Polar Code Decoder Parallel Branching Before First Split
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mechanisms for decoding polar codes are prone to error propagation and limited parallelization due to dependencies between processes, leading to inefficiencies in decoding performance.
Innovation Solution
The method involves decoding an encoded sequence using a polar code with a given list size, where the sequence is decoded by at least half the list size of processing units in parallel until the first branching, reducing the need for inter-process communication and enabling independent parallel processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional successive decoding is used for polar codes, then the decoding follows the bit-order of the polar codes, but error propagation occurs and decoding performance deteriorates
Solution Approach 1:
The patent segments the decoding process into multiple independent parallel paths instead of a single sequential path. By dividing the successive decoding into parallel branches that can be processed simultaneously, the system maintains decoding accuracy while improving efficiency. Each parallel path processes different portions of the code independently, preventing error propagation from affecting the entire decoding process.
2Productivity
If branching with limited candidate branches is used, then the number of candidate decoded sequences is controlled, but parallelization is limited due to dependencies between processes
Solution Approach 1:
The patent segments the decoding process into multiple independent parallel paths instead of a single sequential path. By dividing the successive decoding into parallel branches that can be processed simultaneously, the system maintains decoding accuracy while improving efficiency. Each parallel path processes different portions of the code independently, preventing error propagation from affecting the entire decoding process.
Solution Approach 2:
The patent performs preliminary actions by pre-determining the branching structure and candidate paths before parallel processing begins. This allows the system to set up independent parallel processes with predetermined branching criteria, reducing the need for complex inter-process communication during execution and enabling more effective parallelization.
3Measurement precision
If communication between processes is required at every bit-position, then candidate branches can be compared, but computation stalls and parallelization gain is limited
Solution Approach 1:
The patent performs preliminary actions by pre-determining the branching structure and candidate paths before parallel processing begins. This allows the system to set up independent parallel processes with predetermined branching criteria, reducing the need for complex inter-process communication during execution and enabling more effective parallelization.
Solution Approach 2:
The patent extracts the comparison and metric evaluation operations from the critical parallel processing path. By performing metric calculations and comparisons at predetermined branching points rather than continuously during parallel execution, the system eliminates computation stalls and maintains high parallelization gain while still ensuring accurate metric comparison for selecting candidate branches.
Data Source
AI summary
There is provided mechanisms for decoding an encoded sequence into a decoded sequence. A method is performed by an information decoder. The method comprises obtaining the encoded sequence. The encoded sequence has been encoded using a polar code. The method comprises successively decoding the encoded sequence into the decoded sequence. The decoding is performed for a given list size, LS, where LS>1, defining how many candidate decoded sequences in total the thus far decoded sequence is allowed to branch into during the decoding. The encoded sequence is decoded, until its first branching, by at least as many processing units in parallel as a factor, f, of the given list size. The factor is at least half the given list size, f≥LS/2.


