研究目的
To develop a hybrid decoding algorithm for low-rate LDPC codes in 5G that improves performance closer to belief propagation decoding with only a slight complexity increase compared to the normalized min-sum algorithm.
研究成果
The proposed H-MIN algorithm achieves performance much closer to belief propagation decoding with only a slight complexity increase over the NMS algorithm, providing a better trade-off for low-rate LDPC codes in 5G. It is robust to channel variance estimation errors and can be efficiently applied with layered scheduling.
研究不足
The algorithm's complexity increases slightly compared to NMS, and it may require optimization of parameters like the degree threshold and multiplicative factors for different code rates. The linear approximation might not perfectly match the exact BP decoding, and the study focuses on specific 5G LDPC codes, limiting generalizability to other codes.
1：Experimental Design and Method Selection:
The study designs a hybrid decoding algorithm (H-MIN) that applies normalized min-sum (NMS) simplification for high-degree check nodes and linear approximation for low-degree check nodes based on a predefined degree threshold. It uses protograph-based EXIT (PEXIT) analysis for threshold evaluation and simulations for performance comparison.
2：Sample Selection and Data Sources:
LDPC codes from 5G NR base graphs BG1 and BG2 are used, with specific rates (e.g., rate-1/4 and rate-1/5) and information bit length K=2048. Data is generated through AWGN channel simulations.
3：Data is generated through AWGN channel simulations.
List of Experimental Equipment and Materials:
3. List of Experimental Equipment and Materials: No specific physical equipment is mentioned; the work is computational, involving algorithms and simulations likely performed on computers.
4：Experimental Procedures and Operational Workflow:
The algorithm initializes LLRs with a multiplicative factor, updates check nodes based on degree (NMS for high degree, linear approximation for low degree), performs variable node updates, and iterates until parity checks are satisfied or maximum iterations are reached. Simulations are conducted to measure block error rate (BLER).
5：Data Analysis Methods:
Performance is analyzed using PEXIT for iterative thresholds and BLER simulations. Complexity is evaluated based on the number of operations per check node update.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
