PAC Code SSCL Decoding With Special Node Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The application of special node decoding in existing channel coding schemes is not possible for polarization-adjusted convolutional (PAC) codes due to the rate-1 convolutional code encoding, leading to issues with output consistency and complexity in decoding.
Innovation Solution
A simplified successive cancellation list (SSCL) decoder is developed for PAC codes, which processes special nodes at the node level by considering the impact of convolutional code encoding, using techniques such as constant codeword outputs for rate-0 nodes, two constant codewords for repetition nodes, and least likelihood ratio scrambling for SPC nodes, while implementing inverse CC encoding with lower complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If special node decoding is applied to PAC codes, then decoding latency is reduced, but output consistency is compromised due to CC encoding
Solution Approach 1:
The patent applies different decoding strategies to different node types within the decoding tree. Rate-0 nodes use direct zero-output, repetition nodes use majority voting, rate-1 nodes use standard polar decoding, and SPC nodes use parity check constraints. This localized adaptation allows each node type to be decoded efficiently while maintaining overall output consistency despite the presence of convolutional code encoding.
Solution Approach 2:
The patent inverts the traditional approach by not trying to make PAC codes conform to standard polar code decoding, but rather by adapting the decoding tree structure and node processing methods to accommodate PAC codes' unique convolutional code encoding characteristics. This inversion enables special node decoding to work correctly for PAC codes while maintaining output consistency.
2Measurement precision
If standard SCL decoding is used for PAC codes, then decoding accuracy is maintained, but device complexity increases
Solution Approach 1:
The patent segments the decoding process into distinct node type handlers within the decoding tree. Each node type (rate-0, repetition, rate-1, SPC) has a specialized decoding routine that can be applied independently. This segmentation allows the decoder to maintain high accuracy for each node type while reducing overall complexity by avoiding unnecessary processing steps for each node type.
Solution Approach 2:
The patent changes the decoding parameters and methods based on node type characteristics. For example, rate-0 nodes skip all processing and output zero, repetition nodes use simple majority voting, and SPC nodes apply parity constraints. These parameter changes allow the decoder to maintain accuracy while significantly reducing the computational complexity compared to uniform SCL decoding.
3Productivity
If special node processing is implemented, then decoding speed is improved, but handling of rate-1 nodes becomes non-linear and complex
Solution Approach 1:
The patent makes the rate-1 node decoder universal by implementing a single decoding routine that handles both standard polar code rate-1 nodes and PAC code rate-1 nodes. This multi-functional approach maintains linearity and simplicity while improving decoding speed, avoiding the need for separate complex handling procedures for different node types.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and devices are provided in which a channel encoded with a polarization-adjusted convolutional (PAC) code is received. A decoded codeword is generated based at least in part on simplified successive cancellation list (SSCL) decoding performed on the channel via a decoding tree. The decoding tree includes a node that generates candidate codeword output based on predefined processing using convolutional code (CC) state input and channel vector input. A sub-tree of the node remains unprocessed.