Polar Code Construction Using Short-Sequence Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing polar codes face challenges in constructing longer sequences due to high complexity and poor performance when extended using approximation algorithms, particularly in 5G communication systems, where channel encoding demands lower complexity and better performance for various communication scenarios.
Innovation Solution
Construct polar codes based on short sequences with a length of N0, reducing the length of sequences to be stored and simplifying the construction process by extending these sequences to required lengths using a method that reuses existing technology to a maximum extent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If the polarization weight sequence extension method is used to extend the optimal short sequence to required length, then the polar code can be constructed with longer code length, but the performance of extended sequence is poor and the complexity of extension is high due to floating-point operations
Solution Approach 1:
The patent segments the sequence extension process into two distinct parts: integer operations for determining information bit positions and floating-point operations only for calculating polarization weights. This segmentation allows the system to use efficient integer arithmetic for the majority of operations while limiting floating-point operations to only where necessary, thereby reducing overall computational complexity.
Solution Approach 2:
The patent changes the parameter representation by using integer-based indexing instead of floating-point values for determining information bit positions. The system transforms the sequence extension problem from a continuous floating-point domain to a discrete integer domain for position determination, which significantly reduces computational complexity while maintaining code construction capability.
2Length of stationary object
If the polarization weight sequence extension method is used to extend the optimal short sequence to required length, then the polar code can be constructed with longer code length, but the performance of extended sequence is poor
Solution Approach 1:
The patent performs preliminary action by pre-calculating and storing polarization weight values in a lookup table before the actual code construction process. This preliminary preparation allows the system to retrieve pre-computed weights during sequence extension, avoiding repeated floating-point calculations and ensuring consistent, high-quality weight values are used, thereby improving the performance of extended sequences.
Solution Approach 2:
The patent uses copying by retrieving pre-computed polarization weight values from a lookup table rather than recalculating them. This copying approach ensures that optimal weight values are consistently applied during sequence extension, maintaining high performance while reducing computational complexity.
3Ease of manufacture
If a sequence with length of 1024 is used for polar code construction, then the existing polar code can be applied, but it cannot be directly applied to construction of a longer polar code
Solution Approach 1:
The patent implements dynamics by creating a flexible sequence extension mechanism that can adapt to different code length requirements. The system dynamically extends the base sequence of length 1024 to various longer lengths (2048, 4096, 8192, etc.) by combining the base sequence with extended sequences, allowing the polar code construction to be easily adapted to different application scenarios without being limited to fixed lengths.
Solution Approach 2:
The patent achieves universality by developing a general-purpose sequence extension method that works for multiple code lengths. The extension technique using integer operations and lookup tables can be applied uniformly to extend sequences from length 1024 to any required longer length, making the construction process universally applicable across different polar code configurations and scenarios.
Data Source
Figure 1
Figure 2~4(c)
Figure 5~6
AI summary
This application provides a polar encoding method and apparatus. The method includes: obtaining a basic sequence, where the basic sequence includes N0 subchannel numbers; sequentially reading first subchannel numbers from the basic sequence; sequentially reading 2m second subchannel numbers from the basic sequence starting from an Mth subchannel number based on the first subchannel number read each time; and adding q∗N0 to each of the 2m second subchannel numbers to obtain 2m third subchannel numbers, and constructing a polar code by using subchannels corresponding to the 2m third subchannel numbers as information bits. A polar code with another code length is constructed based on a sequence with a length of N0. This method can reduce a length of a polar code that needs to be stored, reduces complexity, and is easy to implement.