LDPC Decoding for Variable-Length Data Without Padding Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LDPC codes struggle with adapting to variable-length data in communication systems, as they require padding or puncturing to match fixed-length blocks, leading to changes in coding rate and redundant sequence transmission.
Innovation Solution
The development of time-varying LDPC-CCs with parity check polynomials of coding rate (n−1)/n, allowing for encoding and decoding of arbitrary-length information sequences using a prime period greater than 3, which maintains high error correction capability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LDPC block code is applied to variable-length Ethernet frame, then error correction capability is improved, but coding rate changes or redundant sequences are transmitted due to padding or puncturing
Solution Approach 1:
The patent transitions from static LDPC block codes with fixed block lengths to dynamic LDPC convolutional codes that can adapt to variable-length information sequences. The encoder processes data continuously without requiring fixed block boundaries, allowing the code to dynamically adjust to different data lengths while maintaining a constant coding rate, thereby resolving the contradiction between error correction capability and adaptability to variable-length data
Solution Approach 2:
The patent changes the fundamental parameter of code structure from block-based fixed length to convolutional variable length processing. By using parity check polynomials with time-varying coefficients and processing information sequences of arbitrary length, the system maintains consistent coding rate parameters while adapting to different data lengths, eliminating the need for padding or puncturing operations that previously altered the coding rate
2Adaptability or versatility
If padding or puncturing processing is applied to adjust transmission information sequence length, then LDPC block code can be applied, but coding rate is changed or redundant sequences are transmitted
Solution Approach 1:
The patent extracts and eliminates the unnecessary padding and puncturing operations from the encoding process. By using LDPC convolutional codes that naturally accommodate variable-length sequences, the system removes the redundant sequences that were previously added through padding, thereby preventing information loss and maintaining the intended coding rate without modification
Solution Approach 2:
The patent performs preliminary design of the convolutional code structure to inherently support variable-length sequences from the outset. By configuring the encoder with appropriate parity check polynomials and memory elements before transmission, the system prepares to handle any input length without requiring subsequent padding or puncturing operations, thus avoiding redundant sequence transmission
3Ease of manufacture
If time-varying period is limited to specific values (2, 3, 4 or multiples of 3), then implementation is simpler, but error correction capability is reduced
Solution Approach 1:
The patent creates a universal LDPC convolutional code framework that can accommodate any prime number as the time-varying period. The encoder structure using parity check polynomials with time-varying coefficients is designed to be flexible and adaptable to different period values, allowing the system to maintain high error correction capability for any chosen prime period while preserving implementation feasibility through the standardized convolutional encoding approach
Data Source
AI summary
A decoding method includes inputting coded data, and decoding the coded data to obtain decoded data. The coded data are generated by using an encoding process at an encoding apparatus, and the encoding process includes: (i) repeatedly-selecting and collecting first packets included in the decoded data to generate at least one second packet; (ii) dividing at least one third packet included in the decoded data into fourth packets; and (iii) allocating fifth packets included in the decoded data to respective sixth packets without collecting the first packets or dividing the at least one third packet, and performing an error correcting encoding on the at least one second packet, the fourth packets, and the sixth packets in accordance with a coding rate selected from a plurality of coding rates to generate parity data.