Polar Encoding Rate Matching with Precomputed Puncturing Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional polar code encoding technologies face challenges in meeting low-latency and low-complexity requirements, particularly in scenarios with frequent channel changes, due to high computation complexity in real-time density evolution and Gaussian approximation constructions.
Innovation Solution
A method and apparatus for polar encoding that determines a puncturing pattern and frozen set independently of channel conditions, allowing for quick generation and implementation, reducing complexity and latency by using specific patterns and sets based on code rate and length, such as |P|=N−M elements for puncturing and |F2|=└|P|×R1┘ elements for frozen subsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real-time density evolution or Gaussian approximation construction is used to determine frozen set based on channel conditions, then decoding performance is improved, but computation complexity increases significantly
Solution Approach 1:
The patent pre-determines frozen sets for different code rates and lengths before actual transmission. These pre-determined frozen sets are stored and directly selected based on channel conditions without requiring real-time computation. This preliminary action eliminates the need for complex real-time density evolution or Gaussian approximation calculations while maintaining good decoding performance.
Solution Approach 2:
The patent prepares multiple frozen sets in advance corresponding to different code rates and lengths, acting as a cushion against the need for real-time computation. When channel conditions change, the system can switch between pre-prepared frozen sets without performing complex calculations, thus cushioning the system from high computation complexity.
2Adaptability or versatility
If frozen set is reselected based on channel changes or code length switching, then adaptation to channel conditions is improved, but latency increases due to real-time computation requirements
Solution Approach 1:
The patent performs frozen set selection in advance for various code rates and lengths, storing these selections for quick retrieval. When channel conditions or code length change, the system simply selects from pre-computed options without performing real-time calculations, thus maintaining adaptability while minimizing latency.
Solution Approach 2:
The patent enables dynamic switching between different pre-determined frozen sets based on channel conditions and code length. While the frozen sets themselves are static and pre-computed, the system dynamically selects the appropriate frozen set from the pre-prepared collection, achieving adaptability without real-time computation.
3Adaptability or versatility
If puncturing or shortening is performed during rate matching, then code rate flexibility is improved, but the intersection between puncturing set and shortening set causes complexity
Solution Approach 1:
The patent pre-determines both the puncturing pattern and frozen set together in advance, ensuring that their intersection is properly handled before transmission. This joint pre-determination eliminates the need for complex real-time coordination between puncturing and shortening operations, simplifying the rate matching process while maintaining code rate flexibility.
Data Source
AI summary
The present disclosure provides an encoding method and apparatus, and relates to the field of communications technologies, to reduce an encoding latency and complexity, and the amount of computation of real-time construction. The encoding method includes: obtaining information bits; determining a puncturing pattern, where the puncturing pattern includes an element in a puncturing set and an element in a shortening set, and the puncturing set and the shortening set have no intersection set; and performing, by using the determined puncturing pattern, rate matching on data obtained after the information bits are encoded.