Polar Code List Decoding with Threshold-Based Branching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mechanisms for decoding polar codes are prone to error propagation due to the successive decoding procedure, which can lead to incorrect decoding and inefficiencies, especially when the probability measures of binary digits are close but not exact.
Innovation Solution
A method that performs a threshold check on successive binary digits in the encoded sequence, branching into two candidate decoded sequences when the threshold check fails, based on instantaneous bit uncertainty, allowing for parallel decoding and improving accuracy and complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If successive decoding is performed following the bit-order of polar codes, then decoding complexity is reduced, but error propagation occurs leading to decreased decoding accuracy
Solution Approach 1:
The patent segments the decoding process by introducing a list of candidate sequences instead of a single sequential decoding path. The binary tree of possible sequences is divided into multiple candidate branches, allowing the decoder to explore alternative paths when errors are detected, thereby preventing error propagation while maintaining manageable complexity through controlled list expansion.
Solution Approach 2:
The patent implements dynamic adaptation in the decoding process by adjusting the list of candidate sequences based on instantaneous channel conditions. The decoder dynamically expands or prunes the candidate list depending on the reliability of received bits, making the decoding process adaptive rather than static, which improves accuracy without proportionally increasing complexity.
2Reliability
If a fixed list of binary digit indices is used to mitigate error propagation, then decoding accuracy improves, but the list may be mismatched to instantaneous channel realization reducing effectiveness
Solution Approach 1:
The patent transforms the static fixed list approach into a dynamic process where the candidate sequence list is adaptively adjusted based on instantaneous channel conditions. The decoder monitors channel reliability metrics and dynamically expands the candidate list when uncertainty is detected, ensuring the decoding process adapts to varying channel realizations rather than relying on a predetermined fixed list.
Solution Approach 2:
The patent incorporates feedback mechanisms where the decoding process continuously monitors the reliability of received bits and uses this information to adjust the candidate sequence list. This feedback loop allows the system to learn from instantaneous channel conditions and adjust its decoding strategy accordingly, improving adaptability to different channel realizations.
3Measurement precision
If probability measures are used to evaluate binary digit reliability, then decoding decisions can be made, but errors in probability measurement may result in selecting wrong paths
Solution Approach 1:
The patent segments the single probability measurement approach into multiple candidate sequence evaluations. Instead of relying on a single probability measure that may be inaccurate, the system maintains multiple candidate sequences with different probability assessments, allowing it to select from multiple possibilities rather than committing to a single potentially erroneous path.
Solution Approach 2:
The patent prepares multiple candidate sequences in advance as a cushion against probability measurement errors. By maintaining a list of alternative candidate sequences before making final decoding decisions, the system creates a safety buffer that protects against selecting wrong paths due to inaccurate probability measurements, allowing for correction if the initial probability assessment is erroneous.
Data Source
AI summary
A method is performed by an information decoder. The method comprises obtaining (S102) an encoded sequence having been encoded using a polar code. The method comprises successively decoding (S104) the encoded sequence into the successive bits of the decoded sequence. Successively decoding the encoded sequence comprises performing a threshold check (S106) for evaluating a bit uncertainty criterion. Successively decoding the encoded sequence comprises branching (S108) the decoded sequence into two candidate decoded sequences whenever the threshold check fails.