Lifted LDPC Code Generation Across Multiple Block Lengths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems face challenges in efficiently generating high-performance low-density parity-check (LDPC) codes for new radio (NR) technology, which is essential for improving encoding and decoding processes in 5G networks, particularly in handling errors and ensuring reliable data transmission.
Innovation Solution
The method involves selecting integer lifting values to generate multiple lifted LDPC codes, where the lifting size value is greater than the maximum range, allowing for efficient encoding and decoding of information bits, and transmitting the resulting code words.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple lifted LDPC codes are generated for different lifting sizes, then code performance and adaptability are improved, but generation complexity and computational overhead increase
Solution Approach 1:
The patent applies universality by designing a single LDPC base graph that can generate multiple lifted codes with different lifting sizes (e.g., Z=128, 256, 512, 1024) through parameter configuration rather than separate code designs. The base graph structure serves multiple functions across different code rates and block lengths, eliminating the need for multiple specialized code designs and reducing generation complexity.
Solution Approach 2:
The patent utilizes parameter changes by modifying the lifting size parameter Z and code rate parameters while keeping the base graph structure fixed. By changing these parameters, the system generates optimized LDPC codes for different communication conditions without redesigning the entire code structure, thus improving adaptability while controlling complexity.
2Productivity
If lifting size value is increased beyond maximum range, then encoding and decoding efficiency are improved, but memory requirements and processing overhead increase
Solution Approach 1:
The patent applies segmentation by dividing the LDPC code into a compact base graph structure that can be systematically expanded through lifting operations. Instead of storing complete large-scale parity check matrices for each lifting size, the system stores only the base graph and generates expanded matrices through structured repetition and permutation, significantly reducing memory requirements while maintaining encoding/decoding efficiency.
Solution Approach 2:
The patent implements the nested doll principle by embedding the base graph structure within larger lifted code structures. The base graph serves as a template that is nested and repeated Z times to form the complete parity check matrix, allowing efficient memory utilization where the same base structure is reused across different code sizes rather than storing entirely separate matrices for each size.
3Adaptability or versatility
If multiple lifted LDPC codes are generated with different lifting sizes, then adaptability to different code rates is improved, but the time required for code generation and configuration increases
Solution Approach 1:
The patent applies preliminary action by pre-defining the base graph structure with optimized connectivity patterns that work across multiple lifting sizes and code rates. This preliminary design eliminates the need for time-consuming code generation and optimization processes when adapting to different code rates, as the base structure is already prepared to support various configurations through parameter adjustment alone.
Data Source
AI summary
Techniques and apparatus are provided for efficiently generating multiple lifted low-density parity-check (LDPC) codes for a range of block lengths and having good performance. A method for wireless communications by a transmitting device generally includes selecting integer lifting values for a first lifting size value Z, selected from a range of lifting size values, wherein the selected integer lifting value is greater than a maximum lifting size value of the range of lifting size values; determining one or more integer lifting values for generating at least a second lifted LDPC code having a second lifting size value based on an operation involving the second lifting size value and the selected one or more integer lifting values for generating the first lifted LDPC code; encoding a set of information bits based on the second lifted LDPC to produce a code word; and transmitting the code word.