Channel Encoding With Uniform Source Symbol Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication/broadcasting systems face inefficiencies in information recovery due to high system overhead and decreased reliability when using erasure codes, particularly in scenarios with low feedback channel fidelity or multiple transmitters/receivers, where ARQ schemes are not viable.
Innovation Solution
The implementation of a Fountain code-based encoding method using Luby-Transform (LT) codes, which generates encoding symbols uniformly from all source symbols, ensuring each symbol is used at least once and optimizing their selection based on the number of times they have been used, thereby reducing overhead and improving recovery efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If erasure codes are used for information transmission, then reliability of information recovery is improved, but system overhead increases
Solution Approach 1:
The patent changes the parameter of symbol selection probability from non-uniform to uniform distribution. By making each source symbol equally likely to be selected for encoding, the system achieves more balanced contribution of all symbols to encoding symbols, which improves recovery reliability while reducing the number of redundant encoding symbols needed, thus lowering system overhead.
2Productivity
If ARQ scheme is used for packet loss recovery, then information transmission efficiency is improved for point-to-point communication, but the scheme becomes non-viable for multi-to-multipoint transmission or when feedback channel fidelity is low
Solution Approach 1:
The patent segments the encoding process into multiple independent encoding operations where each encoding symbol is generated independently using uniform random selection of source symbols. This segmentation allows the system to function without feedback mechanisms, making it adaptable to multi-to-multipoint transmission and scenarios with low feedback channel fidelity, while maintaining transmission efficiency through the Fountain code structure.
3Device complexity
If source symbols are selected with non-uniform probability for encoding, then encoding complexity is reduced, but information recovery reliability decreases
Solution Approach 1:
The patent applies homogeneity by making the selection probability of all source symbols equal (uniform distribution). This homogeneous treatment of all source symbols ensures that each symbol contributes equally to the encoding process, which improves information recovery reliability by preventing any single symbol from being underrepresented, while maintaining manageable encoding complexity through the simplicity of uniform random selection.
Data Source
AI summary
An encoding apparatus and method in a communication/broadcasting system are provided. When an encoding symbol is generated using at least one source symbol from among total source symbols to be encoded, encoding symbols are generated by using each of the total source symbols with a uniform probability before repeatedly using any of the total source symbols.