Polar Code Subblock Replication for Lower HARQ Decoding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current polar code encoding and decoding methods in communication technologies, particularly in HARQ systems, face high encoding and decoding complexity due to the information coupling relationships between subblocks, which increases computational overhead and reduces system performance.
Innovation Solution
The proposed method involves generating an input vector with subblocks where information bits are replicated based on a mapping relationship between subchannels, allowing for a coupling relationship between subblocks, thereby reducing encoding and decoding complexity while maintaining performance similar to IR-HARQ systems through the use of polar encoding and decoding processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If information coupling relationships between subblocks are established in HARQ retransmission, then encoding gain and decoding performance are improved, but encoding and decoding complexity are increased
Solution Approach 1:
The input vector is segmented into multiple subblocks, where each subblock contains information bits that are replicated from corresponding positions in previous subblocks. This segmentation allows the decoder to process each subblock independently while maintaining the coupling relationship, thereby reducing decoding complexity compared to processing the entire codeword as a single unit.
Solution Approach 2:
Information bits are replicated in advance during the encoding process to establish coupling relationships between subblocks. The encoder pre-positions these replicated bits in the input vector before polar encoding, which simplifies the decoding process by providing the decoder with predetermined relationships between subblocks that can be exploited without increasing computational complexity.
2Reliability
If information coupling relationships between subblocks are established in HARQ retransmission, then encoding gain and decoding performance are improved, but decoding complexity is increased
Solution Approach 1:
The codeword is divided into multiple subblocks that can be decoded sequentially or in parallel. Each subblock contains replicated information bits from previous subblocks, allowing the decoder to process smaller units independently while maintaining the benefits of coupling relationships, thus reducing overall decoding complexity.
Solution Approach 2:
Information bits are copied from previous subblocks to current subblocks at corresponding positions. This copying mechanism establishes the coupling relationship without requiring complex processing during decoding, as the replicated bits provide redundant information that can be used directly by the decoder to improve reliability without increasing computational burden.
3Device complexity
If first information bit of first subblock is obtained by replicating second information bit of second subblock, then encoding/decoding complexity is reduced through coupling relationship, but hardware overhead increases
Solution Approach 1:
The encoding process merges the information from previous subblocks with current subblock data by replicating information bits at corresponding positions. This merging occurs in the data domain rather than requiring additional hardware components, thus reducing encoding complexity while the overhead is minimal and confined to storing and accessing the replicated bit values.
Data Source
AI summary
Embodiments of this application disclose a polar code encoding method, a polar code decoding method, and apparatuses thereof, to reduce encoding and decoding complexity. The method in embodiments of this application includes: generating an input vector, where the input vector includes T subblocks, a first information bit of a first subblock is obtained by replicating a second information bit of a second subblock, the first subblock and the second subblock are subblocks of the T subblocks, a sequence number of the first subblock is after a sequence number of the second subblock, and T is an integer greater than or equal to 2; and performing polar encoding on the input vector to obtain an encoded bit.