Soft-Decision Linear Code Decoding With Early Cost Pruning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current soft decision decoding methods, such as the ordered statistics decoder (OSD), face high computational complexity due to the need to check all combinations, which hinders efficient decoding performance.
Innovation Solution
A method and apparatus for fast decoding a linear code based on a soft decision that involves arranging received signals by magnitude, performing a hard decision, identifying a most reliable base, and using an error vector to calculate a cost function, with speed-up conditions to reduce unnecessary computations, such as omitting operations when the Hamming distance exceeds a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ordered statistics decoder (OSD) checks all combinations to find the most similar codeword, then decoding performance is improved, but computational complexity increases significantly
Solution Approach 1:
The patent segments the decoding process into distinct stages: hard decision stage, error vector generation stage, and cost calculation stage. By dividing the exhaustive search into manageable segments with early termination conditions, the computational complexity is reduced while maintaining decoding performance through systematic exploration of candidate codewords.
Solution Approach 2:
The patent applies partial action by calculating cost functions for only the most promising candidate codewords rather than all possible combinations. The method generates error vectors of varying weights and evaluates candidates in order of likelihood, performing partial evaluations that suffice to achieve near-optimal decoding without exhaustive computation.
2Reliability
If soft decision decoding is used instead of hard decision, then decoding performance is improved, but data processing requirements increase
Solution Approach 1:
The patent extracts the essential soft decision information by generating error vectors based on reliability metrics from soft decisions, then takes out only the critical components needed for decoding. The cost function calculation focuses on the most significant error patterns, extracting and processing only the necessary data elements rather than all soft decision information.
Solution Approach 2:
The patent changes parameters by representing soft decision information through error vectors with specific Hamming weights rather than processing full soft decision values. The cost function uses weighted sums that transform the parameter representation, enabling efficient computation while preserving the performance benefits of soft decision decoding.
3Measurement precision
If all cost function calculations are performed for every candidate, then decoding accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary actions by generating and sorting error vectors in advance based on their Hamming weights, and pre-calculating which candidate codewords are most likely to be correct. This preliminary organization allows the decoder to evaluate candidates in optimal order and terminate early when the minimum cost is found, avoiding unnecessary calculations.
Solution Approach 2:
The patent implements skipping by implementing early termination in the cost calculation process. When a candidate codeword achieves the minimum cost value, the algorithm rushes through to completion without evaluating remaining candidates. The method skips unnecessary cost calculations for candidates that cannot improve upon the current minimum, significantly reducing processing time.
Data Source
AI summary
Disclosed are a method and an apparatus for high-speed decoding of a linear code on the basis of a soft decision. The method for high-speed decoding of a linear code on the basis of a soft decision may comprise the steps of: obtaining an alignment signal by aligning received signals in order of magnitude; obtaining a hard decision signal by making a hard decision on the alignment signal; obtaining a higher-level signal corresponding to most reliable bases (MRB) from the hard decision signal; obtaining a permuted and corrected codeword candidate by using an error vector according to a current order and the higher signal; calculating a cost for the current order by using a cost function; determining the permuted and corrected codeword candidate as a permuted and corrected codeword according to a result of comparing the calculated cost and the minimum cost; and determining a predefined high-speed condition.