Variable-Length Block Coding With Precomputed Puncturing Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for channel-coding in mobile communication systems, particularly in the 3GPP LTE system, face challenges in efficiently handling control information with varying lengths using block codes, as they require extensive calculations to find optimal puncturing patterns for maximum Hamming distance, leading to increased complexity and resource usage.
Innovation Solution
A method is proposed that reuses and modifies a code generation matrix from the conventional 3GPP system to generate (20,A) block codes by puncturing specific rows and columns, adding additional sequences, and permuting rows and columns to achieve maximum or minimum Hamming distance, thereby simplifying the coding process and reducing computational requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional block coding methods are used to handle control information with varying lengths, then the coding can be performed using standard algorithms, but the computational complexity increases and resource usage increases due to extensive calculations required to find optimal puncturing patterns
Solution Approach 1:
The patent pre-calculates and stores optimal puncturing patterns for different information bit lengths in lookup tables during system initialization. When channel coding is needed, the system simply retrieves the pre-computed patterns based on the current information length, eliminating the need for real-time optimization calculations and significantly reducing computational complexity.
Solution Approach 2:
The patent transforms the continuous optimization problem of finding optimal puncturing patterns into a discrete lookup operation by parameterizing the solution space. Different information bit lengths (A=4,5,6,7,8,9,10) correspond to different pre-stored patterns, converting a complex computational task into a simple parameter-based retrieval operation.
2Reliability
If optimal puncturing patterns are calculated in real-time to achieve maximum Hamming distance, then the coding performance is optimized, but the number of calculations increases significantly
Solution Approach 1:
Optimal puncturing patterns that maximize Hamming distance are pre-computed and stored in lookup tables before actual communication operations. The tables contain pre-optimized patterns for various information bit lengths, allowing the system to achieve maximum coding performance without performing real-time optimization calculations.
Solution Approach 2:
Instead of recalculating optimal patterns each time, the patent creates and stores copies of the optimal puncturing patterns in lookup tables. These pre-computed solutions are then reused across multiple coding operations, eliminating redundant calculations while maintaining optimal performance.
3Reliability
If extensive calculations are performed to find optimal puncturing patterns, then the Hamming distance is maximized, but the resource usage increases
Solution Approach 1:
The energy-intensive calculation of optimal puncturing patterns is performed once during system initialization and stored in lookup tables. Subsequent channel coding operations simply retrieve pre-computed patterns, dramatically reducing energy consumption while maintaining maximum Hamming distance optimization.
Solution Approach 2:
The system performs the computationally expensive optimization task once during initialization, and the resulting lookup tables serve themselves for all future operations. The pre-computed patterns automatically provide the necessary guidance for optimal puncturing without requiring additional real-time computational resources.
Data Source
AI summary
A channel coding method of variable length information using block code is disclosed. A method for channel-coding information bits using a code generation matrix including 20 rows and A columns corresponding to length of the information bits includes, channel-coding the information bits having “A” length using basis sequences having 20-bit length corresponding to columns of the code generation matrix. If “A” is 10, individual basis sequences of the code generation matrix correspond to column-directional sequences of a specific matrix composed of 20 rows and 10 columns. The specific matrix is made from 20 rows of the (32,10) code matrix used for TFCI coding were selected.