Layered LDPC Decoding Using Reduced-Node Syndrome Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LDPC code decoding methods in wireless communication systems face challenges with high power consumption and latency, particularly in achieving low error rates and low latency radio requirements, especially in future-generation communication systems.
Innovation Solution
A method for layered decoding of LDPC codes that reduces memory reads, power consumption, and latency by optimizing the updating and syndrome check processes, specifically by modifying the relational expressions for syndrome value calculation to minimize the number of variable nodes connected to each check node and using a hierarchical decoding approach.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional LDPC decoding methods are used to achieve low error rates, then reliability is improved, but power consumption and latency increase
Solution Approach 1:
The decoding process is divided into multiple layers where variable nodes are processed in sequential groups rather than all at once. Each layer updates a subset of variable nodes based on check node syndromes, allowing the decoder to make progress toward error correction while consuming less power per iteration and reducing overall decoding latency.
Solution Approach 2:
The decoding algorithm dynamically adjusts the updating order of variable nodes across different layers based on syndrome information. This dynamic approach allows the system to focus computational resources on the most critical error patterns first, improving reliability while optimizing power consumption by avoiding unnecessary updates to already-corrected nodes.
2Reliability
If conventional LDPC decoding methods are used to achieve low error rates, then reliability is improved, but latency increases
Solution Approach 1:
By segmenting the variable nodes into multiple layers and processing them sequentially, the decoder achieves low error rates through systematic error correction while reducing latency. The layered structure allows early termination when syndromes become zero, avoiding the need to process all variable nodes in every iteration.
Solution Approach 2:
The algorithm performs preliminary syndrome checks before completing full variable node updates. If syndromes are already zero after processing some layers, the decoding can terminate early, significantly reducing latency while maintaining reliability by ensuring all necessary error corrections have been applied.
3Use of energy by moving object
If the number of variable nodes connected to each check node is minimized for syndrome calculation, then power consumption and latency are reduced, but decoding complexity increases
Solution Approach 1:
Variable nodes are segmented into layers with controlled connectivity to check nodes. Each layer is designed to update a specific subset of variable nodes based on syndrome requirements, reducing the number of connections per check node while maintaining decoding capability. This segmentation lowers power consumption by limiting the active node set at each step.
Solution Approach 2:
The connectivity between variable nodes and check nodes is dynamically adjusted across layers. The updating order and connection patterns are optimized to minimize simultaneous active connections, reducing power consumption. The dynamic reconfiguration of the decoding graph across layers manages complexity by processing nodes in manageable groups rather than all at once.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An improved layered decoding method for a low density parity check (LDPC) code and a device therefor are disclosed. Disclosed is the layered decoding method for an LDPC code, capable of determining whether decoding is successful by performing a syndrome check on each check node at every variable node update. In addition, the syndrome check can be performed by using reduced variable nodes, thereby reducing decoding power consumption and decoding time consumption.