Polar Coding Bit Placement Using Reliability Sequence Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current polar code reliability estimation methods, such as Bhattacharyya, Density Evolution, and Gaussian Approximation, have limitations in accurately estimating channel reliability for non-binary erasure channels, leading to low performance and high storage complexity due to excessive storage overheads.
Innovation Solution
A method to determine the location of information bits in polar codes based on pre-defined sequences of polar channel sequence numbers, reducing storage complexity and improving performance by using sequences from tables or calculated using reliability measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Bhattacharyya parameter method is used for polar channel reliability estimation, then the method is simple to implement, but it cannot accurately estimate reliability of non-binary erasure channels resulting in low performance
Solution Approach 1:
The patent transitions from using Bhattacharyya parameters to using Log-Likelihood Ratio (LLR) statistics as the fundamental parameter for channel reliability estimation. This parameter change enables accurate modeling of non-binary erasure channels while maintaining computational feasibility through efficient LLR update algorithms.
Solution Approach 2:
The patent replaces the traditional Bhattacharyya parameter-based mechanical estimation approach with a probabilistic model based on LLR statistics. This substitution allows for more accurate representation of channel characteristics in non-binary erasure channels by capturing the statistical distribution of channel outputs.
2Measurement precision
If Density Evolution or Gaussian Approximation methods are used for polar channel reliability estimation, then estimation accuracy is improved, but calculation complexity increases making them not applicable to online calculation
Solution Approach 1:
The patent extracts the essential reliability information from complex DE or GA computations by focusing only on the first-order LLR statistics (mean and variance). This extraction approach captures the dominant reliability characteristics without requiring the full computational machinery of DE or GA methods.
Solution Approach 2:
The patent performs preliminary channel polarization analysis offline to determine the optimal information bit positions based on LLR statistics. These pre-computed positions are then stored and directly applied during online encoding operations, eliminating the need for real-time reliability calculations while maintaining high estimation accuracy.
3Reliability
If DE method or GA method is used for offline storage with varying parameters, then reliable estimation is achieved, but storage overhead becomes excessively large
Solution Approach 1:
The patent develops a universal LLR-based reliability estimation framework that can handle various channel types (binary erasure, non-binary erasure, AWGN) and different code rates through a single unified model. This universality eliminates the need to store separate reliability data for each parameter combination, significantly reducing storage overhead while maintaining estimation reliability.
Solution Approach 2:
The patent changes the storage representation from storing complete reliability curves or multiple parameter-specific tables to storing compact LLR statistical parameters (mean and variance) that can represent multiple channel conditions. This parameter transformation enables a single compact table to serve multiple purposes across different operating conditions.
Data Source
AI summary
Embodiments can provide a coding method, a coding apparatus, and a communications apparatus. The method includes: determining N to-be-coded bits, where N is a positive integer; obtaining a first sequence that includes N polar channel sequence numbers; determining a location of an information bit in the N to-be-coded bits based on the first sequence; and performing polar coding on the N to-be-coded bits to obtain coded bits. The location of the information bit in the to-be-coded bits is determined based on the obtained first sequence that includes N polar channel sequence numbers, so that performance of a polar code can be improved.