Variable-Length Block Coding with Extended 32-Bit Generator Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the 3GPP mobile communication system, finding an optimal format for block-coded control information of varying lengths is challenging due to the complexity of evaluating individual coding performances for different lengths of information bits.
Innovation Solution
A method for channel-coding information bits using a code generation matrix with 32 rows and A columns, where additional basis sequences are added if A is greater than 10, and the matrix is modified to ensure a minimum Hamming distance of 10, allowing for efficient support of various bit lengths by reusing existing code generation matrices for TFCI and PUCCH transmissions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional block coding methods are used for control information of varying lengths, then coding can be performed for different information bit lengths, but finding an optimal format becomes challenging due to the complexity of evaluating individual coding performances for different lengths
Solution Approach 1:
The patent changes the parameter of code generation matrix by adding column-directional sequences (additional basis sequences) to the conventional (32,10) block code generation matrix. When the information bit length A exceeds 10, (A-10) additional basis sequences are added as column-directional sequences to extend the matrix from 10 columns to A columns, thereby adapting the fixed-rate code to variable-length information while maintaining the original 32 rows and achieving support for various information bit lengths without complex performance evaluation
Solution Approach 2:
The extended code generation matrix serves multiple functions: it maintains compatibility with conventional (32,10) block coding for TFCI information when A=10, supports PUCCH transmission coding when A≤10, and enables variable-length information coding when A>10 by adding column-directional sequences. This universal matrix structure eliminates the need to evaluate and select different coding formats for different information lengths
2Adaptability or versatility
If additional basis sequences are added to extend the code generation matrix for longer information bits, then support for variable length information is improved, but the minimum distance between codewords may be reduced
Solution Approach 1:
The patent applies local quality by carefully designing the additional column-directional sequences to have specific properties. The added basis sequences are constructed to maintain or improve the minimum Hamming distance while extending the code capacity. By locally optimizing the quality of additional sequences rather than uniformly extending the matrix, the patent achieves extended length support without sacrificing reliability
Solution Approach 2:
The patent performs preliminary action by pre-defining the structure and properties of additional basis sequences before actual coding operations. The column-directional sequences are designed in advance with appropriate Hamming distance characteristics, allowing the extended matrix to maintain reliability requirements without needing to re-evaluate or adjust parameters during runtime coding operations
Data Source
AI summary
A method for channel-coding information bits using a code generation matrix including 32 rows and A columns corresponding to length of the information bits includes, channel-coding the information bits having “A” length using basis sequences having 32-bit length corresponding to columns of the code generation matrix, and outputting the channel-coded result as an output sequence. If “A” is higher than 10, the code generation matrix is generated when (A-10) additional basis sequences were added as column-directional sequences to a first or second matrix. The first matrix is a TFCI code generation matrix composed of 32 rows and 10 columns used for TFCI coding. The second matrix is made when at least one of an inter-row location or an inter-column location of the first matrix was changed. The additional basis sequences satisfy a value 10 of a minimum Hamming distance.