CRC-Polar Encoding with Segmented CRC for Error Floor Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, the concatenation of a Cyclic Redundancy Check (CRC) code with a polar code experiences an error floor phenomenon due to short CRC code lengths, leading to performance degradation and increased error detection failure probabilities.
Innovation Solution
The method involves performing CRC coding on a determined number of information bits and polar coding on both the CRC-coded and remaining information bits, optimizing the CRC code length to improve error rate performance by adjusting the number of bits subjected to CRC coding based on the polar code generator matrix and Hamming distance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a short CRC code length is used in the concatenation of CRC code and polar code, then the device complexity is reduced, but the error rate performance deteriorates due to error floor phenomenon
Solution Approach 1:
The information bits are divided into multiple groups, and CRC codes are independently generated for each group rather than for the entire information block. This segmentation allows each CRC code to be shorter while collectively providing comprehensive error detection coverage, thus reducing individual CRC length without sacrificing overall error detection capability.
Solution Approach 2:
The error detection approach transitions from a single-dimensional check (one long CRC for all information bits) to a multi-dimensional check (multiple short CRCs for different groups of information bits). This dimensional change in the error detection structure enables shorter CRC codes while maintaining or improving error detection performance.
2Reliability
If a long CRC code length is used in the concatenation of CRC code and polar code, then the error detection capability is improved, but the device complexity and processing overhead increase
Solution Approach 1:
Instead of using one long CRC code, the information bits are segmented into groups with separate CRC codes for each group. This segmentation achieves comprehensive error detection capability equivalent to or better than a single long CRC, while significantly reducing the length and complexity of each individual CRC code.
Solution Approach 2:
Multiple short CRC codes are applied to different portions of information bits rather than one comprehensive CRC. This partial application of CRC to segments provides sufficient error detection coverage without the overhead of a single extensive CRC code.
3Reliability
If CRC coding is performed on all information bits, then the error detection coverage is maximized, but the processing time and complexity increase
Solution Approach 1:
The information bits are divided into multiple groups that can be processed independently with separate CRC codes. This segmentation enables parallel processing of different bit groups, reducing the total processing time while maintaining comprehensive error detection coverage across all information bits.
Solution Approach 2:
CRC codes are generated for each group of information bits independently and in advance before the final polar encoding. This preliminary generation of multiple short CRCs reduces the overall processing complexity compared to generating one long CRC for all bits, as the shorter codes can be computed more quickly.
Data Source
AI summary
An encoding method for encoding input information bits using an encoder implemented with concatenation of a CRC-α coder and a polar coder is provided. The method includes performing Cyclic Redundancy Check (CRC) coding on as many information bits as a determined number of CRC coding bits among input information bits and performing polar coding on the CRC-coded information bits and other information bits than the CRC-coded information bits.