Polar Code CRC Interleaving for More Reliable SCL Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The performance of CA-polar codes in wireless communication systems is limited by the deletion of correct paths during decoding due to low metrics at intermediate nodes, which affects error correction and detection efficiency.

Innovation Solution

The proposed solution involves interleaving CRC bits between information bits and setting them based on the values of preceding information bits, using these CRC bits as parity check bits to aid SCL decoding, thereby increasing the probability of correctly deleting incorrect paths and improving decoding performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional CA-polar coding is used, then the decoding process is simpler, but the error correction performance is limited due to correct paths being deleted at intermediate nodes

Engineering Contradiction:
Improveerror correction performanceVSAvoiddecoding process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The CRC bits are segmented and interleaved with information bits, creating distributed parity check bits throughout the codeword. This segmentation allows intermediate nodes to perform local parity checks without waiting for the complete decoding process, preventing premature deletion of correct paths while maintaining manageable complexity through localized operations

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Parity check bits are calculated and inserted into the codeword before the decoding process begins. These pre-placed parity bits enable intermediate decoding nodes to perform validation checks in advance, identifying and preserving correct paths before the final decoding stage, thereby improving reliability without significantly increasing overall complexity

Inventive Principle:
Principle #10Preliminary action

2Reliability

If CRC bits are placed only at the end of the codeword, then the encoding process is simpler, but the error detection capability is reduced

Engineering Contradiction:
Improveerror detection capabilityVSAvoidencoding process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The single block of CRC bits at the end is segmented and distributed throughout the codeword at regular intervals. This segmentation transforms the error detection capability from a final-stage check to continuous distributed validation, enabling intermediate nodes to detect errors locally without significantly complicating the encoding process

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The error detection function is extended from a single temporal point (end of codeword) to multiple spatial dimensions (distributed throughout the codeword). This dimensional transformation allows error detection to occur at multiple positions simultaneously, enhancing detection capability while maintaining encoding simplicity through systematic placement patterns

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3579423B1Polar code encoding and decoding methods and devices
Publication Date: 2024.10.02 HUAWEI TECH CO LTD
  • EP3579423B1 patent drawingFigure 1~2(b)
  • EP3579423B1 patent drawingFigure 3~5
  • EP3579423B1 patent drawingFigure 6~8

AI summary

Embodiments of this application provide a polar encoding method, including: performing CRC encoding on an information block to obtain a CRC encoded block with a length of B, where a CRC length is Lcrc, an information block length is K, and B=K+Lcrc; interleaving the CRC encoded block, where Lpc CRC bits in an interleaved encoded block are located between bits of the information block, each CRC bit of the Lpc CRC bits is located after all bits checked by using the CRC bit, and Lpc is an integer greater than 0 and less than Lcrc; and mapping the interleaved encoded block to information bits, setting a frozen bit to an agreed fixed value, and performing polar encoding on the information bits and the frozen bit to obtain a polar encoded codeword, where positions of the information bits are positions corresponding to B polarized channels with optimal reliability, a position of the frozen bit is a position corresponding to one of remaining N-B polarized channels, and N is a mother code length of a polar code. The encoding method can further improve performance of a CA-polar code.