LDPC Transport Block Segmentation With Minimal Shortening Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LTE turbo codes exhibit limited performance improvement beyond a certain signal-to-noise ratio, leading to high error rates and increased complexity, which is detrimental in future-generation communication systems requiring ultra-reliable and low-latency data transmission.
Innovation Solution
A transport block segmentation method for LDPC codes is introduced, where the block is segmented based on multiple information bit lengths, with the processor determining the smallest number to divide the block, selecting appropriate information bit lengths, and adjusting the length of shortening bits to minimize overhead and prevent performance degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LTE turbo code is used, then error rate is reduced, but complexity increases and performance improvement is not remarkable beyond a predetermined SNR region
Solution Approach 1:
The transport block is divided into multiple code blocks, each of which is further divided into sub-code blocks. This segmentation allows the LDPC encoder to process smaller blocks independently, reducing overall coding complexity while maintaining reliability through distributed error correction across multiple blocks.
Solution Approach 2:
The patent employs multiple parity check matrices with different parameters (different sizes and structures) to encode different sub-code blocks. By changing the parameters of the parity check matrices based on the transport block size and code rate, the system optimizes performance across different SNR regions while controlling complexity.
2Reliability
If code size is increased to reduce error rate, then reliability improves, but latency increases due to processing time
Solution Approach 1:
By dividing the transport block into multiple smaller sub-code blocks that can be encoded and decoded in parallel, the patent reduces processing latency while maintaining the overall error correction capability equivalent to a single large block. Each sub-code block can be processed independently and simultaneously.
Solution Approach 2:
The patent applies partial processing by encoding only the necessary portions of the transport block into multiple sub-code blocks with appropriate sizes. This allows the system to achieve the required reliability level without processing the entire block as one unit, thereby reducing latency.
3Reliability
If transport block is segmented into multiple code blocks, then error correction performance improves, but overhead increases due to multiple CRC bits
Solution Approach 1:
The patent merges the error correction functionality across multiple sub-code blocks while strategically placing CRC bits. By combining the error correction benefits of multiple blocks with an optimized CRC placement strategy, the system achieves improved error correction performance with minimized overhead compared to traditional approaches where each block has full CRC protection.
Data Source
AI summary
Disclosed are a method for dividing a transport block of a low density parity check (LDPC) code and an apparatus therefor. A method for dividing a transport block of an LDPC code according to the present disclosure can improve the performance of the LDPC code by dividing the transport block using a minimum number of code blocks. In addition, it is possible to minimize shortening bits by making the size of some of the code blocks smaller than the size of the other code blocks. Further, it is possible to prevent performance degradation due to a minimum size code block by minimizing the number of the code blocks and performing shortening on a large size code block.