Adjusted Min-Sum LDPC Decoding with Adaptive LLR Magnitudes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current decoding algorithms for LDPC and Turbo codes, such as BCJR/SP decoders, are complex and not easily implementable, leading to degraded performance at lower rates, especially in wireless communication systems like 5G NR, which require efficient and high-performance encoding and decoding solutions.
Innovation Solution
The implementation of an adjusted minimum-sum (AdjMS) algorithm that approximates an update function to determine magnitudes of outgoing log likelihood ratios (LLRs), allowing for efficient decoding of both LDPC and Turbo codes, and providing a unified decoder with high performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If BCJR/SP decoding algorithms are used, then decoding performance is improved, but algorithm complexity increases
Solution Approach 1:
The patent modifies the minimum-sum decoding algorithm by introducing a scaling factor parameter that adjusts the magnitude of outgoing LLRs. This parameter change allows the simplified algorithm to achieve performance closer to BCJR/SP while maintaining lower complexity. The scaling factor is optimized based on channel conditions and code characteristics to balance performance and complexity.
Solution Approach 2:
The patent uses a simplified approximation of the BCJR/SP algorithm that requires less computational resources and can be discarded or reinitialized more easily. The modified minimum-sum algorithm uses simpler operations that are computationally cheaper, making it suitable for resource-constrained environments while still providing acceptable decoding performance.
2Device complexity
If modified minimum-sum algorithm is used, then algorithm complexity is reduced, but decoding performance deteriorates
Solution Approach 1:
The patent incorporates feedback mechanisms where the decoder monitors decoding performance and adjusts the scaling factor accordingly. By observing the relationship between incoming and outgoing LLRs, the system dynamically tunes the algorithm parameters to maintain optimal performance while keeping complexity low. This feedback loop allows the simplified algorithm to adapt and compensate for its inherent approximations.
Solution Approach 2:
The patent makes the minimum-sum algorithm dynamic by introducing adaptive scaling factors that change based on operating conditions. Rather than using fixed thresholds or static parameters, the algorithm dynamically adjusts its behavior based on channel conditions, code rate, and iteration number, allowing it to maintain high performance across varying scenarios while keeping implementation complexity manageable.
Data Source
Figure 1~2
Figure 3
Figure 4~4A
AI summary
Certain aspects of the present disclosure generally relate to techniques for efficient, high-performance decoding of low-density parity check (LDPC) codes, for example, by using an adjusted minimum-sum (AdjMS) algorithm, which involves approximating an update function and determining magnitudes of outgoing log likelihood ratios (LLRs). Similar techniques may also be used for decoding turbo codes. Other aspects, embodiments, and features (such as encoding technique) are also claimed and described.