equations at the same time. There is a new SISO decoder which allows for the updating of the Log-likelihood-ratio's in a single operation, as opposed to the two pass traditionally associated with the Tanner Graph's. In the decoder, there is a mapping structure that correctly aligns the stored estimates, the stored differences and the SISO's. There is also the ability to deal with multiple instances of the same data being processed at the same time. This structure manages the updates and the differences in such a manner that all calculations on a single piece of data that are processed in parallel are incorporated correctly in the new updated estimates.