Distributed CRC Polar Codes for Early Decoding Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current polar coding methods for 5G eMBB and mMTC communications face challenges in early termination of decoding processes, leading to increased energy consumption and latency due to reliance on single CRC or parity bits, which can result in higher miss detection rates.
Innovation Solution
A method is introduced to distribute error detection and correction bits between control information bits, enabling early termination by arranging them in an upper triangular structure within generators, allowing for error detection checks before or after the first error correction check bit, using CRC, parity, or hash bits for improved reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a single CRC or parity bit is used for early termination of decoding, then decoding speed is improved, but miss detection rate increases
Solution Approach 1:
The patent segments the error detection function by distributing multiple CRC bits (J bits) and error correction bits (J′ bits) throughout the information bit sequence rather than using a single check bit. This segmentation allows multiple independent check points during decoding, enabling early termination while maintaining reliability through distributed verification across different segments of the decoded data.
Solution Approach 2:
The patent applies preliminary action by strategically positioning CRC and error correction bits at specific locations within the bit sequence before decoding begins. The upper triangular structure pre-arranges check bits such that they can be computed and verified at intermediate decoding stages, allowing early termination decisions to be made in advance rather than waiting for complete decoding.
2Use of energy by moving object
If distributed CRC bits are used for early termination, then energy consumption is reduced, but implementation complexity increases
Solution Approach 1:
The patent changes the structural parameter of the generator matrix to an upper triangular form, which enables efficient computation of distributed CRC bits. This parameter change allows check bits to be computed sequentially using previously decoded bits, reducing the computational complexity that would otherwise result from distributed CRC implementation. The triangular structure transforms a potentially complex distributed computation into a series of simple sequential operations.
3Reliability
If multiple error detection and correction bits are distributed throughout the sequence, then reliability is improved, but decoding complexity increases
Solution Approach 1:
The patent segments both error detection (J bits) and error correction (J′ bits) functions and distributes them throughout the information bit sequence. This segmentation creates multiple independent verification points during decoding, improving reliability through distributed checking while allowing early termination when sufficient check bits are available, thereby managing decoding complexity.
Solution Approach 2:
The patent transforms the generator matrix into an upper triangular structure, which simplifies the computation of multiple distributed check bits. This parameter change enables check bits to be computed using only previously decoded bits in a sequential manner, reducing the complexity of implementing multiple distributed error detection and correction functions.
Data Source
AI summary
A method for encoding a sequence of control information bits comprising: generating a sequence of error detection bits based on the sequence of control information bits; generating a sequence of error correction bits based on the sequence of control information bits; and distributing the sequence of error detection bits and the sequence of error correction bits between the sequence of control information bits to form a combined sequence of bits, such that the bit order of the combined sequence of bits following the distribution enables an error detection check to be performed before or after a first error correction check bit.