Non-Binary LDPC Extrinsic Calculation Without Memory Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional LDPC decoders face inefficiencies in computing decoder extrinsic information, particularly in non-binary systems, as they rely on memory storage and subtraction processes that can be cumbersome and resource-intensive.
Innovation Solution
A method and system for decoding vectors using non-binary LDPC decoders that compute R messages through a message passing process, forming decoder extrinsic information by linearly weighing R messages with inverse coefficients from the parity check matrix, and combining these with detector extrinsic information without relying on memory storage, employing a layered decoding approach and symbol metric to error metric conversion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional LDPC decoders store detector extrinsic information in memory and load it for subtraction, then the decoder can compute extrinsic information, but the memory access and computational complexity increase
Solution Approach 1:
The patent extracts and eliminates the memory storage and subtraction operations from the traditional LDPC decoding process. Instead of storing detector extrinsic information in memory and performing subtraction, the invention directly computes decoder extrinsic information by combining R messages with inverse coefficients from the parity check matrix, thereby removing the cumbersome memory access and subtraction steps while maintaining error correction capability
Solution Approach 2:
The patent inverts the traditional computation approach: rather than computing P message and subtracting detector extrinsic information to obtain decoder extrinsic information, the invention directly forms decoder extrinsic information by linearly combining R messages using inverse coefficients. This inverted approach eliminates the need for subtraction and memory storage, reducing computational complexity and memory access requirements
2Reliability
If traditional LDPC decoders use memory storage for detector extrinsic information, then the computation can be performed, but the processing speed decreases due to memory access overhead
Solution Approach 1:
The patent removes memory storage and access operations from the decoding process by directly computing decoder extrinsic information through linear combination of R messages with inverse coefficients. This extraction of memory-dependent operations eliminates access overhead and latency, thereby improving processing speed while maintaining decoding accuracy through the mathematically equivalent computation path
3Productivity
If non-binary LDPC decoders compute R messages and form decoder extrinsic information without memory access, then the computational efficiency improves, but the algorithm complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing the inverse coefficients from the parity check matrix in a lookup table or pre-processed structure. This preliminary preparation allows the main decoding algorithm to simply perform linear combinations of R messages with these pre-computed coefficients, thereby reducing the algorithmic complexity during actual decoding operations while maintaining high productivity and efficiency
Data Source
AI summary
Systems and methods are provided for decoding a vector from a communications channel using a non-binary decoder. The communications channel may correspond to a wired or wireless channel. A message passing process computes R messages corresponding to a variable node of the non-binary decoder. Decoder extrinsic information is formed for the variable node by combining the R messages. The decoder extrinsic information is provided to a soft-detector.